An altcoin can have an active community, shipping developers, rising transactions, partnerships, and even growing protocol revenue while its token falls 60%, 80%, or more.
That is not necessarily a contradiction.
The mistake is treating project health, token demand, and market liquidity as the same variable. They interact, but they are not interchangeable. Product activity is an economic variable. Token price is an asset-pricing variable. Liquidity is a balance-sheet and market-structure variable.
A useful way to frame the problem is:
Price does not need a fundamental failure to collapse. It only needs the marginal supply of tokens to exceed the marginal balance sheet willing to absorb them at the current price.
That distinction becomes especially important in altcoins because liquidity is concentrated. ESMA’s granular analysis of crypto order books found that liquidity differed substantially across venues, was considerably better for the most heavily traded assets, and was concentrated alongside trading activity. Less-traded assets had lower volume and wider spreads. Its 2023 sample also found that the five largest exchanges accounted for 73% of trading volume. See the official ESMA market structure analysis.
So the first principle is simple:
Market capitalisation measures valuation. It does not measure exit liquidity.
A token can be “worth” billions on paper while only a few million dollars of executable bids stand between the current price and a very different clearing price. ESMA’s own order-book work illustrates the distinction: even for Bitcoin, measured 2% depth varied sharply by exchange, while smaller assets generally exhibited lower volume and larger spreads.
The central mistake: treating supply as a tokenomics statistic instead of a liquidity problem
Data: An unlock does not automatically mean that its recipient will sell. Nor does an unlock necessarily create new total supply. A vesting unlock usually changes the amount of existing supply that is transferable, while an emission can create genuinely new units. Treasury distribution can move already existing inventory into the hands of actors more likely to spend or sell it.
Interpretation: From the order book’s perspective, however, the relevant question is similar in all three cases:
How much potentially sellable inventory is becoming economically mobile relative to the amount of capital willing to absorb it?
That is why looking only at “unlock as percentage of circulating supply” can be misleading.
Consider two hypothetical tokens:
| Variable | Token A | Token B |
|---|---|---|
| Market capitalisation | £2bn equivalent | £300m equivalent |
| Next 30-day unlock | 2% of circulating supply | 5% of circulating supply |
| Unlock market value | £40m | £15m |
| Aggregate 2% bid depth | £3m | £12m |
| Unlock value / 2% bid depth | 13.3x | 1.25x |
| Likely liquidity problem | Potentially severe | Potentially manageable |
Token A has the “better” headline tokenomics because only 2% unlocks. Yet the newly mobile inventory is enormous relative to its visible market depth.
That is the variable I care about.
An industry study by crypto market-maker Keyrock analysed more than 16,000 unlock events and reported widespread negative price pressure around unlock windows, with effects often appearing before the nominal unlock date. It also found substantial differences by unlock type and recipient. The methodology normalised token performance against ETH and followed prices before and after events. This is useful evidence, but it should be treated as observational market research rather than proof that the unlock itself caused every move. Their methodology and results are laid out in Keyrock’s 16,000 unlock event study.
The pre-event effect matters. If market participants know that a large investor distribution is approaching, they do not have to wait for tokens to become transferable. Existing holders can reduce exposure, arbitrageurs can alter positioning, recipients may hedge where derivatives permit it, and market makers can change their willingness to warehouse inventory. Keyrock’s study explicitly observed substantial price and volume changes before many large unlocks and described pre-hedging as one possible mechanism.
So:
Unlock date is not necessarily sell date.
And:
Sell pressure can arrive before physical token supply arrives.
Details / Collapsible Block: The supply distinction that prevents a lot of bad analysis
An unlock usually makes previously restricted tokens transferable. It changes liquid or potentially liquid float.
An emission creates or distributes recurring token supply according to protocol rules, such as validator rewards, liquidity incentives, ecosystem incentives, or inflation schedules.
A treasury sale converts a project’s existing token reserves into another asset to fund salaries, operations, grants, acquisitions, taxes, legal costs, or ecosystem programmes.
A staking withdrawal can convert economically immobilised inventory into transferable inventory even when there is no conventional vesting event.
These mechanisms should not be added together blindly. Tokens can move between categories, and an unlocked token is not necessarily sold.
The better framework is to estimate three cases:
Low supply case: only clearly identifiable structural sellers monetise.
Base supply case: some unlocked, emitted, or treasury inventory reaches the market.
Stress supply case: several holder classes attempt to monetise while market depth is simultaneously shrinking.
The stress case is important because supply and liquidity are not independent. A falling price can increase the desire to sell at the exact moment that liquidity providers want to reduce exposure.
The plumbing beneath the price
The failure mechanism usually involves several systems at once. This is why I am suspicious of explanations such as “VC unlocks killed it”, “the market maker dumped it”, or “the narrative died”.
Those explanations may describe part of the event. They rarely describe the system.
Market depth is the scarce resource
A quoted price tells you where the last marginal transaction cleared.
It does not tell you how much can transact there.
For market structure, I would rather know the amount of bid liquidity within 1%, 2%, and 5% below spot than the token’s market capitalisation. ESMA uses order-book depth and bid-ask spreads as explicit measures of crypto market liquidity and found that the most heavily traded assets and largest venues generally had better liquidity.
There is another complication: displayed liquidity is conditional liquidity.
A bid sitting 1% below price today is not a contractual commitment to remain there during a 20% sell-off. If volatility increases, inventory becomes harder to hedge, cross-exchange arbitrage becomes less reliable, or informed sellers appear, a rational liquidity provider can reduce quote size or move the bid lower.
That produces the familiar liquidity paradox:
The depth you measure in calm conditions is often most likely to disappear precisely when you need it.
So market depth should be stress-tested, not merely observed.
Market-maker inventory can turn a stabiliser into an amplifier
Market makers are not bottomless buyers.
Suppose a market-maker starts approximately neutral. Unlock recipients begin selling. The market-maker buys from those sellers to maintain a two-sided market.
Its inventory becomes increasingly long.
Now it has choices: hedge elsewhere, transfer tokens across venues, find an OTC counterparty, sell futures, widen its quotes, reduce bid size, or allow the market to clear lower.
If external buyers are abundant, inventory transfers smoothly.
If external buyers are scarce, the same market-maker that initially absorbed selling eventually has to protect its balance sheet. Bids become smaller or cheaper.
This is one reason a market can appear healthy immediately after an unlock and weaken later. The first buyer may not represent new fundamental demand. It may represent an intermediary temporarily warehousing someone else’s supply.
The relevant question therefore is not simply:
“Is there a market maker?”
It is:
“Who ultimately takes the market maker’s inventory, and at what price?”
Exchange concentration creates hidden single points of failure
A token can be listed on ten venues and still effectively have one market.
What matters is where executable depth sits.
ESMA’s empirical work found substantial concentration at the exchange level, with liquidity likewise concentrated rather than evenly distributed.
Suppose 65% of a token’s genuine 2% bid depth exists on one exchange.
A regulatory restriction, market-maker departure, technical suspension, removal of margin eligibility, or complete delisting does not remove only “one out of ten exchanges”. It can remove most of the economically relevant market.
That leads to a particularly unpleasant feedback loop:
low liquidity → reduced market-maker economics → lower volume → listing review → pair removal or delisting → fewer arbitrageurs and buyers → still lower liquidity
This mechanism is not hypothetical exchange policy. Binance explicitly says that it periodically reviews tokens and trading pairs and lists poor liquidity, low market capitalisation, low trading volume, legal issues, security concerns, and other risks among potential reasons for removal. Its current criteria are set out in Binance’s delisting guidelines.
That gives listing status a reflexive element.
Liquidity can affect listing security, while listing security can affect liquidity.
A product can therefore remain alive while the token’s financial distribution network deteriorates.
Derivatives can create more risk capacity than the spot market can settle
Perpetual futures and other derivatives complicate the picture because speculative exposure can expand without equivalent spot-market capital entering.
Open interest is not itself bullish or bearish. Every derivatives contract has counterparties, and gross open interest should never be mechanically treated as future sell volume.
But leverage changes the response function of participants.
BIS research on crypto futures finds evidence that past crypto price surges coincided with buying pressure from smaller, trend-following investors seeking leveraged upside, and argues that this can create feedback loops that amplify mispricing. The paper also highlights segmentation between venues and participants as an important feature of crypto derivatives markets. See the BIS research on crypto carry.
The dangerous configuration for an altcoin is not simply high open interest. It is something closer to:
large derivatives exposure + shallow spot depth + crowded positioning + collateral sensitivity + few independent liquidity venues
Consider the positive loop first:
price rises → collateral values rise → risk capacity expands → leverage increases → futures buying increases → arbitrage and hedging affect spot → price rises
Then reverse it:
price falls → leveraged equity shrinks → margin headroom falls → liquidations or voluntary deleveraging begin → liquidity providers face larger directional flow → quotes retreat → spot price gaps lower → more positions breach risk limits
BIS evidence on leveraged trend-following pressure is consistent with the possibility of such reinforcing feedback in crypto derivatives, although the exact liquidation mechanism depends on the exchange, collateral rules, leverage, and position distribution.
This becomes more reflexive when the same altcoin is accepted as collateral.
Now the asset being sold is also part of the system financing the leveraged position.
That creates a double hit: the position can lose money while its collateral simultaneously becomes less valuable.
Narrative rotation is a balance-sheet event, not just a social-media event
Narratives sound soft. Their consequences can be very mechanical.
Suppose capital decides that one sector no longer deserves a premium and reallocates to another.
The immediate effect is not that the abandoned project’s code stops working. The effect is that the marginal buyer disappears.
Then second-order effects follow.
Lower speculative demand reduces turnover. Lower turnover worsens the economics of allocating market-making capital to that pair. Lower depth raises execution costs for larger investors. Higher execution costs make the asset less attractive to those same investors. Weak relative performance prompts portfolio rebalancing. Declining volume can make exchange support less economically attractive.
So:
attention → flow → volume → market-maker economics → depth → execution quality → institutional accessibility → flow
That loop can operate positively or negatively.
Product activity may continue throughout the entire process.
The token only escapes the loop if activity creates sufficiently strong token demand, cash-flow capture, collateral demand, staking demand, governance demand, fee demand, buybacks, burns, or some other mechanism capable of absorbing the marginal supply.
This is why “users are growing” is incomplete token analysis.
The missing question is:
What specifically forces, incentivises, or economically motivates those users or another group to hold the token?
How the liquidity cascade forms
The most severe altcoin declines usually become understandable once several individually manageable risks overlap.
A stylised cascade looks like this:
| Stage | Initial change | First-order effect | Second-order effect |
|---|---|---|---|
| Supply | Unlock, emissions, treasury distribution | More inventory can be sold | Existing holders anticipate selling |
| Market making | Dealers accumulate inventory | Bids absorb initial flow | Dealers hedge, widen spreads, or reduce bids |
| Spot liquidity | Depth falls | Equal sell size causes more slippage | Volatility rises |
| Derivatives | Leveraged longs lose equity | Deleveraging begins | Forced flow accelerates |
| Collateral | Token collateral falls in value | Borrowing capacity falls | Additional collateral or deleveraging required |
| Narrative | Relative performance deteriorates | Marginal buyers rotate elsewhere | Volume and market-maker economics deteriorate |
| Venue | One exchange loses liquidity | Concentration rises elsewhere | Delisting or pair-removal risk can increase |
| Price | Market clears lower | Market cap falls | Risk limits tighten and further capital exits |
Interpretation: The crucial feature is endogeneity. Sell pressure worsens liquidity, and worsening liquidity increases the price impact of subsequent sell pressure.
That means a 10% increase in tradable supply does not imply a 10% fall in price.
There is no fixed conversion.
If deep buyers exist, a large unlock can be absorbed with modest price impact.
If bids disappear, comparatively small selling can produce an enormous price adjustment.
Keyrock’s unlock dataset is consistent with the importance of this distinction: its results vary materially by unlock structure and recipient class, rather than showing a simple one-for-one relationship between unlock percentage and subsequent performance.
This gives us a more useful model:
Price impact is a function of marginal supply divided by resilient marginal liquidity, multiplied by leverage and reflexivity.
Not a literal pricing equation, but a much better mental model than:
unlock = dilution = price falls by X%
A liquidity stress test before thinking about position size
This is not a trading signal and it cannot identify a fair value. Its purpose is narrower:
Can the market plausibly absorb your scenario without liquidity itself becoming the dominant risk factor?
The inputs should ideally be collected repeatedly, because a single order-book snapshot can give false comfort.
Build the liquidity map
Measure bid depth at roughly 1%, 2%, and 5% below the mid-price on every economically meaningful venue.
Use the median of several observations rather than the best snapshot.
Then calculate:
[ D_{2%} = \sum \text{visible bids within 2% below mid-price} ]
Call that aggregate 2% bid depth.
Now calculate venue concentration:
[ C = \frac{\text{2% bid depth on largest venue}}{\text{aggregate 2% bid depth}} ]
A high value does not automatically make the asset bad. It tells you that apparently diversified liquidity may actually have a single point of failure.
ESMA’s findings give a strong reason to perform this calculation rather than assume liquidity is evenly distributed across exchanges.
Calculate potentially mobile supply
For the next 30, 60, and 90 days, separately estimate:
[ S = U + E + T + W ]
where:
U = unlocked tokens that become transferable
E = new emissions or incentive distributions
T = plausible treasury distributions or sales
W = staking or other economically locked inventory becoming withdrawable
Do not assume 100% gets sold.
Run multiple cases instead.
For example:
| Assumption | Potential monetisation |
|---|---|
| Low case | 10% of newly mobile supply |
| Middle case | 30% |
| Stress case | 60% |
| Extreme test | 100% |
Those percentages are deliberately heuristic scenarios, not empirical rules.
Convert each case into market value and divide it by depth:
[ \text{Supply Absorption Ratio} = \frac{\text{potentially sold new supply in £ or $}} {\text{median aggregate 2% bid depth}} ]
This ratio is not expected slippage. It asks a different question:
How large is the approaching inventory relative to the market’s current visible capacity?
That is more informative than comparing the unlock only with market capitalisation.
Destroy some of the liquidity on purpose
Now make the test unpleasant.
First, remove the largest exchange.
Then reduce the remaining displayed depth by 25%, 50%, and 75%.
Define:
[ D_{stress}
(D_{aggregate} - D_{largest\ venue}) \times h ]
where h might be 0.75, 0.50, or 0.25 depending on the severity being tested.
These haircuts are not forecasts. They are hypothetical tests of dependence on displayed liquidity.
Then recalculate:
[ \text{Stressed Supply Ratio}
\frac{\text{potential sell supply}}{D_{stress}} ]
If the asset looks extraordinarily liquid in the normal snapshot but extraordinarily fragile after one venue disappears and quotes are halved, you have learned something important.
Measure the position footprint
For any hypothetical exposure P:
[ \text{Liquidity Footprint}
\frac{P}{D_{stress}} ]
Again, this is not a recommendation about what P should be.
It tells you whether an exposure that appears small compared with market capitalisation is large compared with stressed executable liquidity.
A portfolio might own only 0.02% of a token’s market capitalisation while representing a substantial fraction of the bids available near the current price.
Those are completely different risk statements.
Stress the derivatives layer
Record spot volume, open interest, funding rates, basis where available, and changes in those variables.
I would not create a universal “safe” open-interest ratio. Market structure varies too much.
Instead, ask:
Does leveraged exposure appear to be growing faster than spot-market depth?
Is price appreciation accompanied by deeper spot bids, or primarily by more derivatives activity?
If open interest were rapidly cut, is there enough independent spot liquidity to accommodate the associated hedging and deleveraging flows?
Is the token itself widely used as collateral, making falling prices potentially procyclical?
BIS research provides evidence that leveraged, trend-following demand can participate in reinforcing crypto price dynamics, which is why derivatives should be treated as a liquidity multiplier rather than analysed in isolation.
Run the delisting test
Remove the dominant venue completely.
Then ask:
Can price discovery migrate somewhere else?
Are meaningful spot pairs available elsewhere?
Can arbitrageurs transfer the asset quickly enough?
Does most derivatives liquidity depend on the same venue?
Would removal from margin or collateral programmes matter even if spot trading continued?
This is not an absurd stress. Binance explicitly identifies poor liquidity and trading volume among the factors it may consider when reviewing trading pairs, alongside regulatory, ethical, security, and project-related considerations.
Run the narrative test
Finally, assume nothing is technically wrong with the project.
Only change capital allocation.
Halve speculative spot volume.
Reduce visible depth by 50%.
Assume sector-relative performance remains weak for several months.
Keep token emissions and unlock schedules unchanged.
Then rerun the test.
This isolates the exact issue the original thesis is trying to understand:
Could the token remain fundamentally active while financial demand becomes insufficient to absorb its financial supply?
A worked hypothetical
Imagine an altcoin with:
| Variable | Normal state |
|---|---|
| Market capitalisation | $1.5bn |
| Aggregate 2% bid depth | $6m |
| Share of depth on largest venue | 55% |
| Newly mobile supply next 30 days | $30m |
| Assumed middle-case monetisation | $9m |
| Hypothetical exposure being evaluated | $500,000 |
The headline market capitalisation makes a $500,000 exposure look tiny.
Now remove the largest venue:
[ $6m \times 45% = $2.7m ]
Assume the remaining quotes shrink by 50% during stress:
[ $2.7m \times 50% = $1.35m ]
The stressed supply ratio becomes:
[ $9m / $1.35m = 6.67 ]
And the hypothetical exposure’s liquidity footprint becomes:
[ $500,000 / $1.35m \approx 37% ]
Nothing in those calculations says the token must fall.
Nothing says the entire $9m will be sold simultaneously.
Nothing says the displayed book predicts actual execution perfectly.
But the exercise reveals something that market capitalisation hides:
The token’s valuation may be large while its marginal risk-bearing capacity is small.
That is what a liquidity stress test is supposed to expose.
Scenario map: what could happen and why
Because this is a general framework rather than an analysis of one token, assigning precise percentages would create false precision. I would frame probabilities conditionally.
| Scenario | Relative probability | Mechanism | Important triggers | Invalidation signals |
|---|---|---|---|---|
| Base case: slow liquidity erosion | Medium to high when supply growth exceeds organic token demand | Unlocks and emissions are absorbed gradually, but at progressively lower clearing prices | Persistent supply, falling depth, weak sector-relative volume, market-maker inventory pressure | Depth broadens, supply is absorbed without weaker spreads, token demand grows faster than liquid float |
| Upside case: absorption and rerating | Plausible when new demand arrives before liquidity breaks | New buyers absorb unlocks, listings broaden access, spot depth improves and narrative rotates back | Rising spot depth, broader venue participation, healthy spot-led volume, improving token sinks | Price rises while depth stagnates, leverage grows much faster than spot, supply schedule accelerates |
| Downside case: liquidity air pocket | Lower frequency but high severity | Unlocks coincide with venue loss, market-maker withdrawal or leveraged deleveraging | Top venue dominates depth, spreads jump, bids disappear, funding or leverage is stretched, collateral terms tighten | Cash buyers replenish books, open interest falls without spot stress, cross-venue depth remains stable |
| Structural recovery | Depends on token design | Product activity finally translates into persistent token demand | Fee capture, staking demand, reduced emissions, buybacks, burns, stronger collateral or settlement use | Product metrics improve but token velocity and net sell supply remain high |
Base case
Data to watch: depth, spreads, unlock schedules, emissions, treasury movements, volume, spot-versus-derivatives activity and venue concentration.
Interpretation: An active protocol can continue growing while the token gradually reprices lower because every rally meets structurally available supply. The market need not experience one catastrophic unlock.
The more common failure can be persistent small imbalance.
Thirty days of modest selling into gradually weakening demand can matter more than one dramatic event that everyone has prepared for.
Probability: This becomes my highest-probability conditional scenario when new liquid supply persistently exceeds observable token-demand growth but leverage remains sufficiently contained to avoid a disorderly cascade.
Upside case
The upside case is not simply “the project succeeds”.
The market needs a balance-sheet event.
A new listing may broaden access. Sector rotation may attract capital. Economic token utility may start creating real holding demand. Supply growth may slow. Unlock recipients may choose not to sell. Market makers may observe enough two-way flow to allocate more capital.
Then the feedback loop reverses:
buyers absorb supply → volatility falls → market makers quote more aggressively → depth improves → execution becomes easier → larger participants can enter → turnover rises → market-making economics improve → depth increases again
Probability: This becomes meaningfully more likely when price appreciation is accompanied by broader spot depth and diversified venue liquidity rather than merely increasing perpetual open interest.
That distinction matters.
Price up with deeper cash markets is different from price up with leverage up.
Downside case
The dangerous case is correlation between stressors.
An unlock arrives.
At roughly the same time, the market narrative weakens.
A major market-maker reduces exposure.
Depth falls.
Leveraged longs remain crowded.
Price begins declining.
Liquidations increase.
The token is used as collateral somewhere else.
Risk systems demand deleveraging.
An exchange then removes a margin pair, applies additional monitoring, restricts users in some jurisdictions, or eventually delists the asset.
Each event is manageable independently.
Together they create a nonlinear system.
That is why the deepest drawdowns can appear “irrational” when viewed solely through product fundamentals.
The product may not have changed much.
The financial network supporting the token has.
The research question that would actually test this thesis
Primary research question:
After controlling for BTC and ETH beta, sector performance, token age and circulating market capitalisation, does the ratio of newly liquid 30-day token supply to stressed aggregate 2% bid depth predict subsequent abnormal returns, spread widening and depth deterioration better than unlock size measured simply as a percentage of circulating supply?
This is testable, and its answer could materially change the framework.
The core hypothesis is:
[ \frac{\text{newly liquid supply}}{\text{resilient market depth}} ]
should explain more about liquidity stress than:
[ \frac{\text{unlock}}{\text{circulating supply}} ]
alone.
A serious test would collect historical unlock events, recipient categories, contemporaneous 1%, 2%, and 5% order-book depth across major venues, venue concentration, emissions, treasury flows, spot volume, derivatives open interest and funding. Keyrock’s findings provide evidence that unlock behaviour differs across event structures and recipients, while ESMA independently demonstrates the importance of depth, spreads, asset liquidity and venue concentration.
The dependent variables should not be price alone. I would examine abnormal token return, spread change, 2% depth change, realised volatility, venue concentration change and spot-to-derivatives volume after the event.
The strongest version of the hypothesis would predict that a nominally “small” unlock can be dangerous when depth is weak, while a nominally “large” unlock can be absorbed when resilient liquidity is deep.
What would falsify or weaken the thesis?
If unlock percentage of circulating supply consistently predicts outcomes just as well after depth, recipient type, leverage and market factors are introduced, then the liquidity-adjusted model adds little.
Likewise, if tokens with extremely high supply-to-depth ratios systematically absorb distributions without deterioration in spreads, depth or abnormal performance, the proposed mechanism needs revision.
That is the useful falsification test.
The thesis should be allowed to fail.
The broader conclusion is therefore deliberately narrower than “unlocks are bearish”:
Altcoin crashes are often best understood as clearing events. The token does not need to lose its community or product. A growing quantity of economically mobile inventory can encounter a shrinking quantity of risk capital, while leverage, collateral, market-maker inventory, exchange concentration and narrative rotation amplify the adjustment.
Data: Crypto liquidity is measurably concentrated across assets and venues, less-traded assets tend to have worse spreads, leveraged trend-following activity can create reinforcing dynamics, unlock effects vary substantially across events, and exchanges explicitly consider liquidity when maintaining markets.
Interpretation: Community strength can keep a network alive without providing enough marginal token demand to support the previous clearing price. Fundamentals determine what an asset may eventually deserve. Market structure determines whether holders can get from here to there without a liquidity accident.
Probability: No single variable is sufficient. The highest-risk configuration is the intersection of new supply, shallow depth, concentrated venues, dependent market makers, high leverage and fading marginal demand. Remove several of those conditions and the same unlock may be absorbed uneventfully. Combine them and a fundamentally active project can still experience a severe token drawdown.
That is calm uncertainty in its proper place: not asking whether an unlock is “bullish” or “bearish”, but asking who must absorb the inventory, with what balance sheet, through which venues, under what leverage, and what happens if those buyers step away.