Although it was funny to imply that they were getting dumped on MEXC, in reality most of the funds were getting transferred back on chain.

Where they got dumped.

But just some of the arena! The rest of it of it was held in various accounts. I tagged all the accounts (as one does) and moved on with my life.

Until now.

MEXC is not a MIXER

Just an FYI for the more retarded of you out there (all of you).

Don’t transfer funds to MEXC and expect that they will stay anonymous. It was relatively easy to track what happened to the funds in the screenshot above.

Y no fdn support????

Every time I see one of these dumbass posts it reminds me to check in on $ARENA.

Not sure why but I though maybe I should check in on what’s been going on with those mixed funds, and I stumbled across something very funny!

Get moonbags’d, pleb

Man, man, man. That’s a lot of arena loans getting taken out on Salvor in a short period of time! I wonder why? What’s happening to them?

Loans 1 (0x127)

Swapped/bridged to SOL, bought $testicle lmaooooo

Loans 2 (0x370)

‘Mixed’ via Debridge to 0x4ae86d58abdb1413da0adbd3c185d9d64f181db2 and 0x065ecbe6dbbd13cfd39378350dbff0abed3cbf8d.

Puzzling! Funds were used to pay back previous loans on the target wallets. Bizarrely, these two accounts were both funded directly from the StarsArena Deployer wallet.

Pure speculation, but perhaps the goal was to shift the intended default from a wallet closely linked to SA to one with more plausible deniability.

Or perhaps… something more nefarious, like trying to build up credibility for future loans. Who can say!

Loans 3 (0x2de)

Lol. Lmao, even.

Loans 4 (0xcb7)

Another high quality Debridge mixer.

Similar to Loans 2, this one also has existing Salvor loan fuckery. Very curious.

Loans 5 (0x348)

Another case of Salvor loan fuckery. The proceeds from the earlier loan were sent to some wallet that is just holding them.

But, why?

Lot of weird shit going on here. I’m writing this on Tuesday, very curious to know what happens when these loans are up!

The borrows and repays remind me of how Moonbags was trying to juggle loans on Juicy for a time, before he gave up and just let them default.

Anyway, the reason for this is that they wanted to ‘sell’ $150K of arena without hitting the LP. The LP has about $200K in avax across all pools.

I believe arena is currently using a paid LP service, based on the way their liquidity is being moved. I wonder what their contract terms are?

But, who?

Great question!

Feels unlikely to me that Arena leadership knows enough to do this kind of stuff. Not sure how they manage to tie their shoes in the morning.

So my guess is that these are payments to employees trying to cash out without nuking the token.

Partychad? Are you there?

Anyway, the other ‘who’ is the lender. Who could be lending this much liquidity for a shitcoin?

This guy AGAIN? I’d say ‘poor bastard’ but honestly at this point maybe he deserves it.

How sure are you about this stuff?

Really sure that the funds came from the Arena team. Really sure about where the funds ended up.

Not sure who specifically took any of these actions. Good opsec!

Deadly sure that there are about $120K of arena loans outstanding by either the same person or closely related persons.

Late breaking update! Jan 20, 9PM ET

Loan 2 bridge target 2 sent 160A to some wallet, and then bridged funds to SOL.

It sent more funds to a wallet initially funded by ChangeNow, the CEX of champions (criminals).

Unsurprisingly, it bought more shitcoins.

Late breaking update 2! Jan 21, 12 PM ET

I wrote a Dune query to summarize the status of loans on Salvor, since their UI is.. inflexible. Then I filtered by outstanding and sorted by loan amount descending, and found ANOTHER arena-linked loan wallet.

I also found World2020 and Jony in there, which is very funny. World2020 already transferred his avax out and he’s gotta be giga poor by now, high risk of default there.

Wonder if Jony will default on his by the time this article posts? He’s borrowing about $35K avax against his arena and he’s about to get liquidated on his massive avax long.

Not looking good, bruh. My guess is that he gets liquidated, uses the new headroom to repay the arena loan, and dumps it to recoup the avax, sending arena down a solid 20%.

Anyway, let’s take a look at Loans 6!

Another one directly linked to SA, doing insane things. Using shady CEXs (ChangeNow) and not-actually-mixers (deBridge) to move funds, juggling funds. Funny stuff.

Looks like this one just bridged to a Solana wallet buying testicle (this will never get old). You know what, let’s see how he’s doing on testicle.

And minecraft grandma fund?

Not great, but could be worse!

N/m I’ve got things to do

I was going to wait until after they started defaulting to post this, but I want to be on to something else soon. First round of loans is due tomorrow.

Good luck to the lenders, and good luck to the people still offering an outstanding (checks notes) $50K of loans for arena.

This person (or persons) still has plenty of arena to lend out. Probably not worth those points, buddy.

In the future perhaps I should wait until closer to the date to compile the article instead of just appending updates to the end. Oh well.

Recently, Moonbags has been delivering some bangers that got my jimmies all rustled up. So let’s take a deeper look at things.

Context

Yesterday, he posted a lot of yapping.

It’s way, way longer than this, you can read it here. This is what you might call an “unforced error”. No one made him post this bullshit, and it really served no purpose.

Anyway, I’ve long been aware of him dumping $juicy, in fact I’ve posted myself about him dumping it via Salvor loans. So one small thing in here jumped out at me.

It’s actually quite common for founders of tokens who care to lose money on their coins because they use funds “supporting the price”. So I figured I would take a little look to see if that was the case here.

And I did find buys! Wallets with actually more buys (in USD) than sells. I’m a fair guy, so I made sure to highlight them here.

Unfortunately, I found A LOT of wallets with sells. X limits you to 4 screenshots per post so I cut it off there, but I posted more in Discord.

This post got a lot of commentary! I scrolled through looking for some bangers so you don’t have to.

This one is curious, I’m not that familiar with the tokenomics, we’ll come back to that

Someone came with receipts, spicy!

OTC, you say?

Oof.

The empire rebuts

So later that day he followed up with another post, yapping about the negative response to his earlier yapping.

Some really quality shit in here, but let’s focus on the relevant part for this post.

Had to double check, but it looked like I was the only person who pointed to specific wallets in the last post. You know, other than that guy who posted screenshots where Moonbags tried to get him to put up MORE collateral for him to dump into.

I’m not the only one who saw this and thought it might be about me.

We’ve also got this guy, a notorious rugger strongly in support of Moonbags.

Whew, good thing I have a ton of free time to research this and prove my case!

ChANALisis

I tried to put an anal joke in there but I don’t think it worked. Moving on.

You might wonder how someone can be sure that an account is owned by someone else. Well despite what you might think from reading Zach’s posts, it’s actually quite difficult and rarely cut and dry. I’ve made mistakes before, for sure! I’ve become quite conservative in attributing ownership, but even still I occasionally make mistakes.

When you’ve got 50 wallets tagged for a person, it’s (probably) likely that some of them are wrong. You do the best you can, you be conservative with your claims.

Let’s take a look at some of the methods that you can use to estimate ownership.

Sending funds from A to B

There are plenty of reasons to send funds! As Moonbags notes, sending funds to someone doesn’t imply ownership of the receiving account. This is very weak evidence. It’s stronger evidence if it’s a substantial amount of money but still.

You might think I’m smitty, but he’s actually much better looking than me.

Repeatedly sending funds from A to B to A to B

OK, well this is better evidence that two accounts share ownership. Still not perfect though. Plenty of reasons to send money back and forth between friends.

Sweeping funds

OK this is where it starts to get easier. It’s very common for people to use a small amount of “hot” wallets at a time, and people sweep funds out of older addresses and remove them from their wallet software to keep things from cluttering up.

Just make sure you don’t catch people doing presales, or paying for services, or some such! This kind of pattern where it is smallish, random amounts of money, all sent in a short timeframe, is pretty good evidence.

A web of entwined accounts, A, B, C, D

This is another good one. Most people have TERRIBLE wallet hygiene. Wallets don’t have specific purposes, they interact with each other randomly over a long period of time.

Once you start putting people into an entity and you click on them and see the transfers between accounts in that entity. Oof.

Well known tags and payments

Obviously some tags are straightforward. It doesn’t take a genius to know who owns this.

The place this wallet sweeps funds to? Yeah probably also the same guy. Everyone’s got Arena wallets with verified X identities now, makes my life so much simpler.

This is an easy one obviously, Juicy has contracts and this is the account that deployed them.

Here’s a payment for the work on AIVAX, I was told at the time that it was $30K, lines up perfectly.

Corroborating evidence

Here we’ve got these accounts that are receiving funds directly from the Juicy Gnosis Safe. This isn’t perfect evidence of ownership but it does indicate intent, since the TX had to be signed by someone important!

Here’s a Gnosis Safe for $RED, the flagship launch on FizSocial (a Moonbags product) which I am told Moonbags was behind (went to zero).

Could be someone else, could be. Who can say?

I’m not sure what this is for, some kind of MPC wallet? I didn’t look that close. But if you look at the top counterparties, wow that’s some corroboration.

Shared CEX deposit addresses

This is a good way to find candidate wallets, but it’s once again just corroboration. I blanked out a third party who sent him some funds below!

But generally speaking, finding reused CEX addresses is a great way to ID wallets. It’s less common (but not unknown) for people to send directly to other people’s CEX wallets.

Common behaviors

This is just more corroborating-level evidence in general, and the specific type of ‘common behavior’ could vary from person to person. Maybe they only use MetaMask swap like a retard. Maybe they only use Pharoah. Maybe they exclusively trade some obscure NFT. It will vary from investigation to investigation.

But I wanted to call it out specifically because it’s funny and will come up again later. There’s a really funny thing you can see over, and over, and over, and over across these wallets.

Personally, I’ve never felt the urge to loan out tokens on Salvor. I’m sure people do. I’m a bit less sure about how many people are loaning out Juicy.

Harder evidence

And finally we’ve got stuff like this

Not really up for debate, given that he admitted to it on X, right?

Another fun thing is when you have a billion other people in the eco tagged and you see them in counterparties and you reach out to them for confirmation.

People love doing pre and post sales!

And then there are obvious payments for undisclosed shilling, where the awkward response is an admission of guilt.

Yeah IDK bro people are always sending me thousands of dollars of coins and I just sell them without knowing why.

I got tired of scrolling back on this query, honestly. Isn’t undisclosed paid shilling a crime in the US?

The facts

Wow, that was quite a digression! I hope you learned something today.

Let’s move on to the facts. I’ll ignore the wallets that I claimed bought more Juicy than they sold (why would he complain about that?) and just focus on the dumpers.

Specifically, the 4 dumpers that I posted; I want to be clear that there are A LOT of his wallets that dumped Juicy. I just don’t care to do that much work and total it up.

Here are the wallets, my tags for them as I was investigating, and the evidence. I’m not going to bother with the charts or the dumps again to save space. I ordered them to make the story clearer.

Dumper 4 - “Big JUICY seller, AIVAX Pay”

Well this one is an easy ID, between the AIVAX pay and 3rd party confirmation.

Also I want to note up front that Dumper 4 interacts with Dumper 1, 2, and 3.

Dumper 3 - “MASSIVE sells, RED Gnosis Signer”

Evidence:

• RED Gnosis signer

• EIGHT PAGES of interactions with other wallets I have tagged as Moonbags

• Initially funded by Dumper 1

• Interacts with Dumper 1, 2, and 4

• Gets funds from Juicy Gnosis

• Interacts with Moonbags-deployed vault contract

• Sends Juicy to Hazy (employee)

• Gets funds from the AIVAX wallet

Dumper 1 - “Lots of sells, 2 small buys”

Evidence:

• Funded Dumper 3

• FOURTEEN pages of interactions with Moonbags wallets

• Interacts with Dumper 2, 3, 4

• Come on man

Dumper 2 - “All sells, mid sized”

Evidence:

Actually why the fuck am I still doing this? Ahahahaha. Let me just post a screenshot.

The TYPES of dumping

So you know, you can find lots of evidence of Moonbags literally dumping right into the LP. But what about the other kinds of dumps?

Like the OTC sales? If you recall from earlier, he was openly posting about OTC’ing Juicy to people. You can call it what you want but that’s selling (eventually).

And how about the Salvor loan dumping? I’ve mentioned this several times now.

Soooo, after I did all this research to confirm things, I had Claude write this for me! The results will SHOCK you!

Somehow almost ALL of the defaulters were already tagged as Moonbags, color me surprised. I went to check those last two untagged ones, and yup, also Moonbags.

Claude is so helpful, I love him.

So to be clear:

• He sold tokens into the LP

• He sold tokens OTC

• He sold tokens by defaulting on SIXTEEN FUCKING PERCENT OF THE FLOAT into Salvor loans

We shall now begin Ethics

I’m not a lawyer! But as a layperson, I think that if your tokenomics say you have X% of your tokens unlocked, and then you sell those X% into the market, you’re probably in the clear.

Ethically? IDK, this is all gambling, sometimes you fuck and sometimes you get fucked.

I even already posted about how I personally think defaulting on the loans is kosher.

I’m a little confused though; someone mentioned 15% of Juicy was the dev allocation. How did he manage to default on 16.5%? That’s not even including the OTC and the direct sales.

Curious.

I attempted to find the tokenomics for Juicy.

Wow really weird how all their medium articles are gone. Thank god for the wayback machine!

Well, I guess he’s in the clear, everyone knows there’s no difference between team and treasury.

There’s no way he dumped more than 26% right?

Right?

I enjoy being rubbed, but not in the wrong way

Few things piss me off, honestly.

First, the martyr complex. Cry me a river.

You’re running a business. You can gamble on it being successful if you want. If you win, you get a huge payoff. If you lose, you don’t.

You can sell tokens to pay bills. That’s cool. But lying and pretending that you didn’t, less cool.

Second, the weasel words.

“By the treasury wallet”. I didn’t even verify that was all it sold, I’ll just say sure, that’s all it sold and not look any deeper. But what about the OTHER WALLETS.

HOW MUCH DID THE OTHER WALLETS SELL (a lot)?

HOW MUCH DID YOU OTC (other people sold)?

HOW MUCH DID YOU DEFAULT (16.5% of the float).

Third, implying I don’t know what I’m talking about.

Do you want a 10 page article exposing your bullshit? Because that’s how you get a 10 page article exposing your bullshit.

The end

I’ll leave you with this.

Undaunted by the challenges faced by our predecessors, Deep Water Studios took up this challenge and delivered a quality product to the 2025 Avalanche Summit.

Here’s a behind the scenes look at how it all went down.

Ancient history

It begins again

Disco, WTF

Lock in

Noodles identified

Ferdy Coin

What did you accomplish yesterday

Shipping

Dry run

Expenses

We love stickers

Taste test

It’s go time

Design work

Expenses #2

Co-marketing

Brando’s crashout arc begins

Finalizing packaging

Tragedy strikes

Measurement woes

Coins are being French

Wrapping up

At the summit

Crashout arc complete

The end (of brandi)

Regardless, people persist. In this article I’ll discuss a wide variety of topics related to token launches based on my observations from the last few years, and firsthand experience repeatedly helping people launch tokens over the last half year.

Scoping

I’m not qualified to discuss some things, like launching a token for an L1 with first day CEX listings. I’m barely qualified to discuss launching a token for a protocol that actually intends to do things. I’m not familiar enough with the weird bundling / pump and dump scams on Solana, although I’ve probably heard the same rumors you have.

This article will focus on launch mechanisms for smaller, honorable token creators that want to make a community, or a meme, or a small protocol with fee sharing or something.

But first, there are a few topics you should understand before considering how the token should be distributed.

Why a token

Amazing question. Why DO you need a token?

If you have a successful product already, you probably don’t need one, and the only reason to create one is to cash out. Which can be fine, if done properly.

You might need one as part of your business model. For example, $joe existed so that they could incentivize LPs, and people could speculate on the price going up faster than people dumped it. And also so the founder could dump it.

If you have an L1, you need a token for economic security, but that’s out of scope for this article.

Many tokens are launched explicitly for speculation and/or blatant theft (e.g. bundled and dumped on more gullible speculators), this describes basically every token launched on pump fun.

More recently there have been ‘community’ tokens launched, where the idea is that shared ownership in the token binds people together to create a community. This is similar to what people were doing with NFTs last cycle, except the art gets posted on Twitter instead of attached to your share of the float.

‘Successful’ tokens

It seems clear that a successful token is one where the price goes up. This is a casino and people are happy if they make money, angry if they lose money. Doesn’t matter how hard you work or how honorable your intentions are, price go up is the final decision maker on this.

And to be fair, the founding team does have some (marginal) ability to affect price upwards. Stuff like how they launched, how they manage liquidity, how they market, how they behave in public. But 99% of the people here just want to dump for a profit, and most coins will go to zero.

Realistically, the founding team only has the ability to send the token downwards (other than shenanigans like pulling LP and using it to buy). Like maybe by issuing violent threats. Or going on vacation immediately after launch and never coming back. Or getting caught fucking a horse.

It’s funny, and a bit sad, when comparing against tokens like Fartcoin. No one has any expectations for those tokens. Meanwhile on Avax, you have to have a goddamn 5-year dev plan for how you plan to succeed, madness.

The ‘developer’ vs the ‘dev’

Just want to pop in here and mention that the guy pushing the buttons is a software developer (typically) but people often use ‘dev’ to refer to the person launching the token project, and sometimes conflate the two.

They’re different things. As a developer, I push buttons for money. As long as it’s not crime, I’ll take money to do it (some caveats apply). I have pushed buttons for many projects, and I will push more buttons in the future. I take no responsibility for the future of those projects, other than if any mistakes occur during launch.

Price action

So obviously, everyone wants to buy low, see a huge profit spike, and then dump for generational wealth. And this will happen to some people. For memecoins, this is basically what you’re hoping for.

For the founder of any kind of legitimate ‘utility’ coin, you’d prefer a slow and steady ‘up and to the right’ pattern where as a token generates more value, that gets recognized, and the price reflects it in a rational way without going insane.

Good luck with that.

In reality, many tokens will go right to zero, most tokens will go to zero over a few weeks, and even legitimate utility tokens will swing wildly up and down as people reflexively buy and sell shifts in momentum.

Launch hazards

Keeping your token alive long term is one thing, but the few seconds, minutes, and days just after launching is a whole other bag of snakes.

Launch failure

We haven’t seen too much of this recently, but it’s very easy to fuck up your launch. This stuff is complicated, relatively hard to test, you only get one chance, and a lot of ‘devs’ are literally 20 year old retards who read a solidity tutorial once.

There are a lot of things you can do wrong when pushing your token live, the more complicated the launch, the more things that can go wrong. I’ve seen people airdropped multiple times, not airdropped at all, accidentally send the dev funds to the wrong place, burn or lock the LP incorrectly.

We even have Defi Advisors adding LP in the wrong range and accidentally nuking their token!

Sniping

The most well-known hazard is getting your token ‘sniped’. There’s a lot of discussion about what ’sniping’ means, and I’ve written about this quite a lot, but the end result is that a lot of supply is acquired cheaply and gets sold after ‘real people’ have bid the price up much higher.

Sniping is very challenging to prevent, and honestly I think you should mainly focus on ensuring that there’s no motivation to snipe the token instead of fighting the snipers.

Distribution

This refers to the amount of float that is in the hands of any individual person, the concern being that that person could decide to exit and nuke the token, aka FSH (Full Stack Hitler).

Historically people have attempted (poorly) to prevent this using per wallet limits on token ownership, something easily bypassed technologically.

More recently, presales that tie social identity to allocation have come in vogue, which are more effective.

Airdrops

This specifically excludes presale allocation airdrops.

I have no idea what makes an airdrop effective. People want airdrops so they can dump them, full stop.

For some tokens, an airdrop is a way of compensating for fees collected. Basically an over complicated presale mechanism, more recently has become transparent via ‘points’. This seems right and fair to me.

But other tokens that airdrop to users who haven’t contributed anything are hoping to get exposure, I guess? It rarely works out. People get free tokens, they dump them, they don’t suddenly become invested in the ecosystem.

Vesting

There have been a number of presale launches on Avax recently with zero vesting. At first I questioned this, but after some discussion I was convinced that this was a good thing for a few reasons:

1. Get people who don’t want to be a part of your community, out of your community ASAP

2. Higher float available at launch discourages sniping and normalizes the chart faster

3. Dripping float out into the hands of people just ensures a slow death as people claim and dump like clockwork

Liquidity

People have all kinds of crazy ideas about liquidity. They say things like “there’s too much liquidity” or “dev needs to add more liquidity”. These people have no idea what they’re talking about.

But neither do I, honestly. The most intelligent-sounding thing anyone ever said to me on the topic was “there needs to be enough liquidity to facilitate trading without suppressing price discovery” which sounds really smart but I have no idea how that translates to actual numbers.

Presales

If you do a presale, the ratio of token to Avax that gets added to the LP needs to match or exceed the price of the presale. If you mess that up, it’s basically a scam (people buying from the LP get a better price than presale).

For a presale, you need to have some kind of sell-side liquidity, typically people put in the range of 50%-70% of the raise (if it’s not a scam).

Single sided liquidity

Another way to launch is to pick a starting price, and add tokens on a curve to infinity. If you’re not doing a presale, I suggest this mechanism because if you do a 50:50 LP you’re likely to start at an insufficient price relative to demand.

Tokenomics

I don’t have a lot of commentary on this. How much the dev team keeps, whether they keep raise funds, token, or both. How much gets airdropped and to whom. There’s a lot of stuff that basically up for debate.

Really the only thing I have any advice for you on is initial liquidity, which I’ve covered above.

Token Generation Events (TGE)

There are a lot of ways to do the initial distribution of your token, and they all have pros and cons, but some are worse than others.

Note that I’m just discussing the initial distribution here! I’ll discuss some modifications to the launch afterwards.

Stealth

This is when you announce your project with no marketing and add LP at some random time. It is the worst kind of launch. Especially if you do a classic 50:50 LP, since you lose that money forever and you will inevitably set the price way too low.

You’ll also get accused of crime forever (sniping your own launch) and unfair distribution, etc, no matter how noble your intentions. Unsurprisingly, this type of launch is very popular with criminals (Delaunch).

Just don’t do it.

Bonding curves

Stuff like pump.fun is effectively a stealth launch, except even more scammy. The curve for acquiring tokens prior to ‘migration’ is even worse than a very thinly added LP.

There’s really no reason for a bonding curve to exist. You can just launch a token single sided on a full curve. The reason people use those products is to acquire supply cheaply and dump on people. The epitome of the casino.

Fair

This is only slightly less bad than stealth, you’ve announced your project and told people when it will launch, and then you add a small amount of LP and all the tokens. This was very popular a few years ago, notably $coq launched this way.

This is a terrible way to launch. These launches commonly have 1% ownership limits on them, and launch with < $10K in LP. They are easily botted to zero, the protections do nothing. The only reason $coq survived was that it came first, no one expected the response, and also because of exogenous factors we don’t discuss in polite company.

Dutch auction

I haven’t seen too many of these recently, but the idea is that the price per token starts high and then decreases over time. People can buy at any point, but they only end up paying the ‘final’ price for the token, essentially when all tokens have been sold (they get refunded any overpayments).

This is an interesting idea, I think the issue is that it’s hard to gauge sentiment and set the parameters in advance properly. People either end up buying out the launch immediately at way too high of a price, or the token inevitably closes at the lowest possible price.

There’s also no mechanism to ensure smooth distribution, unless you KYC or something. Perhaps this could be solved through whitelisting verified social accounts or something (new meta!?)

Liquidity bootstrapping pool (LBP)

There are a number of sites that provide this service (Fjord), where a project can pair their token with some other asset and let users buy and sell prior to launch to find the launch price.

This is a bit harder to understand than a Dutch auction, but the price is actually higher for early buyers. The risk of trying to buy later at a lower price is that you might miss out entirely.

You can trade in and out of a LBP prior to the launch completing, which is a big difference from a Dutch auction. Anecdotally, this kind of launch seems more common, I’ve seen a lot more LBP launches than DA launches.

Decreasing tax

This is a similar idea to Dutch auction, where the launch occurs on a Dex like usual, but the token itself ‘taxes’ purchases during the launch window. Effectively, you pay a decreasing price over time to buy tokens (*, until the tax is fully removed. Everyone can decide for themselves the price that the token is worth buying at, and earlier buyers pay more (similar to a LBP).

Notably, this harshly punishes snipers who buy in the first few blocks of launch. I’ve seen this happen, and them barely break even or lose money, good tech.

It avoids issues with setting the Dutch auction parameters correctly by effectively setting an infinity price to start and eliminates the ‘everyone gets lowest price’ degenerate behavior.

This does have some downsides though; mechanically it is quite difficult to pull off smoothly, and confusing for end users (needs a custom launch site even though it’s already on a Dex). Some users who are not aware of the mechanics will get burned.

Presale

In a presale, users supply funds in exchange for a promise of tokens at launch. They are happy to do this because it gets them in ‘early’ and the expect the price of the token to go up.

Recently, in an attempt to even out the distribution, presale allocations have been (at least theoretically) tied to social profiles, which are much harder to fake than wallets. This has inevitably resulted in a lot of drama about who gets what allocation.

Regardless of what you think about the drama, a presale is strictly superior to many kinds of launches. A lot of people get the majority of the float at the same price, there is no mechanical advantage to those purchases, and the distribution is guaranteed to be at least more even than buys from an open market.

I think one of the most significant problems with this kind of launch is the fact that the presale size available hasn’t met the demand for the presale, leading to a massive pump on launch, and an inevitable decline.

Initial token trading

People are understandably concerned about how to make token launches ‘more fair’. The key concern being a sniper taking a lot of tokens, and then dumping them immediately for a huge profit.

I already mentioned that I think this is a fool’s errand, but for completeness I’ll discuss some of the options you can ‘add on’ to your TGE.

Custom Dex

Launch your token on a custom Dex with limits of your choice. I wrote a whole article about this, read that post.

Decoy LP

Many launches have been adding fake LP that they create, leave for X seconds, and then remove. The intention is to bait snipers into incorrectly buying the fake LP, and scare them off buying the real LP.

There has been some controversy over how legitimate this tactic is, but regardless it’s become significantly less effective recently, as snipers have pivoted to scraping Twitter for the CA announcement.

You can read this article on the $wink launch.

Factory-based launches

This is common for ‘launchpad’ type things, where a contract is created that will subsequently create the token at a later point, add liquidity, etc. But it can also be done for a standalone launch to ensure that the token creation and provisioning takes place in a single transaction.

This doesn’t do much to prevent a dedicated sniper, although many non-technical TG snipers will require a CA in advance to ‘snipe’ and it will defeat those.

Token modifications

There are a variety of ways that you can modify your ERC-20 token to attempt to prevent sniping at launch, mostly bad, and problematic because they flag Dexscreener.

Here I’ll list a bunch of things that are relatively trivial to evade for a botter. That doesn’t mean you shouldn’t do them, just realize that they’re not an effective strategy on their own. But most of them also negatively impact real users.

Note: ‘wallet’ below is shorthand for either the wallet address the token goes to, or the transaction initiator (they can be different things, especially when botting).

• Blackout period - token cannot be transferred in the first X blocks

  • Lockout period - if the token is transferred in the first X blocks, it’s locked forever for that wallet

• Time limits - wallet can only transfer X times within Y blocks

• Amount limits - wallet can only contain/transfer X% of the token

• Tax period - see decreasing tax section above

So how should we launch tokens?

I’ll say it again, don’t, it will just end in tears. No matter what you do, some people will make money and some people will lose money, and the people who lose money will be angry at you.

But if you do want to launch a token, then there are some things I think you could do to reduce the number of tears.

Don’t stealth or fair launch

Those things are a fucking train wreck. They will go to zero and your reputation will be shot. You’ll never beat the charge that you sniped it.

Do test obsessively

You only get one chance at launching. Test everything, a lot. Have a checklist. Don’t rush. Don’t only have one developer, get a second opinion.

Don’t vest

Unless you have some kind of strategy with very long term vesting and some kind of fee-driven buyback utility, best to get the tokens out in circulation immediately.

Do pre-launch price discovery

Personally I like the LBP and Dutch Auction mechanisms because they give some kind of a way to evaluate demand prior to the token actually being live.

At a bare minimum, you should do a presale if you have any kind of real demand. A presale will ensure that a large % of tokens (typically this has been 70% or more) get into the hands of a lot of people at an equal price.

Don’t limit presale to pump price

Everyone wants to make their presale exclusive so that people will have to buy on the open market and pump the price. Everyone loves ‘token go up’.

I feel like we’ve seen how that goes, and it’s not pretty.

I think most of the issues relate to a mismatch between demand and supply at launch. If more people bought tokens prior to launch, the price would be higher, and there wouldn’t be as much instant appreciation as everyone rushes to buy.

Do provision single-sided buy LP for your presale

I don’t really get the fascination with launching on 50:50 LPs. Everyone who bought into the presale got those tokens at a specific price. Why should anyone have to sell tokens back under that price?

Launching on a 50:50 LP just means that the people who dump faster, get more. Set a floor for your token instead.

Don’t fuck around with your LP

People are always like ‘devs plz pull LP to pump price’ but they don’t understand that reducing LP also means when people dump, they dump harder.

Just stop fucking around with the LP. You don’t really know why you’re doing it, and your justification is probably stupid.

Do keep a dev allocation instead of sniping your launch

This shit never works out. It’s very easy to detect a dev sniping their own launch. Just keep a dev allocation and put the extra money you were going to snipe with into the LP.

Got milk?

This post has gotten quite long, but I’m sure I’ve forgotten some topics. Feel free to ping me to ask me to add content on anything in particular.

Launching via a custom token

Historically, if a deployer wants to customize their launch (perhaps attempting to dissuade snipers) they put custom code into their ERC-20.

ERC-20 is a standard; all ‘tokens’ created on an EVM follow it religiously (lets ignore rugs for now), to ensure that their token can be used by things like DEXs, wallets, and explorers without any issue.

As a developer, there are a few places that you can hook custom behaviors into a token. The most common one is the ‘transfer’ method. Here’s a famous example from $coq that limits each wallet to have only max 1% during the launch phase.

People have gotten more inventive over the years and created various types of anti-sniper features, but you’re still very limited by the information you have in this method, and there’s really no way to require additional information and still have a vanilla DEX work.

Take my word for it that there are a lot of mechanics you can use to evade anti sniper checks in your token contract.

Some of these can be combated, but there’s really no mechanism to fight against a sniper submitting tx from 10 different wallets.

Additionally, the various different mechanisms occasionally have the unwanted side effect of causing red flags to pop up on Dexscreener, which can hurt a token in the long run. I’ve highlighted a few that commonly get flagged even by well-meaning tokens trying to do anti snipe.

Custom Initial Distribution

One possible way to solve this issue is to have a customized method for initial distribution of the token.

A presale is one example of that; people send money, and then they later receive tokens, typically (but not always) at the same price as everyone else in the presale.

After the presale is complete, trading from that point onwards takes place on a vanilla DEX.

Custom DEX

There are a lot of factors that go into launching a token, but let’s say that you’re in a position where even after the presale you expect that there will be a lot of demand for the token, and you want to ensure that the initial buys are ‘fair’, while allowing people who want to exit immediately to do that. What do you do?

One option is to do something so custom that there’s no way to prepare automated buys (snipes) for it, which is what we did for the $blub, $woof, and $lok launches.

In those launches, we created a temporary DEX + UI for users to buy and sell on prior to migrating the liquidity to their final home on Pharaoh/Trader Joe.

The temporary DEX had a few anti snipe mechanisms that aren’t possible to implement directly in a token.

How buying tokens works

This is basically the way that it works (thanks Claude). The router is what most users use to swap the tokens. Users just give the router the address of the tokens they want to swap, and it resolves them to the Pair that has the liquidity needed for the swap.

A ‘custom DEX’

It’s relatively simple to just deploy your own Router and Factory, and create a Pair for your token. It just won’t be one of the ‘official’ routers that are supported by tools like Dexscreener.

But, problematically, there are many sniper tools (e.g. Sigma TG bot) that use their own custom router for buying tokens (so they can tack on a fee). These routers will automatically interact directly with the Pair when they see liquidity added, regardless of the ‘officialness’ of the factory/router.

Basic custom anti snipe features

The most basic thing you can do to stop people is add a password to the swap method.

Not very imaginatively named, but it gets the job done. You have to also give the password to the pair as well, since remember that they can skip your router entirely.

And, importantly, remember to set a max buy limit. Would be silly to just let the first person buy 2000A of your token manually and call that anti sniping.

With that, you have the basics of a ‘hard to snipe’ LP. Yes, it’s snipable if you know that the LP has a password in advance. And yes, it’s possible to circumvent even with a bit of time to research.

But trying to knock that out on an unverified contract in 5 minutes will be a bit challenging for even a qualified dev. Once the first few minutes are up, are you even in ‘snipe’ territory?

Advanced anti snipe features

But that’s just the basics.

If you’re doing a launch like this you ALSO control the UI, and the UI sets quite a few additional parameters that you can interrogate in a custom DEX, unlike in a vanilla shared DEX.

The amountOutMin, deadline, gas limit, and gas price (caveats apply) are all set by the UI, and available to the router. You can make them specific values, make them relate to each other, make them relate to the user, track state in the contract etc. As long as your UI knows what the magic is, it can make sure the TX it generates complies.

Maybe it’s viable to have an LLM quickly scan the anti snipe mechanism in the UI and build a tool, but I really doubt that beats just pressing buttons as quickly as you can (for now).

Outcome

This has come up again recently, but I’m 99.9% confident that no one automated a purchase on $blub.

Slightly more likely that someone did on $woof, but I added additional anti snipe between for that one and you can even see where someone attempted to snipe and failed.

On $lok, one person did manage to snipe for a 400A profit. Sadly I had thought of a mechanism to prevent that type of snipe the morning of launch, but decided that it wasn’t worth risking the launch, and had my record ruined.

Downsides of a custom DEX

OK so obviously this is a good way to make sure only humans are buying and selling in the first few minutes. But that doesn’t necessarily make it a good plan.

Building and testing

It’s a fair amount of effort to go down this route and even get it working at all. And then you have to do an incredible amount of testing to convince yourself that you did it right. And even then, it might not work right at launch.

One suggestion I have is that you can use Anvil from the Forge project to fork mainnet and do all your testing locally. This eliminates any chance that someone will spot your preparations.

Someone spotted Wink’s anti snipe by catching a single mistakenly verified contract on testnet that had ‘wink’ in the folder name (not even in the token name).

Hosting

You also become responsible for hosting the UI. The Blub presale site and Ferdy Fish were both DDOS’d prior to launch, so be ready to handle that in a worst case scenario. Hell, I’ve seen people fumble website launches even under normal user human load.

One thing we did for these launches was create and publish two different websites, both protected by Cloudflare, but one with the hardcore anti-DDOS measures turned on.

Launching

Doing the actual launch is even more stressful than a standard launch, since there are more moving parts. And there are more things to fuck up. You only get one chance in web3.

Although it got easier over time, we had a 22 point checklist of items to execute for $blub, and I had a terrible migraine when the launch was complete.

Even though we tested extensively, we still had a few steps of the launch plan fail and had to scramble to fix.

Chart

One problem with this mechanism is that Dexscreener displays the history for a token per liquidity pool. And your LP is temporary, only available at launch, before it migrates to a permanent DEX.

This means that the ‘initial PA’ for your token gets chopped off. If you launch at 1M MC, then migrate at 10M MC, and dip to 5M MC after migration, it looks like your token is down-only even though actually it’s 5x up from launch.

Price action

As mentioned above, you do lose that initial PA, which can be very bad. On the other hand, it does basically eliminate what you might call the ‘$$AIGG price action’ where a sniper buys $50K the same block that LP is added and then you’re down only forever.

Conclusion

I think this mechanism does solve the problem it’s asked to, but I’m not sure it’s a good idea, regardless. I’ll talk more about alternatives in my post on how to launch tokens.

But it’s likely that you don’t actually understand too much about the mechanics of how much you’re paying and why. Hell, even the people running one of the largest protocols on Avalanche don’t seem to understand how it works.

So I’ve written up a short explainer on the topic! I’ll try to keep it ELI5, enjoy.

Transaction types

The EVM has two different ‘transaction types’.

Type 0 is the ‘legacy’ type, where you specify the exact gas price you want. You always pay exactly what you specify, assuming your transaction is accepted.

Type 2, aka EIP-1559, where you specify a ‘max price’ and a ‘tip’. The max price is the most you’re willing to pay; in practice you will end up spending less. The tip, on Avalanche, is mostly relevant for in-block transaction ordering.

If you’re writing software to send transactions, you should (almost) ALWAYS use type 2 transactions to save users gas.

Don’t ask what happened to type 1 transactions.

RPC nodes, validators, mempool

When you submit a transaction, you’re typically sending it to an ‘RPC node’, not a validator directly. The RPC node will propagate it to multiple validators, who will also gossip it to other validators.

The validators all have a pool of unmined transactions (the mempool) that need to get included into blocks. Every 1-2s the next ‘leader’ will build a new block from the contents of the mempool and broadcast it out to other validators to reach consensus.

Block building

The leader wants to build a block containing as many transactions as possible (assuming they pay enough gas). It won’t randomly skip a transaction it knows about in the mempool that could be included. Two things limit the TX that can be included.

Gas price vs units vs cost

People say ‘gas’ a lot but they’re not always specific about what they’re talking about. There are four things they could be referring to:

• Gas units - when a TX is executing, it consumes ‘units’ of gas. It costs 21,000 units of gas to do any transaction, e.g. to transfer native. It costs more depending on what you do; e.g. to send native to 2 addresses, it will cost about 26,000 units.

• Gas limit - a TX needs to specify how many gas units it’s willing to consume to complete. If the gas limit is less than the actual amount of gas used, the transaction fails.

• Gas price - the amount the TX spends PER UNIT to execute.

• Gas fee - the total amount the TX actually consumes, which is gas price * gas units.

Block gas price

Each block has a ‘base fee’ that needs to be met in order for a transaction to be included. This fee is dynamic, updating over time depending on load.

Assuming every TX in the mempool is willing to pay the minimum fee, the leader will attempt to fit every transaction available into the block. TX in the pool specifying below the current fee will be delayed to future blocks.

Gas limit

Every block has a limit on the amount of gas that can be included. Historically this limit is 15M units of gas, but recently became dynamically adjustable.

Every TX also has a gas limit, but the thing that matters to the block is the amount of gas USED by the TX. The sum of all gas limits for all TX in the block could be enormous, but the sum of gas used for all TX in the block needs to be below the limit.

The leader

Every validator knows which validator is eligible to produce the next block (their ‘slot’) through some complicated math. If that validator doesn’t produce a block, other stuff happens but it’s not that important, slots are rarely skipped.

All you need to know is that typically it’s the job of one specific validator at a time to stuff transactions into a block.

Inclusion and ordering

On Ethereum and some other chains this is more complicated, but on Avalanche it’s pretty straightforward.

The transactions the leader knows about (its mempool) are filtered by fee (every tx willing to pay the current block fee), then sorted by fee (total amount for type 0, base cost + tip for type 2), and processed in order until the gas limit is reached.

The fee you pay determines which TX comes ‘first’, but as long as the gas limit isn’t reached, any TX willing to pay the minimum fee gets in.

Block fees

Great, so how do we determine what the actual fee is for a block? You might remember that for the last few years, the minimum fee you could pay was 25 gwei.

Recently, ACP 125 updated this fee to 1 gwei.

Even more recently, ACP 176 reduced the minimum fee further (I have no idea how it works now)!

But the fee has always been dynamic above that minimum, in proportion to the amount of gas consumed by recent blocks. The old mechanism used a pretty ridged windowing system that checked the amount of gas consumed against a ‘target’ value.

I don’t fully understand the ACP 176 update, but it seems to use a more flexible mechanism that will scale up more smoothly and scale down more quickly. You can review the simulations if you want.

The goal of increasing the gas price is to keep validator requirements reasonable. The TLDR is ‘more gas used, higher gas price’.

Case study

It was by no means a new and unique idea, but at one point I wrote some scripts to consume all the gas used on the chain for about a minute, effectively blocking out a series of lower priced transactions so I could win a contest.

During this exercise, I ended up spiking the gas price and costing myself a big chunk of avax to pay those fees (worth it). But this was under the old regime of 25 gwei minimums and the fixed window size gas pricing.

Typically nowadays gas price only spikes high when there are violent price fluctuations, as bots race to liquidate people’s positions and arbitrage liquidity pools. The ‘normal’ elevated load rarely exceeds 50 gwei, even during token launches.

But I am curious to know how this works now during a period of significant demand. Rather than speculate, let’s test! I wrote a quick script to consume a lot of gas and send out some TX from a few different accounts.

I consumed enough gas to do about 350 swaps, plus there were some other user TX in there. Let’s check what happened to gas before and after.

It only went up by .5 gwei, not a particularly significant increase.

Back to the start

Alright, so let’s review what actually happened during the $boi launch. What was gas usage like?

Pretty easy to spot where trading opened, and you’ll notice that all the blocks after that are well under target gas, much less the actual gas cap. Starting gas was .12 gwei, 30s later it was .26 gwei, 30s later it was .38 gwei.

I don’t know what you consider a ‘snipe’ but 60s after launch seems well out of snipe range to me.

So very, very wrong

Let’s review the cryptofish special from yesterday (that’s when you double down every time you’re confronted with evidence instead of admitting fault).

Yes, blame the user for your shitty code, off to a great start.

Would take less time to look at the explorer and disprove this than it took to type.

The TX being delayed by a single block seems obviously unrelated to the TX failing.

Why would you give an example that is so blatantly incorrect?

You can pay any fee you want to get included, it has nothing to do with the amount it COSTS to get included.

No… no I did not. And I didn’t ask that because that information isn’t relevant to the issue at hand.

This is obviously someone using Chat GPT (poorly) to reply. My money is on cryptofish but maybe they hired a retarded intern after Blue left, who knows.

Anyway, very ballsy of a web3 protocol to go the route of insulting their users, provide factually incorrect information, and double down after being corrected.

My goodness. The Dokyo team DUMPING on you? Let’s investigate. You can follow along in my Google Sheet but I’ve included the relevant bits below.

Side note: Dokyo videos

These videos are getting INSANE. Is he narrating them now?

Let’s just ignore the fact that he misused the apostrophe in the FIRST WORD IN THE VIDEO after making fun of Goose’s (this is the correct usage) Fiverr editor.

Did this mother fucker literally just compare himself to Aristotle, Plato, and Jesus Christ?

Goddamn. What does ‘revered as fools’ even mean? Is he literally trying to trigger me to death? It’s working.

Back to the mint

I figured I should start from the beginning. Who minted what, where did they go?

OK so a lot went to the presale, a bunch were minted by the team, the Dokyo Hero is there of course, and then a few large individual team member mints, and then just normal mints below (didn’t bother investigating down too far).

About that presale

If you recall, the presale was 5 NFTs for $300. Everyone got the same deal, right?

Looks like that’s not the case!

SECRET UNVESTED ALLOCATIONS, TEAM DUMPING, WARNING, WARNING!

Seems like a bunch of team members and KOLs got EXTRA LARGE ALLOCATIONS.

That’s grossly unethical! I cannot BELIEVE that Brandi would approve of such crimes.

Looks like Brandumb got a lot of value out of this. “Rugs like dokyo”, indeed.

Dumping from the treasury

Wow! Instead of using their mint funds and royalties, they’re literally giving people Dokyo to dump on you as payment for… services?

Let’s take a look.

I guess it’s not that surprising that they started dumping Dokyo on you, given that the mint/royalties have all been spent on ladyboys and drugs.

Lot of comedy gold in here. For example, it looks like Hazy in particular is absolutely RUTHLESS about just nuking them into bids.

One transaction, accept all standing offers, thanks for the avax.

It’s hard to say who some of these are… given that many of them are funded by the Dokyo team, nuked into bids, and then immediately transferred to a CEX.

Very curious behavior. Fresh wallets, zero AVAX ties except funded by team, given Dokyo, straight to CEX.

Keep your heads down and build!

That’s all

Short post! Short and sweet. I love the taste of hypocrisy in the morning.

So glad we have an expert in crime like Brenda available to comment on other crime in the community.

Looks like only another 233 Dokyo left for Brandowns to dump on you. Delist now, before this shit goes to zero!

I wrote this article to crystalize a piece of Avax lore for future generations, and to serve as a warning for anyone who is thinking of buying into whatever grift he works on next.

The before times

I don’t know much about what Brandi was up to before he got to Avalanche. I’ve heard rumblings that he ran some kind of NFT pump and dump group on Solana, but I guess he wasn’t that important since I’ve never been able to find a first hand source who could comment on it for me.

Anyway, we once had this conversation which seems to align with that, you can take your own conclusions from it.

Dokyo

Pre-launch

It’s an open secret that Dokyo was funded from the ground up by a grant. The goal was to attract some attention to the ecosystem from SOL/ETH people and to try to kick off an NFT season in conjunction with the Hyperspace launch.

As part of this, Brawndo reached out to a bunch of KOLs to offer them, gasp, a DISCOUNTED PRESALE to mint Dokyo. On top of that, the money was collected as USDC on SOL or ETH instead of AVAX, how scandalous!

As a niche micro influencer with a tech blog, I was of course included. But I find it hilarious, given the state of things, that Branduh once launched a presale that went out of his way to include KOLs who never posted about it.

But between the slick marketing videos, the ecosystem backing, and the seemingly high quality art for the project, it was likely that this would be a decent gamble.

Launch

The launch did not go well. Hyperspace had issues (surprise!) that affected the ability of people to mint. A combination of this outage, high mint price, and malaise over the market caused minting to stall out after a few minutes.

This is typically the kiss of death for a project; when it doesn’t mint out, people start listing below mint price, and a downward spiral occurs.

Fortunately, a hero arose!

Someone minted 10K avax worth of Dokyo. A very diamond handed someone. So diamond handed, they hold to this day.

It completed minting out shortly after 20% of the NFTs were bulk minted by this user.

Sidenote: what about Joepegs?

Until this point, Joepegs was the place that most new projects launched. So you might wonder why Dokyo launched on this new, poorly tested, honestly kind of shitty marketplace?

Well, it turns out that a few people from the Joepegs team talked shit about Brando in a Telegram chat that actually included Brando. Whoops!

Post-launch

This was an exciting time for Dokyo! ‘Successful’ mintout, basically maintaining floor at mint price, Avalabs employees wearing Dokyo PFPs!

Lots of talk about how holo masks got you access to an ‘alpha’ chat! Someone buys one for 400A! Plans for the future! Noodles!

The mint proceeds were nuked from Avax to stables between Sept 22 and Sept 25, at approximately $9 AVAX.

Right before the bull market returned. Oof. Truly astonishing how accurately they managed to bottom tick it.

Hyperspace era

The thing that really boosted Dokyo as a project was the launch of Hyperspace Trading Rewards. Under this incentive structure, users were encouraged to wash trade their Dokyo (and other NFTs, but primarily Dokyo) with the understanding that they would be rewarded some multiple of the fees they paid.

Although Hyperspace paid a fixed .5% creator royalty, the massive volume created by these incentives resulted in an enormous influx of money to Dokyo as a project. They actually received even more than they minted for, in USD terms. At one point, they were earning $15K per day in royalties.

It also drove the floor price of a Dokyo up to something like 60A, worth more than 1 ETH at the peak.

Slow decline into irrelevance

What goes up must come down. With the end of the incentive programs, the volume dried up, and those 60A Dokyo crashed down to as low as 3A.

Of course the mania was never going to last, but the project wasn’t doing itself any favors.

What went wrong

More like, what went right? Or perhaps, what happened at all?

Noodles

Brambo’s big promise was that Dokyo would ‘disrupt the noodle market’. Which is, frankly, the dumbest thing I’ve ever heard in my entire life. And I am old.

Unsurprisingly, nothing ever came out of this. Ignoring the fact that making noodles for an NFT project is a ridiculous idea, feels like it should be pretty obvious that creating and distributing a food product is a complicated and costly endeavor.

I suppose if your life experience is ‘barber and NFT founder’ maybe you lack the intuition necessary to warn you away from a plan like that.

The best way to learn is to try and fail! Lots of failing going on here. Less clear how much trying is happening.

Team

My understanding is that the studio that created the Dokyo art was contracted to do the work, and then departed. There was, at the start, a community manager? Not for long, I assume, given the amount of activity in this community.

For reference, I count 15 posts on Dokyo’s Twitter from launch until earlier this month, and more than half (9) of those posts happened prior to Dokyo launching.

So confusing. How do you know what you’re doing wrong, and still not take any action to rectify it?

What team changes are we talking about here? Did anyone ever get hired? But why does he even need to hire someone? Just post yourself, Brandope.

Merch

They did manage to ship some merch! They shipped (checks notes) a few jackets?

Hope Dokyo holders are thrilled that they managed to hand out some merch to random girls in Japan.

So they handed out some t-shirts at conferences and a few jackets to special people. What a fantastic run! Hope all of you Dokyo holders managed to fly to Japan to get your free t-shirt. After all, isn’t that what NFT projects are about?

Conference hopping

So Brandork didn’t actually work much on Dokyo. What was he doing with his spare time? Traveling the world, obviously. What else do founders do with your mint funds?

Fantastic.

I see a lot of connecting and collaborating. Really bringing the Dokyo brand into the spotlight. Huge value add.

He’s BD-ing, bro

One retort that Brandull likes to use is that he ‘brings attention and projects to the ecosystem’.

Where are all these projects? The only one that he points to publicly (correctly or not, who knows) is NFT Inspect.

Famously shitty NFT Inspect. Bargain bin Kaito. The project that told people who complained that their PFP wasn’t being recognized to reboot their computer. Their algorithm was so bad that they had people typing one-word replies to shitty KOLs to try and boost their leaderboard ranking.

How about Hyperspace? I believe he’s at least partly responsible for them. Hello? Hyperspace, are you home? Are the servers still running? Can someone page Bryan?

What about Looty (the original Looty team, not the new one). Remember the bottom of the barrel SOL scumbags Looty? I wrote a 4-part series on how shitty they are Looty?

Fantastic BD work buddy.

Fwog arc

This isn’t that funny but I found it really ridiculous that he spent so much time bullposting Fwog instead of posting about his own fucking project.

Founder mode

Let’s not forget all the valuable inspirational quotes he has ChatGPT’d over the last year to keep us all motivated.

This one is so topical too, I love it.

This shit is a failure, obviously

What’s amazing about this whole thing is that it didn’t have to be such a complete failure. They had great art, strong backing from the ecosystem, and an insane amount of incentives/funding funneled to them.

He just… didn’t do anything. That’s the whole story. Complete abandonment of the project in order to fuck around in Asia for a year. Hired professional photographers to follow him around and watch him jerk off.

I mean, I can see why. Who wouldn’t accept $700K to travel around Japan, Korea, and (presumably) fuck ladyboys in Thailand? All while doing a copious amount of drugs (presumably).

I assume the drugs, because I don’t understand how you could be so delusional about everything without them.

For Christ’s sake, the guy doesn’t even remember what his own collection looks like at this point. He confused an honorary from another collection with his own.

Finally up to date

Wow! That was a long story. Given how badly he failed, you’d think he would slink away into the night, right? Well, sadly, it seems like Brandnope’s Dokyo royalties have fallen upon hard times.

All cashed out! No more AVAX to send to Kucoin to nuke for ladyboy and cocaine funds. You know what that means?

I can’t wait for Dokyoke to launch. I NEED this energy drink. Or memecoin. Or NFT. Who can say? I assume there’s a plan! No plan for Dokyo for the last year, but definitely a plan now for this thing.

That’s right, Brenda is BACK baby. And he’s ready to grift, again. He’s got a CM who has been spicy on the timeline and some Fiverr artists (better than Chef Goose’s) to make some new shitty inspirational videos.

Attacks on, well, everybody

As part of his coming out special initiative, Blandork has been slinging shit at literally everyone who has been (unlike him) actually doing things. Anything to get some attention I guess, even if it’s not positive.

Well, let’s be fair. Nobs and ChefGoose probably had it coming. They’ve been dunking on him continuously for the last year, and he mostly ignored them, which takes a remarkable amount of restraint. But, again to be fair, all of those dunks were justified. The guy has literally done NOTHING, and he deserves all the shit he got.

There are so many funny things about this crash out (is it a crashout? or is this a media stunt?) that it’s going to be hard for me to get them all. I’ll do my best!

Public reaction

I laugh every time I read the replies to one of his posts. I’m just going to grab a random one right now and I’ll tally up the sentiment really quickly.

Ahaha, this one is so funny. Mom! I’m a gatekeeper! I decide who wins and loses! Mom, he caught me botting mints! He must not read my substack, where I wrote about it, which makes me sad. Maybe he’ll read this one?

Anyway, lets review the stats:

• 146 ‘Likes’ (totally real people)

• 56 comments

  • Negative: 46

  • Neutral: 5

  • Positive: 2

  • Inconclusive: 2

• All 5 quote tweets are negative

Fascinating. Almost seems like his socials are botted!

And here you can see Blandoh’s alt talking to Brandslow’s alt.

Every accusation is a confession

Ignoring the fact that I have tweeted repeatedly to make fun of the stupid shit Branflake has said, this ‘send Discord minions attack’ line is SO FUNNY.

No one has ever sent anyone to attack him… because there’s no need to. Everyone hates him. The community thinks he’s a literal clown. If he posts something, an army of people will dunk on him of their own volition, no coordination necessary.

He doesn’t understand this because he has no friends. No one comments in support of him, because no one supports him. He has bots reply to his posts and like them. He assumes anyone who is against him must be paid, just like his bots are paid.

Comedy gold

I can’t get over this one where the guy he quote tweets tells him that he’s wrong about his interpretation and then his response is to attack him.

Chain analyzoooor

He’s also discovered Arkham, and I am LOVING this arc.

Yeah that’s right buddy, I bought 25K of ‘WWW.TETHER.COM’.

Stealthy dumping devs

It’s hilarious how hypocritical and inconsistent the accusations are.

I can’t believe I gave myself up like that. How could I know that by creating dashboards that show who is selling, my own selling would be revealed! Foiled again.

Of course, the idea that ‘discovering’ the behavior of someone who built a dashboard that exposes that behavior USING that dashboard might be a tad bit inconsistent is simply too much for Brandont to process.

Do you think he ever figured it out? The math might be a bit too hard for him.

Creepy as fuck

I actually came back to edit this in after publication when someone reminded me. I can’t be the only one who gets a weird, creepy, sexual vibe from this kind of talk right?

KOLs with back-channel

Brandslow has some very strong opinions on KOLs getting into presales. Which is why I find it hilarious that he didn’t bother trying to apply to Blub.

Instead, he DM’d someone trying to get in after applications had already closed.

I’m sure Branfaux appreciates the upstanding integrity of that person trying to preserve the fairness of the process and that it has no impact on his current crashout arc.

This post is soooo looong

I think I’m going to have to cut it off here.

The TLDR is that Branslow got handed a shit ton of money, fucked ladyboys/snorted coke/partied at conferences for a year, and is now spending more effort crashing out on other people than he ever did for his holders.

For fucks sake man. What were you DOING ALL THIS TIME. You had all this creative talent and you just sat on it! We could have been enjoying this clipart the whole time!

Anyway, fair warning that he’s a grifter and grifter’s gonna grift. I assume this latest crashout is just a way for him to justify abandoning Dokyo, probably so he can launch some new scam.

Can’t wait to see what it is!

Full disclosure: although I like a lot of Joe employees, I intensely dislike their founder, Cryptofish. The Pharaoh guys seem nice enough although I haven’t interacted with them much.

As always I’ll try to be neutral and review the facts but assume some bias on my part.

Stakeholders

Some background on the main characters.

• Cryptofish (aka cryptofishx) is the co-founder of Trader Joe (aka LJF), a multi-chain DEX that originated on Avalanche. It expanded to Arbitrum, among other chains.

• LameHillbilly (aka LHBCrypto) is the co-founder of Pharaoh, a DEX on Avalanche that is under the umbrella of the ‘Nile DEX’ organization, which is multi-chain.

• Ramses is the Arbitrum part of the Kingdom, which is where the misconduct was alleged to take place.

• The Arbitrum DAO has a voting system for allocating funds to ecosystem participants.

• Entropy is some kind of paid advisor organization that raised the concerns to Arbitrum DAO after being alerted by a third party.

The drama

Lots of drama coming out of the drama queen himself, cryptofish.

Oops how did that last one slip in there.

The allegations

You can read them in full here. In short:

• Ramses LP are not immutable; the fee tier can be adjusted by an administrator.

• In this case, it was alleged that an administrator was disabling the fees, performing swaps to arbitrage pools, and then re-enabling fees.

• In a traditional DEX, swap fees are the way LPs make money.

• It is alleged that the fees were retained by the swapper, making this a straightforward case of theft from the LP.

Sidebar: does this even matter?

Ignoring the fact that I’m spending all this time writing a summary of this situation over one whackjob crashing out on the TL, is the crashout even aimed at the right person? And how serious is the issue, anyway?

How are Pharaoh and Ramses related?

My understanding is that Ramses is part of the ‘Kingdom’ organization. The leaders of that DEX also run several other DEXes on other chains, but Pharaoh is a franchise and has an entirely different leadership and is only on Avalanche. The developers are shared between all the DEXes though.

So it’s not entirely clear to me how wrongdoing on Arbitrum would be relevant to LHB personally, and he seems to be the target of the current crashout.

I will continue on regardless, because I’ve become curious to know if this is a real issue.

How much are we talking about here?

I’ve heard different estimates of the amount ‘earned’ by the bot here, ranging between $4K to $8K. Maybe it’s an asshole thing to say, but that’s not an enormous amount of money.

I mean, I have firsthand knowledge of TJ misplacing $10K of JOE rewards and no one ever made a big deal about that. I’m sure there are many similar situations over the years, for both dexes, for most protocols even. We’re talking about enormous amounts of money changing hands in web3, 4K-8K is pretty negligible.

But, on the other hand, I do believe that ‘tone from the top’ exists. If you’re willing to wave away small amounts of money instead of ‘doing the right thing’, it’s not a great sign.

A review of the forum post

Here’s the summary of the allegation by Entropy:

It has been alleged that Ramses is lowering fees to zero, enabling the team to capitalize on arbitrage opportunities within its own pools, before raising the fees back to normal levels for other traders. This practice results in LPs receiving no fees during this toxic flow trading, which could be deemed unethical.

Ramses replied to this, but the summary of the response from Entropy is more concise:

LPs on Ramses do not capture most fees, token lockers do.

The team’s plan is to eventually return the extracted value to token lockers and LPs.

The same arbitrage would be captured anyway by other bots.

The following post by Entropy summarizes what seems to be their conclusion

Our overall assessment is that if the team commits to returning past value extracted and any future value extracted to LPs/token lockers, that they never had malicious intentions. They have long contributed to arbitrum, and we think they deserve the benefit of the doubt here. Do we think there could have been more transparency with this MEV experimentation? Yes we do. And Ramses team agrees.

There’s a very post by ‘dk3’ who appears (not familiar) to be someone authoritative to this process that basically says this will be reviewed further, but seems relatively accepting of Ramses’ response.

There’s also a post that references what seems to be a now-deleted post claiming that the funds were bridged off-chain.

No discussion past this point, and I haven’t been able to find any kind of follow-up.

Initial thoughts

The technical

Presuming that this was all done above-board, the explanation for what they were doing and why they were doing it seems very reasonable to me.

With a few caveats of course; all pools involved would have to be gauged, and the majority of the profits would have to be returned to the pools themselves.

Obviously the pools would have eventually been arb’d by someone, and that person would have kept the excess profits. In the meantime, due to the overhead imposed by the fee the prices would be incorrect. This is basically break even for users, some might get a better price, some might get a worse price (net of fee).

So I think that assuming the caveats bear out, they’re cleared on the matter.

The political

Hashing things out in public like this has a lot of different competing incentives for different parties. What’s not said can be as important as what is said.

In this case, the straightforward read to me is that given the lack of follow-up, and the end state of the discourse, the relevant parties accepted the explanations and cleared Ramses of any wrongdoing.

There is a claim (I have not verified) that the original poster apologized for the post.

If you wanted to take the most negative view on this situation, you could maybe say that this just got swept under the rug, given the lack of follow up on the post.

But you could just as easily say that maybe it just got worked out behind the scenes and no one wanted to inflame things further.

Hard to say! People are complicated. Absent other evidence, I’m leaning towards ‘it got resolved’.

Thoughts on the general idea

Arbitrage will happen, and is a good and normal thing. If this value can be captured by a dex and returned to users of a dex, this seems like a straightforward value add to the LPs. That money would have gone to MEV bots. Might as well give it to the protocol users instead.

As long as you can confirm that none of the protocol users (in this case, LPs) are worse off, this is a great feature. In fact, I’m surprised this is not more widely done.

One thing I will note is that I personally dislike that the ability to mutate fees like this exists. They describe in their documentation why this ability exists, but it feels gross to me. The purist in me says that things that DO need to be dynamic instead of immutable should be automated, perhaps via a keeper.

Investigating the facts

Alright, so down to the nuts and bolts. Here are the questions I want to answer.

1. Were all the pools gauged?

2. How much did the bot make?

3. What happened to the money the bot made?

Assuming I like all the answers to these questions, I think they’re cleared of any wrongdoing. I have to do this work myself because Cryptofish is probably Brando quality at chain analysis. I’m going to limit my inquiry to Arbitrum for simplicity.

The account in question: 0x6e639af00851684e85913197a7d36bf3aa6b5a9a

Damn. That is a lot of transactions. Unfortunately it’s above the limit that arbiscan will ship you transactions for, so I’ll have to do some work in Flipside.

It seems to interact with a few contracts to do it’s MEV work:

• 0x50ecbc4ee95869365a688f2cdf977e37ecb66f25

• 0xc1edf4c712d062bce9f3d65178ba828f32be3022

• 0xe8b1cb012b27b3bfcadc8b74ddf00d35d22deafc

After filtering out the MEV, there weren’t too many actual transactions to look through, thank god.

Funds transferred out

First place to start, as always, is in Arkham. Lot of stuff going on here, but I do see some outgoing transactions to what I assume is the bot dev on there.

I found this bridge of 2E to Avalanche, which was mentioned in the post.

Most recently, I see the remainder of the funds getting swept out:

Weird stuff

There are some tx that are tiny value and don’t mater, not sure what they’re about:

• ~.0001E bridging rsETH many times

• Very tiny swaps for fraxETH

There are some small swaps for USDC to/from ETH, it’s not clear what they’re for.

I only found one transaction that I found to be suspicious.

Normally for an arb swap, you expect to see something like this, where the originator doesn’t send anything but receives a small amount of eth as output.

You don’t see that here. Instead you see just a plain swap going through the feeless contract. I inquired about this and was told that this was part of a short-lived test for cross-chain arb support.

Summary

Mixing the cross-chain activity with the local-chain activity makes things difficult to tie out. VERY DIFFICULT. But reviewing the bridge activity, it seems like a very small amount was initial bridged in (<1E) and the other bridges seem to be in pairs and net out.

Two of the contracts stored their balance in the contract and were withdrawn [1], [2] for about 2.5E total.

One of the contracts would swap WETH to the bot, and use part of the leftovers to continue swapping, meaning the WETH would accumulate in the account. This makes it difficult to calculate earnings, but it seems plausible that it’s the amount converted here [3].

So in total, about 6E earned to the bot.

Note that this is BOT EARNINGS; the amount ‘diverted’ from the voters would be less than this, although it’s hard to quantify how much less.

Missing info

Two bits are still missing here:

1. Were all the pools gauged?

2. What happened to the funds

#1 is hard to answer. I’m not sure it’s even worth looking into. I asked and was told that 95% of pools are gauged.

#2 is a bit more obvious. It seems like most of the funds sat in the bot wallet for a long time before being moved only recently. I believe the original post in the thread that ‘the funds were bridged to Avalanche’ was incorrect.

Final thoughts

I was told that some of the funds were given to the bot dev as compensation for the work. Although I applaud paying devs, that seems like an inappropriate incentive structure for work like this. The pay should be the pay, and the funds received from the activity handled appropriately.

It also seems like the remainder of the funds were not used to pay voters, although I await any transaction evidence counter to that. It’s possible they used other funds to do so, perhaps after being confronted with the evidence, but perhaps because they always planned to do it at the end.

As I mentioned above, this is not a lot of money we’re talking about, and I doubt anyone had malicious intentions. This feature seems like a great value add, assuming it’s implemented appropriately.

But it’s very easy to accidentally give the appearance of impropriety when there’s no such thing intended.

So is Cryptofish’s crashout actually justified?

I would say the rough outline of his complaints are accurate, but hysterically overblown, and as far as I can tell mostly aimed at the wrong person(s).

WolvesDAO is… to be honest, I’m not the guy to explain it. I’m not that in touch with gaming stuff, or anything outside of the few Avax communities I frequent. But they have a nice website, and a Substack with a lot of content, which I appreciate.

A lot of people have reached out to me over the last few weeks looking for assistance launching projects, and I’ve had to turn a bunch of them away because I didn’t know them, or they seemed too sus, or frankly I just had too much work to do.

But WolvesDAO came recommended to me from a number of people I respect. So when they asked for help launching their upcoming token $woof, I agreed.

$woof

I’ve made no secret of the fact that I’m skeptical of most AI based coins. Not that it matters much to me as the deployer, the pay for deploying a non-AI coin is the same as for deploying an AI reply guy coin.

But it does seem like they have more interesting plans for this coin. Apparently they have been collecting web3 gaming related data for a while and plan to integrate it into their AI? I assume you’ll need to have X coins to access their token-gated services, or perhaps pay in the token, not clear from their docs.

Anyway, at the time I accepted this job I was stressed out from preparing for $blub; after the success of the $wink and $ket launches, pressure was on to keep ahead of advances in sniping. So I was pretty happy when they said they just wanted a vanilla launch with no shenanigans.

Circumstances change

Unfortunately, once things heat up in an ecosystem, new coins start to become a target for snipers. And we saw that happen with $albert, which got pretty wrecked by snipers even though they had anti-sniper tactics built into the launch.

I’ve said before ‘you only get one chance to do it right in web3’. A bad launch can wreck you, like dead-project-forever wreck you. When they reached out to upgrade their launch from a ‘simple’ one to an ‘anti sniper’ one, my first thought was ‘they bait and switched me’!

But to be honest, I’m glad they did. I have my own professional pride, and having a launch I oversaw destroyed would be very offensive to me.

They agreed to pay me a bit more, and although I wasn’t willing to go to $blub lengths (undisclosed at this point) I agreed to repurpose a bunch of stuff from the $wink launch. I still warned them about the risk of sniping, though.

I asked them to push back their launch date a bit so I would have more time between the $blub launch and $woof launch to prepare, test, and decompress.

Things change, again

From a technical perspective, the $blub launch went off pretty smoothly. A few minor glitches, even after all our testing, but we wrapped up our launch checklist and were done within a few hours. I moved on to preparing for the $woof launch.

While I was working on it, someone asked me some questions about the $albert launch, and while investigating I stumbled across this.

Someone testing their Twitter scraper, including CA detection inline, in images, and in a linked website. Uh oh! I warned them about the increased risk level.

Even worse, someone asked me some questions about the temporary $blub router, and on investigation, I found more worrying stuff.

People manually poking at the router. The page had already been taken down, meaning these were scripted requests. My paranoia (and blood pressure) increased to never-before seen levels. I decided to see if they were willing to pivot, again.

They were, and they even agreed to pay for the extra work required (not extortion, I swear).

Final launch planning

Given the new plan to launch on a private AMM, we opted to scrub all the ‘anti sniper’ fake LP tactics that we’ve used in previous launches. Those things just punish snipers, but don’t actually make the launch any safer for real users. Better to skip the controversy.

VQ quickly reskinned the AMM front end and made a few improvements as well.

He claimed he was going to be very lazy about it, but he’s a liar. Look at this level of detail! What a sucker.

I also added a few novel improvements on the smart contract side to attempt to dissuade snipers. Rabby ended up being a problem, but that’s a story for another time.

Other than that, the only additional hitch was that the requirement was to launch on Pharaoh CL, mimicking a traditional 50:50 LP. I had some code to do this, but out of paranoia I ended up writing a bunch of tests and extra helper scripts in case anything went wrong.

We wrapped up launch planning, and I did a bunch of dry runs with 12 hours to spare. I registered all the domains and set up blank pages to make sure there wouldn’t be any DNS issues this time.

Launch

I absolutely despise people who give multi hour launch windows and then launch near the end.

I was about 20s late to start, but I wrapped up pushing everything and doing a quick test buy/sell on the router at about 10:02. Thanks to excessive testing, things went very smoothly.

Then I hit my first hitch.

The airdrop was to over 3K users, so out of extreme paranoia I had written a custom disperse contract that would refuse to airdrop to the same user twice. This came in very handy when the airdrop didn’t complete successfully.

The issue turned out to be that I had just slightly underestimated the amount of gas required to do all the airdrops. I put 5A in the deployer, we needed about 5.5A. Whoops.

After filling it up with another 5A the script finished the last two drops just in time for me to migrate the LP to Pharaoh at 10:12. I was finished cleaning up the deploy (transferring LP tokens, ownership, dex registrations etc) by 10:30.

One last hitch happened when the Pharaoh UI lagged on finding the new LP as a place to trade against. The aggregators (e.g. Odos, 0x) actually found the LP about 20s faster than the native routing somehow.

Final thoughts

The launch went pretty smoothly, compared to other launches. But I’m not convinced that this will be the case in the future, fighting snipers is a cat and mouse game. The stuff I put in at the last minute only (visibly) blocked one sniper for about 30A of buys.

However, a guy I know from a few years ago who is big on MEV and sniping on Avax (and other chains) did reach out to me to tell me some of my efforts recently had been at least partially successful. Based on some stuff he told me, an unknowable amount of bot activity was prevented; since you can simulate TX before sending them (and abort failures), you can’t use what’s reflected on chain as the final word on what people tried.

As clients go I can recommend WolvesDAO. They were very professional and organized, paid me prior to me even asking for payment, didn’t complain when I asked for pay for VQ to do the unplanned frontend work, and proactively offered me a bonus for the work. Was great working with them.

This post follows the launch of $wink, a meme coin on Avalanche. This is part 2 of 2; you might be interested in reading the guest post on the topic by 0xSmitty.

Additional context can be found in my series on the launch of $ket. I’ll try not to rehash that content too much here.

Anti bot mechanics

As part of the Wink launch, there were a number of fake LPs spun up before liquidity was pulled from them. The team warned about buying LPs, in advance, on Twitter. It was also pretty well known that I had done a similar thing, and had warned people against sniping Ket, and was consulting with the Wink team.

Regardless, people still aped into a variety of fake LPs and lost money. I think this might be partially my fault; during the Ket launch I kind of just bluffed and didn’t bother making LPs with full liquidity because I was worried about screwing it up. Anyone who investigated that and thought it would be the same here took the bait hard when the full LP was added.

I believe that in the end the team withdrew more than twice LP that they actually added due to all the insane sniping that happened. RIP snipers.

For context, the LPs didn’t have liquidity for more than 14 seconds.

What’s a sniper, anyway?

An interesting question to be sure. There’s no fixed, agreed upon definition, as far as I know, so I’ll attempt to define one.

People definitely consider buying in the same block as launch a snipe. First few blocks after, probably easy to call those snipes, gotta be using a bot.

But is it really the bot that makes a sniper? I’m thinking about the $TRUMP and $LIBRE ‘launches’. In both those cases someone bought a million dollars in the first 10 seconds. I’ve heard those buys described as ‘snipes’. What makes them snipes? If you bought a million dollars of a fake coin, would you call it a snipe?

After thinking about it, to me the difference between a sniper and a non-sniper is not the technology used. It’s the information inequality. Doesn’t matter if you use a dex or a TG bot or you roll your own stuff. If you have some way of knowing something is ‘real’ before everyone else does, and you buy before them, that’s a snipe.

Why this is relevant

I saw a lot of dumb posts like this. Ignoring the fact that none of these people will share the example tx where they bought, even if we assume that they somehow watched Snowtrace (laggy as shit), copy pasted the deployment into Trader Joe (laggy as shit), and got a buy through in time, I don’t think that this is a good excuse.

Anyway, here’s a funny montage

All of these people are well known snipers (ok not that last guy, but he’s sniper-associated).

Suck a dick.

The legal implications of this launch

Well, I talked to my lawyer and he said it was fine. Seems like it would be hard to get prosecuted for it. Imagine if a store was legally obligated to buy back anything they sold to you? Or your counterparty in a stock trade? Not that these are securities, no sir.

Kind of strange how it’s so illegal but there are literally no examples of anyone getting prosecuted for it, right? I’m trembling thinking of the consequences.

Funnily enough you can get prosecuted for things like faking volume (and people have been). But given how many rug pulls you see every day, you’d think it should be easy to find news articles about it right? No dice.

Anyway, I asked ChatGPT Deep Research to take a look, here’s what it told me.

The ethical implications of this launch

There are a lot of different ethical frameworks you can apply to any topic. This one seems particularly grey.

Arguments against this

Violation of expectations

People see an LP created from a funding source associated with the project and assume it’s legit. Not fair to remove those funds.

Disclosure doesn’t make actions ok

Would it be OK to tell someone you’re going to punch them in the face, and then follow through?

Others

I’m open to listing more arguments here but I honestly have a hard time empathizing enough to think of them. If you want me to add something reach out and I’ll update the post.

My thoughts

Fuck those assholes. They wanted to buy a huge amount of supply cheaply and then dump it on everyone else. They got dumped on instead. Turnabout is fair play.

Not sure what the exact term for my position is, perhaps crypto anarchist? I think code should be law (I understand it’s not). I also understand that this is an extremist position though.

A more nuanced argument would focus on the fact that they were warned not to do this and then persisted. People need to be able to take precautions to safeguard their projects from malfeasance. If the precautions have no teeth, then there’s no disincentive to refrain from doing harmful things. They should take responsibility for their own actions.

The ethics of botting

I slipped my opinion on my own ethics into a previous post, did you notice it?

Anway that’s just a joke, I do have ethics, but I doubt they align much with most people. I already mentioned before that I’m a ‘code is law’ kind of guy.

I actually think the botters did nothing wrong. Surprise! I still don’t care what happened to them though. The project did nothing wrong either.

This shouldn’t be that unexpected, given how open I am about the fact that I bot things. Botting is an edge, and I like edges. I even told people during the Ket launch that they should use a bot. Level the playing field.

But I’ll be upfront and admit that the things I do are unfair and I am a bad person.

Building a bot for a (hypothetical) future Ket

I obviously had no time to run a bot for Ket. People don’t understand the kind of stress and amount of work that goes into pressing the buttons correctly at launch. If you’re doing things properly you don’t have time to bot your own launches.

I wouldn’t have botted it anyway though. It feels wrong to me, to bot a coin I launch. I went out of my way to make sure the launch was spotless.

But I couldn’t help but think in the days after launching that there were things I could do to bot launches like this in the future. Sniping a launch is easy if you have the CA or even the token name in advance, and the launcher doesn’t take precautions. But launches like the one I did, not so easy.

I’ve used Sigma before, and I’ve at least looked at some of the other TG bots. They’re not really flexible enough to do what I wanted to be able to do. So I decided to build my own.

Interactive sniping suite

I set out to build some tooling that would let you quickly evaluate new LPs, and buy them with a button press. I don’t know how to do TG bots, but I do know Discord bots, so that’s what I went with.

A very minimal tool to evaluate new LPs, and buy if you like it. There are a few extra features that didn’t trigger here, and I have a laundry list of potential improvements.

Works very smoothly. Not automatic obviously, since you want a human in the loop. But you can buy 1-2s after launch if you’re watching like a hawk. If you want to wait for more metrics to come back, takes more time.

I have some great ideas about how to incorporate AI that I’ll never get to!

Testing it out

So obviously I let my friends use my bot (for a fee, suckers). Unfortunately they kept buying random rugs, so I decided to leave it off, except for when we were expecting a launch.

At this point I had a great idea to incorporate a Twitter scraper into the bot to be used whenever a new LP is detected. The idea being that most projects (if they’re legit) will immediately post the CA on Twitter, giving you extra signal about if it’s OK to click buttons.

Unfortunately Elon is an asshole who made this extremely painful. I ended up registering a bunch of bot accounts and cycling through them to fetch tweets, and it was still very flaky.

Other bots

I’m a paranoid asshole, and the best way to prepare is defence in depth. So I actually wrote two other tools. The Discord bot is good for nontechnical people (my friends), but obviously I can run whatever I want myself, which gives me more options.

The first was an automatic Twitter scraper/buyer. The second was a more manual terminal based buyer that used some optimizations to send requests even faster than my Discord bot, with the idea that if the Discord bot failed that I’d still be able to buy something.

I had so many damn terminals and browser windows open during this launch it looked like a spaceship.

Wink launch

I mentioned before that I do have some ethics. Those ethics don’t extend to not trying to bot a launch of a friend. But they do extend to not having any special information to do so.

In this case, I knew that they were planning on doing fake LPs, and that you had to wait for the Twitter post. I knew what the contract looked like, but it didn’t have anything punitive towards people who waited for the CA to be posted.

I guess it’s slightly unfair that I knew that they were really serious about it and not just faking, but that seems like a pretty marginal advantage, given that you should probably have taken their warnings seriously anyway.

As for the people in my Discord? Well, I tried to warn them.

Did they listen? No.

I was assuming that the Twitter integration would give us a huge edge. Did the Twitter integration work? No. I ended up getting rate limited from scraping every time another fake LP launched.

So anyway, 15 minutes later the actual fucking LP finally launched, although rugs continued launching in the background.

The extra ping in there is another embed that I deleted to stop people from buying a rug.

I bought completely blindly based off someone’s random post that it was the right message (which in retrospect, seems crazy, but I didn’t gambol much). Lets see how they did in terms of latency.

Here’s the Discord embed post time.

Here’s the call time.

11 seconds, no idea how he managed it. I didn’t even see the fucking post update with my phone open to the page and a Twitter tab focused on my desktop.

Anyway, I blindly aped with my fastest bot based on that call and then bought the rest of what I had prepared based on other people confirming it.

At the end of the day, people using my bot had mistakenly dumped over 1.5K avax into the wrong LPs.

Rest in piss, brothers. Follow the fucking directions next time.

Editor’s note: this is Part 1 of 2, see part 2 by tactical_retreat here.

My role in WINK

I was only tangentially involved with the launching of WINK of course, but for the purposes of this article I will write as if I was the sole developer in charge of the rollout plan. This is hypothetical and will never be mentioned again.

I DID NOT REVIEW ANY PRESALE APPLICATIONS. Plenty of my friends/community members didn’t get in. I was able to bully Goose into giving my own projects under the Dev and Draw umbrella a presale allocation (nice shameless plug there, am I doing this right?), but that was my only interaction with the selection process.

I proposed launch plans, workshopped them with Goose and other dev friends, then executed them. I’m working with the team on a contractual basis and as far as I’m aware have no token allocation. Maybe it’s not too late? .5% or something? You have my number.

The Problem

So what is our problem?

Thankfully we had a launch discussed on this very substack not long ago that perfectly highlighted our problem! Feel free to leave this subpar post and go watch the master at work discussing the launch of $$AIGG here.

In short, there were tons of buyers in the same exact block the liquidity went live which spiked the price massively. These “snipers” are typically purely profit motivated with no interest in a project in particular and thus are more likely to sell their tokens at any sign of life. At the very least they’re raising the price for the average joes trying to buy on a DEX.

We want as many tokens as possible in the hands of people who are likely to hold them. Snipers are not likely to hold tokens, and are more likely to dump them after actual interested parties have bought in. This is bad. We don’t like it.

What is a sniper?

When people think of snipers, their first image is the type of advanced mempool snipers that AIGG fell victim to. These are people who are buying immediately, likely less than a second after the launch.

However, if our goal is to maximize the amount of tokens that are in normal community members’ hands, we might want to expand this description. But how far do we expand it? Here’s a rough list of different ways people might buy WINK, loosely organized from “most sniper” to “least sniper:”

1. Watch mempool for any $WINK token and buy it

2. Track liquidity funds and have a bot buy any token deployed by team

3. Twitter scanner bot that buys any CA found in tweet

4. Bot tracking all AVAX deployments, has a UI to one-click buy

5. Bot tracking all AVAX deployments, paste CAs into another bot

6. Do snowscan analysis, paste relevant CA into another bot

7. Wait for account to tweet, trade CA on dex

I think the general consensus is that the first 2 options would be traditional snipers. These are the classic bad actors that pretty much every launch will make an attempt to prevent.

After that we get into the territory of what I would consider “soft snipers” or just very sophisticated market participants. They’re still trying to gain an advantage over the majority of buyers, they just don’t have the technical ability of true snipers. This is where things get more grey.

For Wink’s launch in particular, given the insane amount of hype and mindshare it had, we decided that many of these “soft snipers” were likely motivated similarly to traditional snipers. These are probably people from other chains looking to make some quick money on a highly anticipated launch.

With this thesis in mind, we wanted to maximize the amount of normal, strong community members that would be getting in via DEX.

The Plan

With our goals now clear, we needed a plan. Let me start by saying that it’s pretty much impossible to stop bots. Anything a human can do, a bot can do better. If you’re a reader of this blog surely you know this by now. So what can we do to accomplish our goals?

Well I’ll tell you what we can’t do. We can’t just transfer the funds to a wallet and deploy a single, legit $WINK token. This would obviously get eaten up by pretty much any kind of sniper on our list, leaving normal market participants (see: majority of our community) demoralized.

So we need to deploy more than one token. TR pioneered this with $KET before, deploying 10 different tokens along with the actual legit KET token. However, it didn’t do all that much for him–only 1 (I think, not gonna fact check this) person bought any of the decoys.

TR launched all of his decoys with only .1 AVAX in the LP, and it turns out most snipers are smart enough to figure out not to buy those. With so many more eyeballs on our launch, we needed to do better. I modified the KET script a bit to instead launch LPs with full liquidity, wait 4 seconds, then send the remove LP transaction. Our launch LP was set to be 9000 AVAX, so in practice this led to a decoy 9000 AVAX LP $WINK token being launched roughly every 20 seconds.

In addition to this, I thought it would be fun to split our funds into two wallets each with 9000 AVAX to keep people on their toes. The first wallet would be constantly looping decoy LPs. The second wallet would launch a single decoy LP, pull it, then transfer funds to a new wallet and launch the actual token.

With this plan, we’re making it INCREDIBLY hard for any of the snipers to get wind of which contract is a decoy and which is legit. “Soft snipers” or highly sophisticated actors will also find this basically impossible to navigate, paving the way for a large portion of buyers (or at least larger than usual) to be just normal AVAX enjoyooors.

Launch day

Obviously this targeting of snipers of all varieties is a bit of a grey area. We didn’t want to just take unsuspecting people’s funds. It’s also kind of arbitrary which token will be the correct one, so we made sure to tell people ahead of time that we would ONLY be announcing the correct token on our twitter and buying anything else before our announcement would end in trouble.

After seeing Goose’s announcement and knowing our plan, this didn’t seem like a harsh enough warning. I wanted to make ABSOLUTELY CLEAR that people who buy anything without checking our twitter first will get rekt. So I suggested a second announcement:

Ok, well now I’m feeling good. We’ve made it clear that people shouldn’t try to buy before we tweet and that we’ll be doing everything we can to make sure nobody snipes our launch. With that out of the way, we were ready to begin.

After I finished my final pre-launch tests, since we were planning on looping a bunch of deploys anyway, I decided to run just a single manual test of adding/removing LP from the main looping wallet before kicking off the script that does this repeatedly. We’ve now warned people not to snipe with varying levels of intensity, telling them to wait for a tweet. And if you’ll remember from earlier, the absolute most basic thing we could do is launch a single token directly from our main funded account. Surely nobody will fall for this test one right? Right?? RIGHT???

Well at this point you know what happened here. That first token was live for all of 17 seconds before the liquidity was pulled. When I checked the wallet again to start the actual script I thought something had gone wrong. How could the wallet value have 2x’d? Did I accidentally send all the funds to one wallet instead of splitting them into two?

I #CheckedTheChain and saw that somehow the liquidity remove TX returned 18,000 WAVAX to our wallet and shock started setting in. After our multiple warnings and clear messaging that people should wait for confirmation, fellow blockchain enthusiasts dumped NINE THOUSAND AVAX into the very first, most basic possible contract launch we could have done. In 17 seconds.

There was nothing left to do at this point but continue on with our plan. I got a second computer set up with the launch tweet ready to go, ran my scripts, and fired off the tweet right when I saw the main deploy going through.

There were some minor technical issues towards the end of the launch. The airdrop was delayed about 10-15 minutes after launch due to gas spiking immediately.

Also I had accidentally yeeted the team allocation into the $KET multisig. Shout out to the Wyoming local government for getting us those funds back asap.

Editor’s note: This is because we shared the same codebase, not just a completely random accident.

Reactions

Obviously since we had such an extreme launch plan, there was some backlash.

Based on the people who are upset with us it seems like the plan did a good job deterring the kinds of people we were trying to fight (sniper group extractors + French soft snipers).

We very clearly said to just wait for the tweet and buy–anyone who did not do this took a gamble that they could buy the right token early and dump it on your head like they usually do. As TR would say, they bought tokens at the price they were willing to pay. Snipers will complain that we “couldn’t find actual anti-snipe methods so instead did this out of laziness/incompetence.” They’ll make this complaint because they know that they could have bypassed any “actual” anti-snipe like they always do. Meanwhile, most normal people enjoyed this launch strategy and seem to be happy with our approach.

Successful Soft Snipers

The first buy on the actual WINK contract was 16 seconds after liquidity was added, which you’ll notice was about as long as the decoy contracts were live. Interesting that the decoy contracts had 9000 AVAX worth of buys in the same amount of time it took the real contract to have its first buyer. Surely those decoy buys were organic!

As I said earlier, it’s basically impossible to stop snipers in all forms. The best we could do is try to delay them as much as possible and close the gap between them and normal buyers, which I feel pretty confident we did.

The first buyers on the real contract would still be some form of “soft snipers” or very intelligent market participants though. Does anyone know of any very intelligent market participants?

Insider Trading???

I feel compelled to address this because it will certainly come up (and has already) once people do more research into the launch. Yes, some notorious botters (friends) led by the destroyer of projects got decent entries to $WINK.

It would be easy to cry about insider trading here, but if anything the reality is closer to the opposite. Let me give you some more context on the situation and you can make your own decision.

Wink was a massive launch, and I couldn’t afford to give anyone special treatment. It had to go smoothly, and so I used my knowledge of my friends’ setup to inform our release plan. With the hype around this launch, there may even have been others with similar setups.

For starters, you may have noticed an oddly specific bullet point in my list of potential buy methods earlier: “Bot tracking all AVAX deployments, has a UI to one-click buy.” I believe Tac will be publishing (or has already published, this is taking forever to finish) a blog post with more detail later, but this is what he has been running for our friends. Part of the reason for spamming liquid WINK deployments was to clog their feed.

I also knew people would be watching our Twitter carefully, potentially even with bots that auto-buy CAs in Tweets. I was planning on having the CA in a picture in the announcement tweet, but then I saw that Tac’s Twitter bot was displaying images as well so I cut that and used a url shortener to the launch page so that nobody could get the CA from the tweet alone. I know this annoyed people, but I figured there would be lots of other people closely monitoring the Wink page and wanted to slow them down as much as possible.

I had no interaction with any of my friends during the launch (I was too occupied to look anywhere outside my terminal) and gave them no heads up on the launch strategy or timing. I specifically designed parts of the launch plan around sowing confusion for them. These are some of the most sophisticated actors I know (well, Tac is anyway), and so if I could slow them down I was probably slowing lots of others down.

And slow them down we did! I didn’t read the chat until preparing for this article, but it's a gold mine in there.

Eventually the real contract launched with the tweet at the exact same time, and someone in the server resolved the URL fast enough to send it to the rest of them. It probably also helped that the tweet timing was so exact with the liquidity going live, but it had to be that way so there’s nothing we could do about that.

My own girlfriend, who was in the room with me at time of the launch, bought a decoy USING MY OWN MONEY THAT I GAVE HER TO BUY WITH FOR ME. I will be joining the class action lawsuit against Goose.

These are exactly the type of people we were trying to slow down (sorry friends) and despite some of them getting in early anyway, I feel happy with our efforts. Our goal was to give “normal” market participants as fair a shake as possible and I think we did that. If you can point me to another hyped token launch where the first buy didn’t come through until 16 seconds after launch, I’d love to see it.

Editor’s note: Hello? Ket launch? This is libel, I will be contacting the Canadian Government.

Closing Thoughts

Overall I feel pretty good about our launch strategy. I think the people that are upset just got caught with their hand in the cookie jar trying to outsmart us and are mad about it. They didn’t think we’d go to the lengths we did to stop them, took a gamble trying to make money despite our warnings, and are trying to recoup unexpected losses. As I mentioned above, many of my friends bought decoy tokens and are (mostly) accepting that they took a risk and it didn’t pan out.

Obviously the launch was not flawless. The script had a 4 second pause before removing liquidity but with transaction transit times each token was live for about 15 seconds which is probably too long. I’d also prefer to combat snipers in a way that leaves no room for moral ambiguity, but there really just isn’t one. I’ve seen some truly horrible analogies calling this theft and I look forward to seeing how they hold up in court.

I’ll leave you with some wise words of advice from one of our aforementioned sophisticated market actors:

Delaunch launch

A while back, Delaunch ‘launched’, I guess. Their first token was AI 9000, launched via their platform by a request to their automated twitter bot.

I was not impressed at the quality of their product at launch, nor at the way obvious insiders bought a substantial amount and immediately dumped for huge profits.

I half-assed that investigation though. Here’s a picture of the first block that anyone purchased AI 9000 in.

Notice how all the transactions are going to the same address? That’s the Pharaoh router.

Notice how there are two successes and then 10 failures? Well, they’re all transactions to buy the same token, AI9000. All the ones other than the first two failed due to slippage (oops).

Other than the first successful transaction, the others all used the exact same gas. Rookie move! Almost as rookie as setting slippage on your snipe.

Hilariously, they failed A BUNCH MORE TIMES in the next block.

No worries though, they managed to reboot their shit and buy some tokens something like 20 seconds later.

OK, so it was sniped

Big deal! How can you prove it was the team?

Glad you asked. Turns out they’re fucking stupid. But this will take some explanation.

I didn’t actually write this article to talk about the snipe of AI 9000, but it came up while I was tracing accounts, so it’s interesting background.

Peak retardation

See, when I made this post, I thought they were just making shit up because they weren’t ready to launch and they needed to delay.

Then I saw this post.

I looked at the recent deployments to see what was going on.

There wasn’t a single ‘test’ contract that was bought by anyone… but there were a series of rugs.

Haha, there’s no way that they could have launched a series of rugs in prod right. That’s an insane thing to do. They must be lying.

I was honestly really convinced of this. So convinced that I argued with someone in the comments about it.

Turns out they did admit to committing crimes and then continued committing them! Oops! Sorry for calling you a retard buddy. Turns out I’m the retard for assuming that these guys were not retards. I unmuted you.

The evidence that these contracts are linked

This is not good evidence, which is why I refused to believe it at first. But you can view ‘similar contracts’ on Snowscan.

Now, my argument was that it’s not a crime to have a similar contract to someone. I have, many times in the past, deployed a target contract myself in order to test my bots against it.

Which is why I dismissed this evidence originally.

Still, where’s the proof

Getting to that!

Fast forward to today, AIGG is launching (sorry, $AIGG, they fucked up the symbol and included a $, maybe they’re $WIF fans).

You might ask yourself ‘what kind of anti-snipe protection are they putting in’. It turns out that the ‘anti snipe’ was ‘have the team fill the bonding curve using bots’.

I know this because you literally could not have made it more obvious. I found this by decompiling the contract, but apparently they finally got around to verifying it 9 hours later, so here’s a picture.

See that ‘bytes signature’ at the end there? This is typically used when someone is attesting, off-chain, that someone is allowed to perform an action.

In this case, I’m guessing that they added a web service that the front end contacts to fetch the signature to allow a person to participate in the bonding curve. This makes it (marginally) harder for a bot to participate.

Here are a chunk of the first bonding curve buys.

Notice how they’re all in the same block? Literally impossible to do without being an insider. So this is proof that the ‘anti sniper’ mechanism that they put in was just buying their own damn launch off the bonding curve, as stupid as that sounds.

Who cares?

Well, that’s fair. I looked through a lot of these wallets for dumpage, and I didn’t find any. The only actions they took were to obfuscate tracks by splitting and transfering funds repeatedly, the same pattern for every wallet. Can’t have the distribution looking botted!

But no, these wallets didn’t dump AIGG on you. We’ll get to that in a bit.

I did however, notice that some of the wallets had other, non AIGG activity in them. Some tests on Ket, and some other tx on AI 9000. I grabbed all the addresses from the first page and pumped them through a script to find out what they were doing with AI 9000.

A tale as old as time, someone manually increasing the holder count

Where did the tokens come from though.

And where did those tokens come from?

Well well well, block 54217281 we meet again. If you scroll ALLLLLLL the way to the top of this pointless fucking post, you can refresh your memory. This account was the lucky bot that was the second successful buyer of AI 9000.

Linkage confirmed

I guess they were too lazy to make new bot accounts? The team sniped the AIGG bonding curve with at least some of the accounts they used to snipe their token launch. GG to them.

I’m not getting paid for this so I’m not going to track down how much of their token they dumped on you.

OK so what happened to AIGG

It turns out that dark, malevolent forces WERE waiting for the AIGG (sorry $AIGG) launch to happen.

If you peek at the transactions, you can see that someone with a validator sniped the launch with like 2K-3K AVAX. You can tell because it was in the same block, with the exact same gas as the ‘CompleteCurve’ call.

Those two poor bastard TX below that failed also had the same gas, they just got unfortunately shuffled ahead of the TX they were targeting and could not execute.

Anyway, these snipers were the ones that drove the price up to an insane market cap instantly.

That’s not all

Let’s look at the top traders. Every row here without a ‘bought’ is either someone who bought off the bonding curve, or someone who got transferred tokens some other way without buying them.

Digging into those traders, 6 of them are 100+ days old. One is weird, the other 5 all bought from the bonding curve at relatively normal times and got about 1M tokens for 1 AVAX. Possibly legitimate users. I didn't verify no team connections, running short on time here.

What about the other three?

• Buyer 1 became active 4 days ago… it’s funded by an account that bought 50 avax of AI 9000 immediately at launch.

• Buyer 2 became active 54 days ago… and its first buy was AI 9000, at launch.

• Buyer 3 became active 7 days ago… it bought some AI 9000 later today.

• Did I mention that Buyer 2 and Buyer 3 are linked by the same funder?

I’m struggling to pick up any patterns here, who knows what’s going on. Could be a mystery lost for all time.

I didn’t dig into the other ‘real’ bonding curve buyers but I bet you can find some fun tidbits in there with slightly less PNL.

Why this is incredibly stupid

OK let’s put aside the fact that they’re clearly sniping their own launches and dumping on you. And the fact that they launched a bunch of rugs in prod without warning anyone.

Can we stop and talk for a second about how this was the dumbest possible thing anyone could do? They wanted to prevent bots, so they… sniped the launch themselves? And then put it live at like, 80K MC?

How was this ever going to work? What did they think was going to happen? I’m guessing they didn’t care. Maybe attracting launch bots was intentional; I wonder how many of those sniper bots got instantly dumped on by the team dropping their bonding curve allocation.

In summary, I hate you all and I regret wasting my time writing this.

Not even going to edit. Going to just find some ugly picture and hit send. Oh, here’s one.

Why this is incredibly stupid, part 2

Can someone explain to me why the fuck a MENTALLY HANDICAPPED AI GAMING REPLY BOT NEEDS A TOKEN.

Bye.

My (lack of) ethics

I could write a whole post on my thoughts on ethics, but TLDR I think most people (myself included) are bad, the majority of the rest are neutral, and only a very small percentage are good.

So I don’t have a lot of expectations for you, and I don’t really have any moral authority myself. But here are my thoughts.

Taking profit, generally

I hold a very cynical view on this space. Although the tech is (hypothetically) real, the vast majority of crypto is scams and value extraction. Most of us are here to gamble, to try to make money. Most will lose; the game is generally zero sum here, and 80% of the wealth will transfer to 20% of the participants.

It irritates me when I see people hypocritically calling someone out for ‘dumping’. The people who complain the loudest are also the first to boast when they get a huge win. Your 20x came at the expense of other people, it didn’t come from nowhere.

When you sell you’re dumping, period. Don’t give me any bullshit about “selling into an uptrend”. It’s an AMM. A sell is a sell. A FSH has the same impact as ten 10% sells, except you’re not dragging it out.

It might be ironic, given that I helped build the shameboard, but I don’t begrudge people taking profits. You won’t see me calling people out on the TL for selling.

Almost all of the time, you have to put money at risk to make a profit (exceptions like sniping a fair launch or a free KOL allocation aside). That applies to everyone who bought this token.

Taking profit on $ket as a presale participant

It even applies to people who bought the presale. Presales have a bad reputation for going to zero. Accordingly, there were people who turned down the presale, or didn’t take the full amount, or sold their allocation to other people.

OK, it probably wasn’t the case this time. If there ever was an opportunity to make a guaranteed profit on a presale on Avax this is the one. But it wasn’t 100% risk-free.

So in my mind, you took the risk, you get the reward. If you want to FSH your presale for a 3x-10x, that’s up to you. The people you’re dumping on knew this was a presale.

I do think it’s kind of embarrassing to instantly dump a new, interesting project in the first few days though, when you’re up big and have basically no risk. Particularly if you’re a well-known community member, or Avax ambassador, or ex-Avalabs.

Everyone talks about how it’s sad that nothing can pump, and the community sucks, well you sure showed everyone that’s true.

Coping about Ketboard

Everyone knew the rules of the shameboard. All you had to do to not get exposed was sell ONLY 49% of your risk-free (at this point), hugely up portion of your allocation. Take 10K of profit instead of 20K of profit. Hold the rest for now, see if it goes up. I guarantee no one will care 30 days from now.

If you couldn’t do that, and you absolutely have to have the full amount NOW, then deal with it. Quit being such a whiny bitch. Yes, people are going to call you out, especially if you bullposted while dumping. People will be angry with you, unfairly and hypocritically in my opinion.

If you couldn’t see that coming, I honestly have no idea why. It started literally two hours after launch. You can’t claim ignorance, except that maybe you’re an absolute dumbass, which is probably accurate.

Dumb things people said

A few people complained they got outed for “adding LP”. One of those people who complained added LP single sided to TJ LB, and withdrew Avax an hour later.

It’s actually kind of hilarious how many people publicly posted thanks or explanations for why they ABSOLUTELY HAD TO FSH.

If you’re one of the people who said “I bought on other wallets”, please shut the fuck up, no one believes you.

And then we have this masterpiece of bullshit.

It’s all so tiresome. No one forced you to sell 50+%. I don’t care if the price went too high for Avax. Don’t tell me you had to FSH the remainder of your stack because people were being mean to you.

To be clear, I don’t approve of how aggressively people are sell shaming. Everybody sells. I will sell, either way too early, or way too late.

But if you’re out there whining on Twitter about being attacked after you made bag in two days for zero work and low risk, please quit being such a basic bitch about it.

Man up, own your decisions and the consequences that come with them.

-Fin

For now.

The original dashboard

I’m allergic to profits, so I don’t do much on Solana. But I was vaguely aware of some kind of AI project with a presale that those Bankless twats dumped for millions of dollars, mostly because of the people dunking on them.

This dashboard was created to help show who dumped their tiny investment for an enormous amount of profits, and generated a lot of press on Twitter.

Idea for this project

The people voted for it, and Nobs paid Deep Water Studios to build it. We shipped an MVP in a few days and launched it just after the token went live.

The basic requirement was that people should be hidden until they sold at least 50%. We built everything else at our discretion.

It was empty at first, but almost immediately people sold for 50% profit on their presale and showed up on the board.

#Ad

Want a shameboard for your project? Reach out, our rates are very affordable!

Other features

Before we continue on, let me highlight a few other secondary features we shoehorned into the app.

We had a ton of art available but no central location for users to find it (other than searching Twitter). So we added a gallery.

After the bot started trading, we added a way to monitor the positions and P&L.

These features are more specific to this project, not part of the typical shameboard. But it’s nice to have everything in one place.

Origin of ‘shamefi’

I was threatened by zkPrometheus unless I confirmed that he is the originator of the term ‘ShameFi’.

Apparently he is not the first user of the word, but he is the first person (as far as I can tell) to apply it to this kind of dashboard.

Look who we got a shoutout from!

Stats for days 1, 3, and 7 later

ShameFi didn’t stop everyone from dumping. Here are the stats for the people who were unhidden because they sold more than the 50% limit.

• Day 2

  • 58 forcibly unmasked

• Day 4

  • 88 forcibly unmasked

  • 11 voluntarily unmasked

• Day 8 (data not in yet):

  • 108 forcibly unmasked

  • 27 voluntarily unmasked

  • 2 Dokyo masked

A fair number of people also sold under the 50% limit but I didn’t bother to track that.

Public reception

As you would expect, public reception was mixed.

People not involved generally seem to think this is hilarious.

Some people think it’s a bad idea.

More people think it’s a bad idea. I wonder what the pattern is.

People on the board but masked and still holding strong are having a great time.

Lawsuit

Sadly, things took a tragic turn when a user ranked highly on the shameboard took it a bit too seriously.

A very sad turn of events.

Fortunately, Yellow Cat DAO itself has no legal liability for this decision, it’s entirely on Deep Water Studios. And we have a number of lawyer friends, who helped us draft a response to these allegations, which you can read in full here.

Microtransactions

At Deep Water Studios, we pride ourselves on our responsiveness to customer requests and are enthusiastic about making money.

Some users suggest they be able to add ‘flair’ to themselves (or others) and we quickly provided that ability.

For some reason, the ability to mask yourself for a mere 50 avax was seen as inappropriate. I’m not sure why, that seems like a fair price for a banger mask.

Up next

The ethics of dumping.

My involvement in this project

I try to be as impartial as I can, but everything I write has some bias. Full disclosure, I helped to launch the token and build some tooling, but I have little to no impact on the future of this project at this point, my job is done. I’m just a holder now.

The idea

I think Nobs started this as a joke, originally. But he actually worked on it enough to make it into a real thing.

I didn’t read any of the details behind the project before aping into the presale, as is tradition. This is crime szn, after all.

But the rough idea, as I understood it, is to crowdfund a treasury via a token launch and then use an AI-driven approach to trading that includes market sentiment, chart data, and copy trading smart wallets.

Nobs even has the bot actually clicking the buttons on the UI, which is impressive, if stupid. My AI trading bot will obviously be much better than that and do the trading programmatically (I kid).

Presale

It was pretty easy to get into the presale. Basically, if you were a person with literally any AVAX history and you clicked like on the post, you had a good shot at getting in. Some of the shittiest people I know (including people who have had beef with Nobs in the past) got in, because technically they are community members (bad ones).

Obviously some influenza also got in, that’s how the game is played I guess.

Tokenomics

It ended up being 350K raised from 350 individuals, and 175K into the liquidity pool. Raise was done partially in AVAX and partially in USDC, and unfortunately AVAX took a deep shit halfway through the raise.

There’s no actual identified use of the 10% for ‘development’, but I promise it will be FSH directly into the LP once I can trick Nobs into signing the proper Gnosis transaction.

Art

Nobs paid a bunch of artists to create custom art for the project. I’m no expert, but I think the art is good.

The character and composition mimics the approach taken by the $fwog token. I was first introduced to the so called ‘art token meta’ by a research doc that 0xChefGoose wrote on the topic:

I think it’s a pretty cool idea. Good art leads to good memes on Twitter, which leads to more sharing.

Team

The prelaunch ‘team’ was:

• tactical_retreat (me): Pressed the buttons to launch the token, provided some guidance, and built Ketboard (along with VirtualQuery).

• Russell: Helped set up some kind of legal bullshit, waste of time since crime is legal now. Honestly his whole job is obsolete, feel bad for him.

• Nobs: Everything else. Which is a lot of shit that you’re probably not thinking of and I’m too lazy to list here.

None of the three of us were compensated in any way for the token (I got paid for Ketboard), and there’s no token allocation (other than us each having 1 presale slot, which we would have gotten anyway).

I’m sure Nobs is underwater when you consider how much he’s paid for art, for Ketboard, and for legal stuff. He’s kind of a fucking idiot, right? I mean, I am too, but him even more so.

The team now is just Nobs. The fate of our funds rests in his bot dev abilities. May god have mercy on our souls.

Lore

The symbol for the token was chosen via a series of polls on Twitter.

The final name was actually coined by 0xPrimata, who successfully campaigned in the back channel group chats and defeated all other potential names.

The entity for ‘Yellow Cat DAO’ is actually legally incorporated in Wyoming.

Hence the twitter account’s name.

Ketboard

Prior to launch, it was announced that a ‘shame board’ for presale participants would be published.

The idea originates from a dashboard published for AICC, where various influenza made millions of dollars from a $200 investment.

This is a bit different obviously, since the presale was done via Twitter everyone’s wallet was linked directly to their socials.

The rule for the board is that if you dump more than 50% of your allocation, your socials are exposed, along with your PNL.

Launch prep

Obviously the prep work from Nobs was extensive. Gathering and confirming presale data for 350 people is a lot of work. There’s also the planning, actually building the bot, and wrangling the mentally handicapped (me) for the launch.

On my end, I put about 3 days into preparing for the launch. A short list of dev-type things I did, which are common to basically every launch:

• Gather token requirements

• Airdrop prep

• Test Layer Zero bridging

• Verify DexScreener / TokenSniffer won’t have an issue with the token/deployment

• Script the deploy, LP add, airdrop, and other token transfers

• Prepare shenanigans

• Draft a launch timeline and rollout plan

I have absolute paranoia when it comes to doing things in web3. Especially when those things involve moving $175K of other people’s money. It probably didn’t need to take three days, but I did a lot of testing, verification, and pointless worrying.

Launch day

The launch was timed for 11 AM ET, to ensure that most people (fuck Asia) would be awake.

Every launch is concerned with fairness and snipers, and given how much interest there was in this launch, I obviously wanted to give everyone a fair shot. So I made some preparations to punish anyone buying a CA that wasn’t explicitly posted by the main account on Twitter.

Unfortunately, my legal council advised me that rugging snipers might be considered “a crime”. Note that this was prior to crime becoming legal. So I posted some warnings.

Not that “potentially a crime” ever stopped me. I went through with it, I just made it as likely as possible that people wouldn’t lose money doing stupid things.

So anyway, perhaps I over prepared but everything went smoothly (from a technical POV). Only one person tried to snipe the launch, with a very small amount.

Token went up, until people started selling.

Up next

The shamefi arc.

I was too lazy to edit together the Sonic, Fantom, and Avalanche logos together so I asked Grok to do it. Claude and ChatGPT were even worse, if you can believe it.

Foreword

My original note about load testing Sonic got a lot more visibility than I expected. If you’re here from Sonic / Fantom you probably don’t know me, because I’ve spent most of the last two years in Avalanche communities.

I’ll try to be unbiased, even though I know quite a bit more about how Avalanche works. I’m writing this foreword (and most of this article) before actually doing the final tests, no idea how it will turn out. Please don’t assume I cheated if Avalanche looks better when we get to the results.

FYI - I’M JUST SOME GUY

These tests are not official in any way. They don’t test anything standard. There might be bugs. There might be inaccuracies. Please read the details about how the tests were conducted, which are included later on.

This is absolutely not the top performance for any of the chains tested, if testing such a thing is even possible.

I do think that it’s a good directional indication of how the chains perform relatively to each other, using one guy’s arbitrary idea of how that should be done.

Background

I just recently botted the non-public ETH→Sonic USDC bridge for a few friends so they could buy into AnonAI early. While doing some bulk actions, I couldn’t help but notice that transactions zip along pretty fast, so I started reading the docs on the Sonic Labs website.

Personally I think these are a bit too vague and high level to explain what the magic juice is for Sonic. I “read” the Lachesis whitepaper linked from the docs (not smart enough to understand it) but that’s the consensus used in Fantom already, so not something new. I tried asking on Discord for pointers to more information but no response.

So I figured, why not just test both and see how they do relative to each other?

First test

Seemed like a simple idea. I have all kinds of tools from previous projects that should make this quick. I hacked up a load test and kicked it off on 100 different machines.

Just an offhand thing, so I didn’t do the best job of explaining what I did. And honestly I didn’t collect enough details for a really good load test. So I decided to put a bit more work into it.

Huge mistake.

You need a PHD in load testing

I’m exaggerating, but god damn this was a lot harder than I expected it to be. Partially because I wanted to be a bit efficient, I guess. It’s a lot easier if you just spin up 1,000 machines to do a test rather then if you try to make it work with only 100. Let’s talk about some of the problems I ran into.

Rate limiting?

When I did my initial tests, I was pleased to see that the RPCs were super performant and that there were really basically no rate limits. That made testing pretty easy.

Well, I have no inside information, but I wouldn’t be surprised if they made some changes in the last 24H because some asshole kept sending 10K transactions over the span of 2 minutes.

Or maybe not. Could be the phase of the moon, maybe the RPCs are drunk, maybe I hallucinated or ran insufficient tests. Anyway, it went significantly worse on my retests, and I had to make a lot of adjustments, including increasing the bot count from 100 to 200. Probably would have been better with 300-500 but I’m cheap.

Nonces!

When you send a transaction, you need to include a value which is the number of transactions that you’ve sent previously. So if you send 10 transactions from a fresh account, you send 0, 1, 2… 9 with them.

I have some code to handle nonces efficiently, but apparently it doesn’t deal well with trying to send a massive number of transactions at high speed with potentially flaky RPC endpoints. I ended up writing some custom stuff for this experiment.

Still, I bumped into issues. For example, in one test, a bot only got two transactions onto Fantom. After investigating, it seems like the RPC node accepted the TX and then just… lost it? Or whatever, it never made it on chain.

Then it kept sending the next 200 transactions with a nonce above that, and since transactions are guaranteed to be included in order, they just timed out.

Other weirdness and mistakes

Unfortunately I didn’t keep a good log but I ran into all kinds of problems, which I mostly blame on RPC flakiness. It was a frustrating day. Suspicious how much easier that one-off test was, I probably did something wrong.

I made so many dumb mistakes, it’s remarkable. E.g. just now I accidentally re-ran the Fantom load test again after I had already swept the gas out of the accounts. I forgot to fund accounts several times, ran out of gas several times, funded the accounts incorrectly, got rate limited by Google Cloud, you name it.

Test details

You can nitpick any of these if you want, but at least they’re standardized across all the tests, which is (in my opinion) the most important thing. Once again, we’re not trying to measure ‘max performance’, just ‘relative for this test’.

• 200 bots, running on Google Cloud, spread across 6 different regions (against the common wisdom which is ‘run in us-east-1 for best performance’).

• .1s pause between sending each transaction (not including transmission time). This is done to try and minimize RPC flakiness, which was an issue at lower pause times. It puts the absolute max TPS per bot at 10, but 3 is a more realistic number, which makes max effective TPS 600.

• Test runs for one minute. Reduced from two minutes to limit the possibility of rate limiting, to save money, and impact on actual users of the chains.

• Past tests have shown empirically that ‘multicasting’ TX (sending the same TX to multiple RPC nodes) improves time to inclusion. In this case, Sonic only has two public nodes (official and DRPC) so for parity I used the official node and DRPC for all chains.

• The transactions are pre-cached, issued one by one.

• Transactions use a fixed 30K gas limit, EIP 1559 TX with a fixed and different per chain (sufficient that it’s never exceeded) max gas price. Nonces are calculated, not fetched prior to sending.

• Test runs for 1 minute with a fixed start time and end time. The monitor will watch for blocks with TX from the bots in them and record relevant data, and summarize it.

Latency vs throughput

Andre brings up a good point here but I don’t think it matters much for this test. Worth investigating for a really comprehensive attempt at maximizing TPS, but even then I doubt it would matter if you’re willing to throw enough resources at it.

My reasoning is that the latency of the RPCs only really impacts the time to land on chain (which is not a focus of this test). If the RPC responds slowly, or doesn’t gossip fast enough to validators, well, it will still get included eventually.

Over the course of 1-2 minutes that kind of displacement should get lost in the noise; if you can only mine 800 transactions per second, it doesn’t matter if they were broadcast one second ago or five seconds ago.

It would only matter if the public RPC nodes weren’t up to the task of handling the throughput, which is probably worth knowing anyway right? Hard to say the chain can do X TPS when the RPC nodes can’t accept them.

Costs

The compute costs were actually the largest part of running these tests. Gas for 10K minimal transactions is quite low, given the average gas price on these three chains.

Depending on the run I was doing, the compute cost between $5 and $20 to do one of these tests. Could have been optimized lower but whatever. Probably another $5-$30 for gas for each test (Avalanche load tests are expensive).

In total I probably spent about $600 on this, way too much on this for just a random hobby project. I’ll return some of my kid’s Christmas presents, he’s a pain in the ass anyway.

I think that if you did want to try and measure max performance for a chain, you could probably get close by scaling this test up 3x-5x to ensure saturation.

On Avalanche at least, I know that would be pretty expensive as the gas scaling kicks in. I’m not clear on how the scaling works on Sonic/Fantom but I haven’t been able to make it budge with my tests, they have targets but I believe they’re set much higher.

Expectations

My completely unscientific guess is that the results will be Sonic > Avalanche > Fantom for TPS.

One thing to note is that there’s a big difference between ‘transactions accepted’ and ‘transactions included in blocks’. My original test for Sonic showed 850 accepted transactions per second, but they got included in blocks at a much lower speed, and had quite a long tail of smaller TX blocks after the flood had stopped.

The ‘accepted speed’ is really just a measurement of my bots and the RPC nodes themselves, although it does put an upper limit on the TPS that the chain can actually provide.

My expectation is that none of the chains will finish including TX immediately (within 1s-2s) after the flood ceases. It will be interesting to see what the slope of included TX looks like after the end time. I would like to see it remain relatively flat and then dip straight to zero.

Of the three chains, only Avalanche does not have sub-second block times. So I expect to see some seconds with zero TPS on Avalanche.

Quick test results

Before starting I’ll run a with just one bot to see what it looks like (no fancy charts, just bot output).

Fantom

Total bot transactions: 252

Runtime window statistics:
Transactions: 201
TPS: 3.35

Post-runtime window statistics:
Transactions: 51
TPS: 6.38

Time to reach transaction percentiles:
Time to 95%: 66.00 seconds
Time to 100%: 68.00 seconds

Sonic

Total bot transactions: 355

Runtime window statistics:
Transactions: 354
TPS: 5.90

Post-runtime window statistics:
Transactions: 1
TPS: 1.00

Time to reach transaction percentiles:
Time to 95%: 58.00 seconds
Time to 100%: 61.00 seconds

Avalanche

Total bot transactions: 290

Runtime window statistics:
Transactions: 285
TPS: 4.75

Post-runtime window statistics:
Transactions: 5
TPS: 5.00

Time to reach transaction percentiles:
Time to 95%: 59.00 seconds
Time to 99%: 61.00 seconds

Summary

Not a huge difference here, and it can be probably mostly attributed to the speed of the RPCs.

Reviewing the logs for the run, Fantom was the only test to experience RPC flakiness, which probably impacted the total throughput.

Sonic’s RPCs consistently returned faster than Avalanches, ~.16s vs ~.19s, which explains the TPS difference.

Fantom had a surprising and considerable amount of ‘spill over’ after the test. Sonic had a smaller amount than Avalanche, but that’s to be expected given the longer Avalanche block time, I’d say both kept up with the transactions promptly.

Full test results

Fantom

Processed blocks from 100882092 to 100882390

Load Test Summary:
Used native:  0.497300722597066983
Min transactions per bot: 4
Average transactions per bot: 37.90
Max transactions per bot: 271
Total bot transactions: 7580

Runtime window statistics:
Transactions: 6136
TPS: 102.27

Post-runtime window statistics:
Transactions: 1444
TPS: 12.13

Time to reach transaction percentiles:
80%: 59.00 seconds
90%: 81.00 seconds
95%: 125.00 seconds
99%: 166.00 seconds
100%: 179.00 seconds

I think it’s become clear that the Fantom RPCs are quite a bit less good than the Sonic ones. Of the transactions that did land, most of them landed in the test window, but there was a long tail of 120s for the remainder.

Blue line is the blocks per second, you can see it hovers around 1-2 with a few gaps to 3 or 0.

Sonic

Processed blocks from 1364593 to 1365309

Load Test Summary:
Min transactions per bot: 147
Average transactions per bot: 196.69
Max transactions per bot: 395
Total bot transactions: 39337

Runtime window statistics:
Transactions: 24621
TPS: 410.35

Post-runtime window statistics:
Transactions: 14716
TPS: 69.74

Time to reach transaction percentiles:
80%: 88.00 seconds
90%: 118.00 seconds
95%: 157.00 seconds
99%: 206.00 seconds
100%: 271.00 seconds

The Sonic RPCs are pretty good, and the chain handled quite a few more TPS than Fantom.

But the tail for inclusion here is VERY long. Those last few TX between 99% and 100% were lost in the phantom zone for over a minute.

Pretty consistently builds 2-3 blocks per second. Interestingly, it produces fewer but larger blocks at the start, and then smaller but faster blocks at the end. I guess that’s a good thing.

Avalanche

Processed blocks from 54774059 to 54774121

Load Test Summary:
Min transactions per bot: 69
Average transactions per bot: 197.13
Max transactions per bot: 283
Total bot transactions: 39427

Runtime window statistics:
Transactions: 17618
TPS: 293.63

Post-runtime window statistics:
Transactions: 21809
TPS: 42.51

Time to reach transaction percentiles:
80%: 338.00 seconds
90%: 410.00 seconds
95%: 455.00 seconds
99%: 526.00 seconds
100%: 573.00 seconds

Uh oh! I fucked up.

Block 54776336: 0 bot tx, 10 total tx
Block 54776337: 0 bot tx, 7 total tx
Block 54776338: 689 bot tx, 690 total tx
Block 54776339: 678 bot tx, 682 total tx
Block 54776340: 0 bot tx, 7 total tx
Block 54776341: 0 bot tx, 1 total tx
Block 54776342: 0 bot tx, 7 total tx

Looks like I did NOT set the max gas properly for this load test. The blocks with 0 TX before suddenly getting a huge chunk are when the minimum fee for the block exceeded my TX fees.

I see a suspicious number of blocks with ~700 transactions, taking a look at one:

Suddenly it makes more sense. The block is literally full of transactions. Some quick math explains why; 15M (max gas per block) / 21.5K (gas used by my TX) = 697.

I’ve heard people say that Avalanche has hit 850 TPS, which looking at the raw math, now seems a bit suspicious. I’m not sure it can produce multiple blocks per second at the moment. Here’s an example of the first few blocks, including their timestamp.

Block 54776239 at 1734969921: 660 bot tx, 669 total tx
Block 54776240 at 1734969922: 610 bot tx, 620 total tx
Block 54776241 at 1734969923: 651 bot tx, 667 total tx
Block 54776242 at 1734969927: 699 bot tx, 701 total tx
Block 54776243 at 1734969928: 667 bot tx, 673 total tx
Block 54776244 at 1734969929: 451 bot tx, 461 total tx
Block 54776245 at 1734969930: 669 bot tx, 678 total tx
Block 54776246 at 1734969932: 699 bot tx, 701 total tx

I’ve highlighted a few gaps with 1 or more seconds in between blocks. But there are plenty of examples of it hitting exactly one block per second.

Block 54776250 at 1734969940: 696 bot tx, 698 total tx
Block 54776251 at 1734969941: 675 bot tx, 678 total tx
Block 54776252 at 1734969942: 573 bot tx, 582 total tx
Block 54776253 at 1734969943: 393 bot tx, 395 total tx
Block 54776254 at 1734969944: 491 bot tx, 497 total tx
Block 54776255 at 1734969945: 481 bot tx, 484 total tx

So why the gap? I speculate that some of it is caused by under provisioned validators; I know Avalanche saw reduced liveliness during the inscriptions craze because some validators could not keep up with block production. I imagine there’s a visualization somewhere that shows the historical slot misses by validators but I’m not sure where that is.

No chart for Avalanche alone, looks kind of ugly, see below for combined charts.

All combined on a chart

Since I goofed on Avalanche I could only pick the first 30 seconds of data to keep things in parity between the three charts.

Missing timestamps for Avalanche make it ugly, so here’s a 3s moving average version.

Finally, cumulative transaction counts over the first 30s.

Summary

It seems like Sonic is a huge improvement over Fantom, but I think some of the distinction can be attributed to the RPC nodes (eating my words from earlier right now). It does consistently produce more blocks per second, which is a good improvement regardless of throughput.

Avalanche consistently produces larger and fuller blocks (up to the gas limit). Gas price scales up relatively quickly compared to the other chains. A surprising number of timestamps with zero blocks produced. It seems to do better at producing blocks on time (one per second) when the transactions per second included are lower.

The gas limit on Sonic seems insanely high, and the gas target is also pretty extreme, I didn’t experience any gas price increases over the baseline. No zero-block timestamps, a max of four blocks per second.

Thoughts

Note to self: stop doing expensive hobby research.

Location

One interesting outcome is that bots that launched in non-us locations had a distinctly worse time getting transactions included than us-based bots. Those poor Australia bots! But even Europe had a worse time than the US. The tyranny of the speed of light in action.

“Actual load testing”

I’ve said this like 10 times in this post but I’m not really sure what an actual load test even means. But if I wanted to attempt it, I’d probably use the same code but:

1. Only run in us-east1.

2. Use a validator for each chain in addition to the primary RPC and DRPC. Maybe also use QuickNode (paid node service), I’ve had good results with them.

3. Launch more bots, like a lot more. Maybe 800.

4. Be safer in sending TX from each bot, probably shoot for 2/s since none of these chains can currently handle 1600 TPS.

5. Set higher max gas for Avalanche, a lot higher, and fund the accounts more.

6. Just run for 30s, I don’t think 60s is meaningfully more useful.