The thing nobody tells you when you first start watching listings is that the interesting move often happens before the announcement, and it happens somewhere boring. Not a whale accumulating on a DEX, not a founder wallet waking up. It is a plain token transfer from a project treasury or a foundation multisig into a wallet that belongs to a market maker. That transfer has to happen because trading cannot open on day one without inventory. Somebody has to be standing on both sides of the book, and that somebody needs coins in hand before the first buyer shows up. So the coins move early, and if you know which wallets to watch, you sometimes get a heads up that a listing is coming.
I want to be honest up front that this is a noisy signal and a legally touchy one, and most of the value is in understanding the mechanism rather than in trying to front-run anything. But the mechanism is real, and once you see it you cannot unsee it.
Why the coins move before the announcement
A market maker running spot inventory for a new listing needs two things ready before the pair goes live: a float of the token itself and a matching stack of quote currency, usually a stablecoin or the exchange's base asset. The token side is the tell. Projects rarely have the coins sitting on the market maker's books already, so there is a physical transfer from the project side into the MM's operational wallet, and it tends to land in a window that runs from roughly a few days to a couple of weeks before trading opens. Deals get signed, inventory gets seeded, and then the exchange flips the switch.
Some arrangements are loans and some are outright buys or a working-capital model where the MM holds inventory against an option, but the on-chain footprint looks similar either way. A treasury or foundation address sends a meaningful, round-ish quantity of a token to an address that has historically received similar inventory drops before other listings. That last clause is the whole game. A single transfer tells you nothing. A transfer into a wallet that has a track record tells you something.
How to actually identify a market-maker wallet on-chain
You do not get a label handed to you. You build the identification yourself from behavior, and it takes patience. The pattern that separates an MM operational wallet from a random large holder looks like this:
- It receives inventory in tokens it has no long-term relationship with. An MM wallet holds a rotating cast of assets, not one project's bag. A treasury holds mostly its own token.
- It deposits into and withdraws from centralized-exchange hot wallets constantly, in both directions, across many different tokens. This two-way exchange flow is the strongest single tell.
- Its balances turn over. Inventory comes in, gets deployed, comes back, leaves again. A wallet that only accumulates is not making markets.
- It touches multiple exchanges' deposit addresses, because most serious MMs quote on more than one venue.
- It sometimes clusters with sibling wallets that behave identically, which is how these desks segregate books.
Once you have a candidate, you confirm it by looking backward. Pull its receive history and check whether earlier inventory drops line up with pairs that later went live on an exchange. If three or four past transfers into this wallet each preceded a listing of that exact token within a couple of weeks, you have found something worth watching. If the timing is random, you found a large trader, not a market maker, and you should drop it.
What a real pre-listing pattern looks like
The clean version reads like this. A foundation multisig for a token that has never traded on a major venue sends a round-numbered chunk of supply to a wallet you have already tagged as an MM. Shortly after, that wallet or a sibling starts pushing smaller tranches into a specific exchange's deposit addresses. Sometimes you also see the quote-side inventory getting staged, stablecoins flowing into the same operational cluster. Then, days later, the pair opens.
The messy version, which is most of them, looks almost identical until it does not. The transfer happens and then nothing. Or the coins go to an MM but for an OTC deal or a treasury-management mandate that has nothing to do with a listing. Or the listing was already public knowledge and you are looking at confirmation, not a leak. The false-positive rate here is high enough that I would not put a number on it beyond saying most tagged inventory transfers do not resolve into a near-term listing you could have traded. You are hunting for the subset that does, and you are wrong more often than you are right on any single event.
The legal and practical limits, which are the actual point
Start with the law, because it is not optional. Trading on material non-public information about a listing can cross into insider-trading and exchange-policy territory depending on how you got it and what the asset is. Reading a public blockchain is not the same as leaking a listing from inside an exchange, and on-chain data is public by definition. But if your inference is downstream of someone's confidential deal, or if you are trading a token that a regulator treats as a security, you are on thin ice and should get real advice rather than mine. I am describing a research method, not handing you a green light.
Now the practical limits, which quietly kill most of the edge before the law ever comes up. Timing is the killer. Even when you correctly spot inventory seeding, you do not know if the listing is in three days or three weeks, and holding a thin, pre-listing token that long carries plenty of downside. The market frequently prices in the listing before you can act, especially on tokens that already trade elsewhere. Spreads on illiquid names eat you. And the biggest, most liquid listings are exactly the ones where the flow is either private, done off-chain, or so well watched that any edge is gone by the time you see it.
So the honest use of this is not a front-running strategy. It is a monitoring discipline. I keep a watchlist of tagged MM wallets and set alerts on inbound transfers of tokens those wallets do not normally hold, then I treat any hit as a question to investigate rather than a trade to place. Most hits go nowhere. A few give me a genuine early read on where liquidity is about to appear, which is useful even when I never touch the token, because it tells me where attention is heading. On Blockcircle I wire these MM wallets into the same tracking I use for whale flow, so a new inventory transfer surfaces next to everything else instead of living in a spreadsheet I forget to check.
If you take one thing from this, make it the tagging work. The wallets are the asset here, not any single transfer. Build the list slowly, confirm every candidate against its own history, and accept that the signal is a soft lead you verify by hand, never a trigger you fire on blind.