Token Float Design: How Much Supply to Release at TGE

TGE float is a market-structure decision before it is a storytelling decision

Float at TGE is the single most watched number in a token launch because it is the market’s first shorthand for tradable supply, dilution risk, and how serious the team is about price discovery. If you launch below 5%, the market often reads it as manufactured scarcity. If you launch above 25%, you often give away too much headroom and too much inventory too early. The useful range is usually in between, but only after you model liquidity and unlocks together with listing mechanics.

Market cap is based on circulating supply, while FDV prices in the full supply. That means the gap between the two is not cosmetic. It is the first public estimate of future dilution. Binance Research argues that a large FDV-to-market-cap gap implies future inflationary and selling pressure that has to be understood in timing terms, not just headline valuation terms.

Data platforms treat circulating supply as a core market-quality input, not a vanity metric. CoinMarketCap says verified circulating supply generally requires material trading activity or volume on at least three tracked exchanges and enough project disclosure to satisfy its methodology. That matters because float is not only an investor talking point. It directly affects how data providers, exchanges, and market makers interpret the asset at launch.

Low float has become common enough to be its own market regime. CoinGecko found that 21.3% of the top 300 crypto assets by market cap had a market-cap-to-FDV ratio below 0.5 as of May 8, 2024, which it classed as low float. That is not a niche corner case anymore. It is a recurring design choice with recurring consequences.

The failure mode is not “bad tokenomics.” It is bad liquidity and bad timing

Very low float usually fails through the order book before it fails through narrative. Thin float can create the appearance of strength because small net buys move price quickly. But that same thinness leaves the token fragile. The first meaningful unlock, market-maker inventory adjustment, or community realization event can overwhelm the book.

CoinGecko’s 2024 Q2 launch study is a useful reality check. Of the 22 new projects it examined, only two launched with more than 20% circulating supply. The average launch FDV across the set was $4.7 billion. Starknet launched at $19.5 billion FDV, Wormhole at $13.3 billion, and Ethena at $11.8 billion. CoinGecko’s conclusion was blunt: as unlocks occurred, supply hit a market that could not absorb it, pushing down prices and valuations.

Very high float fails differently. It usually does not break through a cliff. It breaks through weak sponsorship. If too much supply is liquid on day one, early holders can distribute into the first wave of demand, price discovery gets messy, and the token struggles to establish a clean post-listing bid. Tokenomist’s launch analysis is useful here because it shows that higher float is not automatically bad. Allocation quality matters. HYPE worked with a moderate float and delayed unlocks. PUMP and PENGU show that a higher float still underperforms if the liquid inventory is held by sellers with low cost basis or weak holding incentives.

This is where the burn-skeptic lens matters. A promise to burn tokens later does not fix a float that is wrong today. Binance Research explicitly frames buy-and-burn as one possible form of revenue sharing. That only matters if the protocol is actually generating fees or cash flows that can fund it. Scarcity without economic throughput is optics.

The right TGE float comes from four hard constraints

The first constraint is exchange distribution strategy. If the launch includes Launchpool, HODLer airdrops, or a broad exchange rollout, the token can support a higher float because access is wider and initial holder concentration is lower. Binance’s Launchpool mechanics explicitly distribute tokens to users who lock eligible assets, which means the launch channel itself can create initial ownership breadth.

The second constraint is liquidity provisioning budget. Float should be back-solved from the amount of stablecoin and token inventory you can commit to market making across venues. Kaiko’s liquidity framework is the right language here. It evaluates assets using 0.1% depth, 1% depth, spreads, volume, and the number of liquid exchanges. If your float decision is not anchored to target depth and spread, it is not really a float decision. It is branding.

The third constraint is community distribution commitments. A launch with a real airdrop, launchpool allocation, or onchain rewards program needs enough float for users to actually trade, stake, or spend. Tokenomist’s review of successful launches found a recurring pattern: the better post-TGE performers reserved meaningful supply for community and ecosystem from day one, while keeping public investor allocations modest. It specifically notes that ONDO and ENA each gave public investors only 2% allocations, mostly unlocked at TGE, while managing broader supply pressure tightly.

The fourth constraint is market-making spread targets. Tier-1 CEX launches generally require tighter quoted spreads and thicker near-mid depth than a DEX-only debut. Kaiko’s work is useful because it shows how professionals actually think about tradability. A token can look large by market cap and still be weak by liquidity rank. If the market maker can only support shallow 0.1% depth, then a tiny float is dangerous because every incremental seller becomes a volatility event.

The practical implication is simple. Float is where tokenomics meets market microstructure. You do not pick it from a generic template. You derive it from venue count, airdrop size, treasury inventory, stablecoin budget, and the spread and depth you want to defend in the first weeks of trading.

TGE float only matters in the context of the next 6 to 12 months

The canonical failure pattern is floor-in-then-collapse. Teams optimize the screenshot at TGE, get the low-float pump, and then discover that month-6 to month-12 supply is what the market was pricing all along. If the first meaningful cliff adds too much new inventory relative to daily turnover and order-book depth, the initial float was never really low. It was just deferred.

Starknet is a clean example of why the unlock path matters more than the headline launch float. Its token docs state that up to 0.64% of total supply, or 64 million STRK, unlocked on the 15th of each month from April 15, 2024 through March 15, 2025. From April 15, 2025 through March 15, 2027, the monthly amount stepped up to 1.27%, or 127 million STRK. That is exactly the kind of schedule that can dominate market behavior long after the TGE headline is forgotten.

Celestia shows the same point from a different angle. Celestia’s docs say the network launched with 1 billion TIA at genesis. Public allocation was fully unlocked at launch, while the 26.79% R&D and ecosystem bucket had 25% unlocked at launch and the rest unlocking continuously from year 1 to year 4. Initial core contributors, seed backers, and Series A&B backers each had 33.33% unlock at year 1, with the remainder vesting over years 2 or 3. Because 2024 was a leap year, the first year unlock landed on October 30, 2024. That is the number that should have been modeled from day one.

Ethena is another useful case because its schedule is explicit. Ethena’s tokenomics page says core contributors hold 30% of ENA and are locked on a 1-year 25% cliff followed by 3 years of linear monthly vesting. Investor tokens follow the same structure, and the unlock schedules started at TGE on March 5, 2024. That means any serious float analysis had to map the first anniversary event, not just day-one circulating supply.

Tokenomist’s launch work reinforces the same lesson from performance data. It highlights HYPE and ONDO as cases where minimal early unlocks reduced sell pressure, while ENA and TIA had smaller starting floats but more noticeable relative supply growth, which created volatility. The point is not that small float is wrong. The point is that small float plus aggressive early unlocks is internally contradictory.

Working float ranges by project type

Most launches should start from a range, not a magic number. The exact number still has to be modeled against your listing plan and liquidity budget, but some ranges are consistently more defensible than others.

Those ranges are intentionally wider for consumer tokens because consumer tokens live or die on actual usage. Pixels’ lite paper is unusually clear on this. It argues that long-term value in a game economy comes from gameplay, entertainment, and premium utility, not speculative future earnings. That is the right instinct. Consumer tokens can tolerate a somewhat higher float when users have reasons to spend, customize, or progress. Without those sinks, a low float just manufactures scarcity and delays the same repricing.

The outer bounds still matter. Under 5% is usually too low unless the launch is deliberately constrained, product-led, and supported by extremely light near-term unlocks. Over 25% is usually too high unless the token already has a genuinely broad user base and a clear reason for that liquid supply to stay engaged. CoinGecko’s 2024 sample is telling here. Most hyped launches came in below 20%, and even that often proved too tight once the unlock schedule started doing the real work.

How to pick the actual number

The correct float is the one that your liquidity plan can defend and your unlock schedule does not invalidate. In practice, that means working in this order.

1. Set the venue map. Decide whether launch is CEX-led, DEX-led, or hybrid as you launch a token. Count how many books and pools need support on day one.
2. Set tradability targets. Define acceptable spread, target 0.1% depth, and expected slippage for your primary pairs.
3. Budget real liquidity. Allocate stablecoins and token inventory to market making and LPs. Do not assume float itself creates liquidity.
4. Map 6-, 9-, and 12-month net new supply. Measure cliff and linear unlocks as a percentage of starting circulating supply and as a percentage of expected daily turnover.
5. Stress-test holder mix. Community airdrop supply, market-maker inventory, foundation wallets, and investor unlocks do not behave the same way. Treat them differently.

If the model says you need a 7% float to create price excitement, but the next two unlocks add 30% to circulating supply and your liquidity budget only supports thin books, the answer is not 7%. It is that the launch plan is internally inconsistent. If the model says you need 22% float to satisfy exchange distribution, launchpool, LP inventory, and user claims, that can be fine, but only if the liquid inventory is not concentrated in short-horizon sellers.

Binance Research makes the deeper point well: without token demand, supply design is of limited value, and a token only sustains value if it captures protocol value or offers real utility. That is the final check on TGE float design and on proper tokenomics more broadly. Scarcity is not demand. Burns are not demand. Deferred unlocks are not demand. The market eventually asks the same question every time: who needs to own this token after the launch trade is over?

At FinDaS Tokenomics, this is the point where token economy design stops being a slide-deck exercise and becomes market design in practice. A chatbot gives you a range. A token economist gives you the number that fits your specific listing plan, liquidity budget, holder mix, and month-6 unlock profile.