Why Most New Token Listings Lose You Money in 2026

On the morning of 2 October 2026, Binance pulled eight tokens (Request Network, WINkLink, Gains Network, Theta Fuel, Gnosis, QuarkChain, Lazio Fan Token and Osmosis) out of its margin and loan services at 10:00 UTC, with no explanation attached to the notice. None of them trended. That is how most listings actually end: not with a scandal and a congressional hearing, but with a line item buried in an operational bulletin that almost nobody reads.

The story crypto likes to tell about a listing is the opposite one. A token graduates from a launchpad or an airdrop, an exchange flips the switch, and the chart gaps up on a wall of green. Traders talk about the Coinbase effect and the Binance effect, screenshot the first candle, and treat the listing itself as a stamp of approval, as though a compliance team somewhere did the homework so you would not have to. The mechanics of that process are worth understanding on their own terms, and we have covered how tokens actually get listed and why exchanges really list them in depth elsewhere.

This piece is about the other half of the trade: what happens after the confetti. The uncomfortable truth, supported by the largest datasets anyone has assembled, is that the first green candle is a trap far more often than it is a gift. A new listing is not the end of a token’s risk. It is the beginning of it. And the ways a listing goes wrong are not random or mysterious; they are specific, repeatable, and in most cases visible before you ever hit buy. What follows is a field guide to the seven failure modes that drain retail wallets, each with a real example, and a checklist for spotting them in advance.

The base rate nobody prints on the announcement

Start with the number that frames everything else. In 2026 the trading firm GSR, led on this work by chief legal and strategy officer Josh Riezman and research analyst Slater Santer, compiled more than 2,300 token listings on major exchanges going back to 2013. Crucially, the dataset kept delisted and dead tokens in the sample, so it does not flatter the results by quietly dropping the failures the way survivorship-biased tallies do. The headline finding is blunt: the median listing fell below its own debut price within three days, and was down roughly 50% within 90 days.

It gets worse the higher the valuation. GSR found that launches which debuted with a fully diluted valuation (FDV) above $1 billion returned a median of about $0.19 on the dollar a year later, an 81% loss. Tokens that came to market with less than 20% of their supply actually circulating lost around 75% of their value over the same year, keeping just $0.23 to $0.26 per dollar invested. Projects that listed with 30% to 50% of supply in circulation did roughly twice as well, retaining about $0.55. The single most predictive variable was not the narrative, the backer, or the chain. It was how much of the supply was really free to trade on day one.

None of this means every new token is a scam, and it does not mean you can never make money on a debut. It means the odds are stacked, the stacking is structural, and the house knows exactly how the structure works. The rest of this guide takes that structure apart.

The pop is real, and that is the bait

Before taking the structure apart, it is worth respecting the thing that pulls people in, because the listing-day rally is not a myth. Messari’s analysis of the so-called Coinbase effect found newly listed tokens gained an average of roughly 91% in the five days around a Coinbase listing, with a range that ran from deeply negative to several hundred percent. Binance debuts show a similar first-move burst. That pop is genuine, and it is exactly why it works as bait: it is the reward that persuades a fresh wave of buyers to show up at the single moment the structure is most primed to sell into them.

The trap is in the timing. The same studies that document the pop also document the give-back, as the market makers, airdrop farmers, and short-term flippers who held supply before you take their profit into that first surge. The initial move often reverses most of the way within days to weeks. By the time an average retail buyer has decided the rally is real and clicked in, the move is frequently over, and what looks like a dip to buy is the opening leg of the slow bleed the GSR data describes. The listing effect and the failure modes in this guide are not opposites. They are the same event viewed at two different speeds: the pop draws the capital in, and the structure decides where it ends up.

Failure mode one: the low-float, high-FDV time bomb

The most common way a listing hurts you is also the most mathematical, and it is baked in before the token ever trades. A project mints, say, 10 billion tokens but releases only 3% or 4% of them at launch, locking the rest for the team, early investors, and a foundation treasury on a multi-year vesting schedule. With so little supply free to move, a modest amount of buying sends the price (and therefore the FDV, which prices every locked token at the live market price) soaring. The chart looks euphoric. The problem is that the euphoria is calculated against a supply that does not exist yet.

The GSR data makes the pattern impossible to miss. Median circulating supply at listing collapses as valuation rises: roughly 97% of supply is live for projects under a $10 million FDV, about 28% in the $10 million to $100 million band, around 16% between $500 million and $1 billion, and only about 13% above $1 billion. The richer the headline valuation, the thinner the real float underneath it, which is precisely the combination that produces a violent first candle followed by a long, grinding bleed as locked tokens vest into a market that has already run out of fresh buyers. Binance Research has estimated that something on the order of $155 billion in tokens were scheduled to unlock between 2024 and 2030, a supply wave that has to be absorbed by someone, and that someone is usually the retail buyer who showed up on listing day.

The counter-case is distribution. When a token launches with most of its supply already in community hands, there is less overhang to sell into and less of a cliff to fall off. Jeff Yan, founder of the perp exchange Hyperliquid, put the philosophy bluntly after its no-investor launch, arguing that if a handful of venture funds own half of a network, that will always be a scar on the network itself, and that ownership should be community-driven. You do not have to agree with the ideology to use it as a filter: a tiny float with a huge FDV and a cap table dominated by insiders is the first and loudest warning sign on this list.

Failure mode two: the market maker who dumps on you

Every new listing needs someone to quote both sides of the order book so buyers and sellers can actually trade. That job falls to market makers, and most projects pay them with a structure called a loan-plus-option: the project lends the firm a slice of supply before trading starts, and the firm is compensated with call options struck well above the launch price. Done honestly, the market maker earns its keep by tightening spreads and smoothing volatility. Done dishonestly, the same inventory becomes ammunition to sell into the crowd.

It helps to see the economics. In a typical deal the project lends the market maker 1% to 5% of circulating supply, and the firm is paid in call options struck 25% to 100% above the launch price over a 12 to 24 month term. If the token rips, the firm exercises cheap options and keeps the upside; if it sags, the firm can sell the borrowed inventory and buy it back lower, pocketing the spread. The largest desks, names like Wintermute, GSR, and DWF Labs, run legitimate versions of this at scale, and a healthy market genuinely needs them. The abuse is not the structure itself but the one-sided, profit-share variant that quietly turns the liquidity provider into the single biggest seller the instant trading opens.

The cautionary tale here is Movement’s MOVE token, and it is worth reading the full post-mortem because it hits almost every mode on this list at once. After MOVE debuted on Binance in December 2024, a market maker working through a middleman entity sold roughly 66 million MOVE, about 5% of the public supply, within a day of listing, netting around $38 million. The contract reportedly allowed liquidation once FDV crossed $5 billion and split the proceeds above that trigger. Zaki Manian, the crypto engineer who reviewed the arrangement, said it created “incentives basically to manipulate the price” and to “dump on retail for shared profit.” Binance offboarded the market maker in March 2025 and froze the profit; the token later fell about 99% from its high, and Movement’s parent company filed for Chapter 11 in July 2026.

Regulators and exchanges have started to react. In March 2026 Binance tightened its market-maker rules, requiring issuers to disclose the identity and legal entity of their market-making partners and the terms of the deal, and banning profit-sharing and guaranteed-return arrangements of the kind that powered the MOVE dump. That is progress, but disclosure is not protection: the structure is still legal, still common, and still invisible to a buyer who only sees a clean-looking order book on listing day.

Failure mode three: the volume was never real

When you evaluate a new listing, the single most reassuring number on the page is trading volume. A token doing tens of millions of dollars a day looks liquid, popular, and safe to enter and exit. The trouble is that a large fraction of reported crypto volume is manufactured. Wash trading, the practice of buying and selling to yourself to paint the tape, is cheap to run with bots and hard to see from the outside, and it is most intense exactly where it matters most: on freshly listed, thinly held tokens that need to look alive.

The scale is not a conspiracy theory. Analyses across several years have estimated that anywhere from half to nearly all of the volume on some unregulated venues is fake, and academic work has put the figure above 70% on many of them. The enforcement record is just as telling. In 2024 the FBI ran Operation Token Mirrors, creating a fake token called NexFundAI to catch manipulators in the act; the SEC subsequently charged three market-making firms and nine individuals whose bots had run what prosecutors described as quadrillions of transactions to simulate demand. One of them, CLS Global, pleaded guilty and was fined $428,059. The lesson for a listing-day buyer is simple and counterintuitive: volume is not evidence of liquidity. If a token reports huge turnover but the order book is thin and the on-chain holder count is tiny, the turnover is the warning, not the reassurance.

Spotting it is not hard once you know where to look. Compare reported 24-hour volume against the token’s real liquidity: if a token claims tens of millions of dollars in daily turnover but its deepest order book holds only a few hundred thousand and its on-chain holder count is in the low hundreds, the volume is theater. Watch for turnover concentrated on a single obscure venue, suspiciously round and evenly spaced trade sizes, and a volume-to-market-cap ratio that would be remarkable for a blue chip, let alone a token that is a week old. Real demand leaves a messy, distributed footprint across many venues and wallet sizes; manufactured demand tends to look oddly tidy.

Failure mode four: the unlock cliff

The low-float launch creates the setup; the unlock schedule delivers the punchline. Most tokens vest their locked supply over one to four years, often with a steep “cliff” after six or twelve months when a large tranche becomes sellable all at once. On paper this is responsible tokenomics, aligning insiders with the long term. In practice, a cliff is a pre-announced sell event, and the market knows the date.

What makes cliffs so dangerous for retail is that the pressure arrives precisely when early attention has faded. The listing pop drew in buyers at the top; six months later, with the narrative cooled and the chart already down, the team, the fund, and the treasury can finally sell, and the thin organic demand that remains cannot absorb it. This is the mechanical engine behind the GSR finding that low-float launches keep barely a quarter of their value after a year. It is not that the project necessarily failed to build anything; it is that the float roughly doubled or tripled while the pool of willing buyers shrank. Before you buy any debut, the unlock calendar is not optional reading. A token with 13% of supply live today and a 25% insider tranche cliffing in April is telling you, in writing, when the next wave of selling starts.

Not all vesting is equal, and the shape matters as much as the size. A hard cliff, where a whole tranche unlocks on one date, concentrates the selling pressure into a single brutal window; linear or daily vesting spreads the same supply out and is usually gentler on the chart. Public unlock trackers now map these schedules token by token, so the information is a search away rather than buried in a whitepaper footnote. The questions worth answering before you buy are simple: what percentage of total supply unlocks in the next 90 days, who receives it (team and investors tend to sell, ecosystem incentives tend to get spent into the market either way), and how that incoming supply compares to the token’s average daily volume. When a single unlock dwarfs weeks of real trading, the price has to find new buyers it does not have.

Failure mode five: death by a thousand delistings

Some listings do not blow up. They fade, lose volume, drift onto a watchlist, and are quietly removed, which is the fate of the eight tokens Binance cut from margin on the morning this article published. The scale of the churn is easy to underestimate. A CryptoRank study of roughly 5,000 delisted tokens found that major exchanges have been removing 400 to 500 tokens per quarter in 2026, up from 67 in the first quarter of 2024, with Gate alone accounting for 573 delistings in the first half of 2026. The shortest-lived token in the sample, a project called TST, survived a single day. We walk through the staged mechanics of this in a dedicated guide to how and why exchanges pull a token.

Before the final removal, big exchanges usually flash a warning. Binance attaches a Monitoring Tag (and a Seed Tag for the riskiest new graduates) to signal elevated risk, and in 2025 it introduced a community Vote to Delist, whose first batch removed 14 tokens in April 2025. The vote is advisory; the exchange still decides. And it has not been free of gaming. Binance co-founder He Yi noted that some projects facing removal were buying votes to push rivals off the platform instead, and said a later round would target tokens that no one bothered to defend. The reverse of the listing effect is brutal: a delisting announcement typically knocks 25% to 30% off a token within hours, and the bulk of whatever value remains tends to drain away over the following month. Older Binance removals illustrate the range, with OMG, WAVES, and XEM each falling somewhere between roughly 26% and 30% on the news.

If you are holding when the notice lands, the mechanics decide whether you escape with anything. A delisting is rarely instant. The exchange usually cuts margin and leveraged products first, then halts new spot trading, then stops deposits, and only later, often several weeks on, closes withdrawals for good, sometimes auto-converting any leftover balance into a stablecoin at whatever price prevails on the cutoff day. That staging is less a courtesy than an orderly-wind-down obligation, but it does hand holders a window. The common mistake is treating the first announcement as a cue to panic-sell into a collapsing order book, rather than as a deadline to move the asset to a venue where it still trades, which for many delisted tokens means a decentralized exchange rather than another major platform.

Failure mode six: when the regulator pulls the plug

A token can be perfectly functional and still get delisted because the rulebook changed around it. This failure mode has become one of the defining stories of 2026, and it does not care how good the technology is. In Europe, the Markets in Crypto-Assets regulation (MiCA) requires a compliant issuer behind any stablecoin an exchange lists; when Tether declined to pursue MiCA authorization, EU-regulated venues including Coinbase, Kraken, and Crypto.com removed USDT for European users while keeping the authorized USDC and EURC. Holders were not accused of anything. The asset simply stopped being listable in that jurisdiction.

Privacy coins face the same guillotine on a timer. The EU’s Anti-Money-Laundering Regulation applies from 1 July 2027 and bars licensed providers from offering anonymity-enhancing coins such as Monero, Zcash, and Dash; Kraken already pulled Monero across the European Economic Area, converting balances to Bitcoin. In the United States the trigger is securities law. Coinbase suspended XRP in January 2021 after the SEC sued Ripple, and Robinhood dropped Solana, Cardano, and Polygon in 2023 after the agency named them securities in its exchange suits, only for XRP to be relisted after a favorable court ruling. The climate has softened under the current SEC, whose Project Crypto proposes a token taxonomy and an innovation exemption, a shift we track in our coverage of the regulator’s thin-staffed scramble to write the new rules. But softer is not settled, and a listing that depends on a particular regulatory reading is a listing with a political variable priced into it.

The broader point to internalize is that listing risk now includes jurisdiction risk. Under MiCA, admitting an asset to trading in the EU requires a compliant, published crypto-asset white paper standing behind it; where none exists, the obligation lands on the exchange, and the practical answer is often to delist rather than absorb the liability. A token that trades freely in one market can therefore be untradeable in another, and the map keeps redrawing itself as each deadline passes. For a buyer, the question is no longer only whether a token is good. It is whether the token is listable where you live, and whether it will still be listable a year from now. A debut that quietly depends on a regulatory gap staying open is carrying a risk that never shows up on the chart.

Failure mode seven: the rug and the quiet exit

The modes above assume a real project behaving badly or a market structured against you. The seventh mode assumes nothing at all: the token exists only to be sold. On permissionless launchpads, where anyone can mint and list a token in a single click with no gatekeeper, this is not an edge case but the dominant outcome. A CoinGecko study of 18.67 million tokens launched on Solana’s pump.fun between January 2024 and June 2026 found that about 68.67% stopped trading on their launch day, and only 4.55% were still alive after 90 days. Solidus Labs data cited alongside the study found that 98.6% of pump.fun tokens showed rug-pull behavior, and barely more than 1% ever graduated to a real decentralized-exchange listing on Raydium.

Not every rug is a cartoon exit scam where the deployer pulls the liquidity pool and vanishes. Some are slow: the team stops shipping, the Discord goes quiet, the treasury trickles out over months. And a close cousin of the rug is the exploit, where the token is fine but the contract or the bridge behind it is not, draining holders through code rather than through a cap table, as the $285 million Drift incident showed. From the buyer’s seat the distinction barely matters. Whether the money leaves through a liquidity pull, a quiet abandonment, or a smart-contract hole, the outcome is the same balance of zero. The defense is the same too: assume a brand-new, unaudited, anonymous-team token on a permissionless venue is guilty until proven innocent, because the base rate says it almost always is.

The seven ways a listing dies, at a glance

Each failure mode has a tell. None of them require inside information to spot; they require only that you look at the right field before the chart distracts you.

Why the venue cannot save you

The instinct that a listing on a big, regulated exchange means someone vetted the token is understandable and mostly wrong. A listing is distribution, not endorsement. Even the most selective desks are reviewing legal classification, compliance risk, and technical security to protect the exchange from liability, not running an investment analysis to protect your downside. A token can pass every one of those checks and still be a low-float time bomb with a six-month cliff, because none of those are the exchange’s problem. The exchange gets paid on volume either way, an incentive we unpack in our look at the business model behind listings.

It is worth being precise about what that review actually covers. A desk like Coinbase runs a candidate token through legal classification (is it a security?), compliance and sanctions screening, and technical and smart-contract security before it goes live. Every one of those questions is designed to protect the exchange. None of them asks whether the float is too thin, whether the unlock schedule will crush the price in six months, or whether you are buying at the top of a market-maker-engineered candle. The review is real work, and it screens out outright fraud more often than not. It simply is not the review you need as a buyer, and no exchange has ever promised that it was.

Decentralized venues remove even the thin protection of a gatekeeper. On a permissionless DEX there is no listing committee, no risk tag, and no one to appeal to; the token is live the moment someone seeds a liquidity pool. That openness is genuinely valuable for access and for censorship resistance, and on-chain order books have matured enormously, as our profile of how one perp DEX took over on-chain futures and our explainer on what happens when you get liquidated both show. But openness and safety are different properties. The venue decides whether you can trade a token. It never decides whether you should. That judgment is always yours, which is why a repeatable checklist beats a gut feeling about a logo.

How to read a new listing before you buy

Every failure mode in this guide leaves a fingerprint in data you can pull up in a few minutes on a token’s market page, its documentation, and a block explorer. The checklist below is not a guarantee of profit; it is a way to refuse the listings that are rigged against you and size the rest honestly.

Run the archetype through it to see how fast the picture resolves. A token lists at a $2 billion fully diluted valuation with 8% of supply circulating, prints $60 million of first-day volume on a single exchange, names no market maker, and has a 20% investor tranche unlocking in five months. Nothing in that description is illegal, and the chart may well be green on day one. But the checklist has already lit up five of seven boxes: a thin float against a huge FDV, undisclosed market making, volume concentrated on one venue, and a near-term cliff. That is not a prediction that the token goes to zero. It is a statement that the odds, and the GSR base rate, are stacked against the buyer, and that any position should be sized as a speculative trade you can afford to lose rather than an investment you expect to compound.

If a listing clears all seven, it can still lose money; markets are markets. But a token that fails three or four of them is not an investment thesis, it is a countdown. The checklist cannot tell you what will go up. It is very good at telling you what is built to go down.

What the rulebook is doing about it

None of these failure modes are secrets to the people writing the rules, and 2026 has seen real, if uneven, attempts to defuse them. Binance’s market-maker disclosure regime attacks mode two directly by dragging the loan-plus-option deal into the light. MiCA’s white-paper requirement forces a named, accountable issuer behind listed assets in Europe, which both triggers the regulatory delistings of mode six and raises the floor for what can be listed in the first place. The SEC’s Project Crypto, with its proposed token taxonomy and a time-limited innovation exemption, is trying to give honest projects a compliant path to distribute tokens without the fundraising theater that produced so many low-float launches, though the agency is attempting all of this while badly short-staffed, a tension we cover in the Q4 countdown at a smaller SEC.

It is worth being clear-eyed about the limits. Disclosure rules make bad structures visible; they do not make them illegal. A monitoring tag warns you a token is at risk; it does not refund you when the risk arrives. GSR’s own recommendation, drawn from its data, is for the industry to move toward fairer pricing, higher initial float, and genuine community distribution. That is the right diagnosis, but it is a request, not a regulation, and the incentives that produce thin floats and fat FDVs have not gone anywhere. Until they do, the rulebook shrinks the worst abuses at the margin while leaving the core math, little supply now and a lot of supply later, fully intact.

The bottom line

A listing feels like an arrival, the moment a project graduates into the real market. The data says it is closer to a starting gun for the risks that actually cost people money. The median debut is underwater within three days; the richest-valued debuts keep less than a fifth of their value within a year; and the ways they get there (thin floats, dumping market makers, fake volume, unlock cliffs, slow delistings, regulatory removals, and outright rugs) repeat with enough regularity to be named, tabled, and checked.

That is also the hopeful part. Because the failure modes are structural rather than random, they are legible. You cannot predict which new token will moon, but you can usually tell which ones are engineered to bleed, and you can tell before you buy rather than after. Treat the first green candle as a question, not an answer. Read the float, the unlocks, the market-maker terms, the volume, the tags, the jurisdiction, and the code. Most of the time the listing will tell you exactly how it plans to end. You just have to look past the confetti to where it is written down.

Frequently Asked Questions

Do most new crypto listings go up or down after they list?

Up first, down soon after. A GSR study of more than 2,300 exchange listings since 2013 found the median token fell below its debut price within three days and was down roughly 50% within 90 days. The listing-day pop is real, but it is usually the high, not the floor.

Why do new tokens pump and then dump right after listing?

Three forces stack up: a thin circulating float that makes the first candle easy to push, market makers and early holders selling into that first wave of demand, and trading volume that is often partly wash-traded, so the liquidity looks deeper than it is. When the buying slows, there is little real support underneath.

What is a low-float, high-FDV token and why is it risky?

It is a token that lists with only a small share of its total supply in circulation but a very large fully diluted valuation. GSR found median circulating supply at launch fell from about 97% for sub-$10 million projects to around 13% for those valued above $1 billion. The danger is that future unlocks keep adding supply for months or years, so the price can bleed even if nothing goes wrong.

What happens to my tokens if an exchange delists them?

A delisting is usually staged: the exchange cuts margin and new trading first, then stops deposits, then closes withdrawals weeks later, sometimes auto-converting leftover balances to a stablecoin. You keep the asset, but liquidity collapses, so prices typically fall 25% to 30% on the news. The practical move is to withdraw before the deadline and trade on a venue where the token still has a market.

How can I tell if a new listing is safe to buy?

There is no safe, only informed. Check the circulating float against the fully diluted valuation, the unlock schedule, whether the exchange carries a risk or monitoring tag, whether trading volume matches on-chain liquidity, and the token’s regulatory status in your region. If the float is tiny, the unlocks are large, and the volume sits on one obscure venue, treat the listing as a trade, not an investment.

Marcus Halloran covers market structure, exchanges, and token launches for HOGE Wire.