Bundled launches and sniper wallets: what they are and how to spot them

A bundled launch is one where the deployer buys a large share of supply across many wallets in the same block the token goes live, so the holder list looks distributed while one person controls the float. A sniper is a bot doing the same thing moments later. Both leave one signature: wallets with no history, one funder, one timestamp.

The distinction matters because the first is the creator lying about distribution and the second is a bot beating you to the trade. Neither is illegal, both are common, and only one of them can be designed against.

What bundling actually is

Deploying a token and buying its supply are separate transactions. If they land in separate blocks, anyone watching can buy in between, and the deployer competes for their own supply at rising prices.

Bundling removes that gap. The deployer submits the pool creation and a set of buy transactions as one atomic unit that executes together, in order, in a single block. Because nothing can be inserted between them, the buys execute at the earliest and cheapest point on the curve. Split across twenty or fifty freshly created wallets, the result is a holder list with no single dominant address and a top-ten percentage that looks unremarkable.

On Solana this is done through bundles submitted to a block engine — an all-or-nothing set of transactions with a tip attached, which either lands complete or not at all. That primitive exists for legitimate reasons, chiefly atomic arbitrage and complex multi-step transactions, and bundled launches are an application of it rather than an exploit of it. Related mechanics are covered in what a priority fee on Solana is.

On EVM chains the equivalent is a private bundle sent to a block builder, or — on an L2 — simply having the transactions ordered by the sequencer. Robinhood Chain is instructive here: L2BEAT's assessment notes it runs a centralised sequencer with MEV capability, and blocks are roughly 250ms. First-block ordering on a chain like that is not something a retail buyer is competing for on equal terms, and it is worth being clear-eyed about that rather than assuming a level start.

The economics are the point. A bundler who takes 30% of supply at the bottom of the curve has an average entry an order of magnitude below the price at which retail arrives an hour later. They do not need the token to succeed. They need enough volume to sell into.

Snipers are a different actor with the same footprint

A sniper is not the deployer. It is a bot watching for new pools and buying the instant one appears.

There is no mempool on Solana, so sniping there means subscribing to the launchpad program's transaction stream and reacting within the same or the next slot. On EVM chains a sniper watches the mempool or the sequencer feed for the pool-creation call. Either way the bot competes on infrastructure — colocation, priority fees, tips — and a human clicking a button is not in that race. Sniping, explained covers why.

For a buyer the practical difference is smaller than the moral one: both sit above you in the entry-price stack and both will be sold into your bid. Distinguishing them matters because bundling tells you something about the creator, and sniping does not.

What it looks like on-chain

The signature is a combination, and no single element is conclusive on its own.

Timestamp. A set of wallets whose first-ever transaction is in the same slot or block as the pool's creation, or within a slot or two of it. Open two or three of them and compare. This is the strongest single element.

Funding. Those wallets received their gas and buy capital from one address, usually shortly before the launch, often in identical amounts. Follow the funder backwards: if it was itself funded by a fresh exchange withdrawal an hour earlier, that is a chain of steps taken specifically to have no history.

Uniformity. Near-identical holdings across many addresses. Real buyers arrive at odd sizes; a script does not, unless written to randomise, which better ones are.

Absence. No prior transactions, no other tokens, nothing. A wallet that exists only to hold this token was made to hold this token.

Behaviour after launch. Several sell in the same window. Coordinated exit is the confirmation, and it arrives too late to help the buy decision, but it is worth logging against the deployer address for next time.

The fastest way to see all of this is a cluster view rather than a holder table — reading a Bubblemaps bubble map covers what the connections do and do not prove. Confirm anything the map suggests in the explorer, and add the cluster's percentages together: ten wallets holding 2.5% each, funded by one address in one block, is a single 25% holder.

Also read the raw list properly, because bundling is precisely the technique that defeats the summary figure screeners show. Spreading a position across forty wallets instead of ten lowers the top-ten percentage without changing who controls anything, which is why how to read a token's holder distribution spends most of its length on what to exclude and what to combine.

The venue-level defences

This is where the market has moved, and it is the most concrete thing in this article.

Pools.trade, launched by Uniswap Labs on 5 August 2026, offers a Crowd Launch mode built specifically against bundling. Instead of a pool going live and being immediately buyable, there is a four-hour bidding window priced by a time-weighted average, so being first in the block confers no price advantage — a bid placed in minute one and a bid placed in hour three settle on the same basis. It also enforces a $10,000 minimum fully diluted valuation for the launch to complete, refunding bidders if it is not reached. Its alternative Instant Launch mode is a conventional bonding curve, live immediately and open to the usual first-block dynamics, so which mode a token used is part of what you are checking. The mechanism is set out in what Pools.trade is, with venue-level launch data on the Pools.trade analytics page.

This is not a fringe design. Pools.trade charges no launchpad fee and takes a 0.25% LP fee that autocompounds into permanently locked, creator-unremovable liquidity in a standard Uniswap v4 pool, against roughly 1% elsewhere, and it took approximately 50% of Robinhood Chain launchpad volume and 40% of new tokens within days.

Other venues shape the problem rather than solving it. Pons.family has no bonding curve and no migration: fixed supply, pool live from block one, so there is no migration event to snipe. hood.fun runs a bonding curve then auto-migrates to a locked Uniswap v3 pool. pump.fun migrates the full LP atomically to PumpSwap at its graduation threshold, removing front-running of the migration itself.

A TWAP window is the only one of these that directly attacks the advantage of being first. The rest change what happens after the launch, not who gets in at the bottom.

Is bundling always malicious?

No, and the honest version of this article has to say so.

A creator who buys a meaningful share of their own token at launch and holds it is aligned with buyers, not against them. Some do it for market-making inventory, some to stop a sniper taking the position instead — a real defensive motivation, not a pretext.

What makes it a problem is concealment. Buying 25% of supply into one visible wallet and saying so is a disclosed allocation. Buying 25% across forty wallets so the holder list reads as organic is a claim about distribution that is not true, and the effort spent hiding it is the informative part. The transactions are identical either way, so what you can act on is the concealment, not the intent.

What these checks do not catch

Everything above finds wallets that were clumsy. It does not find the ones that were not.

Aged wallets. A bundler who funds and warms wallets weeks in advance, through an exchange, gives you no common funder and no launch-day first transaction. The cluster view shows nothing because there is nothing to draw.

Off-list supply. Cluster maps and holder tables both cover a top set of addresses. Split across enough wallets and most of the position sits below the rendered cut-off.

Intermediated funding. A single hop through a centralised exchange breaks the visible link between funder and wallets permanently. This is cheap and well known.

Everything about the contract. Bundling analysis says nothing about mint authority, sell taxes, honeypot behaviour or whether liquidity can be pulled. A token with immaculate distribution and a live mint authority is not safe. The full sequence is in how to spot a memecoin rug pull before you buy.

The outcome. Detecting a bundle tells you supply is concentrated. It does not tell you when or whether that supply will be sold, and plenty of un-bundled tokens go to zero on their own. Concentration is a reason to size smaller, not a prediction.

Our own coverage. Launch and venue figures described as Meme Central's own reflect tokens indexed by Meme Central, not the whole market.

Frequently asked questions

What is a bundled memecoin launch?

One where the deployer buys a large share of supply across many wallets in the same block as the token's deployment, using an atomic transaction bundle. The buys execute at the lowest point on the curve and the resulting holder list looks distributed even though one person controls the float.

How can I tell if a token was bundled?

Open several of the top non-contract holders and compare three things: when their first-ever transaction occurred, which address funded them, and whether they hold anything else. Same block, same funder and no other history across a group of wallets is the pattern. Add their percentages together.

Is a sniper wallet the same as a bundler?

No. A bundler is the deployer buying their own launch atomically. A sniper is an unrelated bot buying in the first block or two after deployment. They leave a similar footprint, but bundling tells you something about the creator's honesty and sniping only tells you the launch was watched.

Can bundling be prevented?

Partly, at the venue level. Pools.trade's Crowd Launch uses a four-hour time-weighted bidding window specifically so that being first in the block gives no price advantage, with a $10,000 minimum valuation or funds are refunded. Classic bonding-curve launches remain open to it by design.

---

What a creator can prove, and what they cannot

Distribution is the hardest thing for a launch to demonstrate, because absence of hidden wallets cannot be shown on-chain. The pool is different: Team Finance's LP lock — built by TrustSwap, which also builds Meme Central — secures liquidity for a fixed term on Ethereum, Robinhood Chain, Polygon, Base and BNB, and surfaces as a verified badge on the token's page in the Meme Central launch feed. It is a commitment about the pool only. It does nothing about bundled supply, and a locked pool full of tokens held by forty wallets you funded is still a bundled launch.

---

Nothing here is financial, legal or tax advice. Memecoins are extremely high-risk: most lose most of their value, and the majority of tokens launched never reach a decentralised exchange at all. Never spend money you cannot afford to lose entirely. Meme Central does not recommend any specific token. Data described as Meme Central's own reflects tokens indexed by Meme Central and is not whole-market data.