Token Graveyard: 99.99% of Cryptocurrencies Will Ultimately Go to Zero

Original: Zeneca
Compiled: Big Pincer | PANews Lobster
The current number of tokens ranges from 37 million to 120 million, and the exact number is not important. What matters is that over 99.99% of tokens have actually failed.
According to different sources, the current number of existing cryptocurrencies ranges from 37 million to 120 million. The specific number depends on the statistical method and data source, but getting caught up in the details at this scale is not significant. No matter how you look at it, this is an astronomical figure.
Over the past year, I have repeatedly mentioned the need to concentrate holdings on a few key targets. Today, I want to share some data and research that support this viewpoint and strategy.
Let's get straight to the point.
How many tokens are there, really?
The answer depends on your data sources and statistical methods; different platforms have different statistics:
CoinMarketCap: Cross-chain indexed token count exceeds 37 million
Dune Analytics: Tracks unique tokens on major chains, about 74.5 million
Tangem: Citing January 2026 on-chain data, over 120 million on major networks
The differences stem from the definition of "token": are all deployed smart contracts counted? Only those that have had at least one transaction? Or only those that are still actively traded? Each filtering criterion will yield different numbers.
But in any case, all sources are highly consistent on three points:
The growth rate is astonishing
Most tokens are dead or nearly dead
Real value is concentrated in a very few tokens
Failure rate approaches 99.99%
A data point circulating in the community — 53.2% of crypto tokens have failed, from a study released by CoinGecko in January 2026. This study is good in itself, but I see several flaws:
It only counted about 20 million tokens that entered GeckoTerminal and had trading records, excluding a large number of tokens that perished before getting started after being created
It defines "inactive" as failure; but in my view, extremely low activity combined with a 99% price drop is also a failure
As mentioned earlier, the actual number of created tokens is far higher than this, and the failure rate is similarly far higher.
Memento Research tracked 118 token generation events (TGE) in 2025. These are typically backed by venture capital, have teams and roadmaps, and go through formal TGE processes (not randomly issued meme coins), of which 84.7% are currently trading below their issue valuation — and I still consider that figure to be low.
The median token drop is 71%. The worst-performing projects were issued with fully diluted valuations (FDV) in the hundreds of millions or even billions, then fell 85%-93%. If the best-resourced projects in the crypto market cause 85% of buyers to lose money, what do you think the fate of the other 74.5 million tokens will be?
Let me tell you: among the 74.5 million tokens tracked by Dune Analytics, about 500 have a market cap exceeding $10 million, accounting for 0.0007%.
Cryptocurrency failure is not limited to any one category but is a total collapse across the board. As I said before: a good starting point is to assume every token is heading to zero and then look for the extremely rare exceptions.
Meme coin failure: 99.67% of Pump.fu tokens have never "graduated" (i.e., have not reached a market cap of $90,000)
ICO failure: 80% of ICOs in 2017 were scams, and by 2020 nearly 90% of surviving tokens fell below their issue price
TGE failure: See the above Memento Research data
Airdrop failure: Most airdropped tokens are sold off within hours of distribution and do not recover
VC project failure: High FDV and low circulation issuance methods have become a hallmark disaster of 2025
Creator token failure: Celebrity tokens from politicians to influencers typically plummet over 90% within days
AI agent token failure: The AI narrative from 2024-2025 spawned hundreds of tokens, the vast majority down over 80% from their peaks
Game token failure: The Play-to-Earn craze of 2021-2022 created dozens of tokens, most of which are now worth less than a cent
L1 failure: Remember when Fantom was at $3? When Luna was at $100?
L2 failure: Most L2 tokens underperform ETH, and ETH itself also lags behind BTC
Secret issuance failure, fair launch failure, governance token failure, utility token failure
This principle runs through every category, every issuance mechanism, every narrative, and every market cycle. Almost all tokens trend toward zero. Exceptions are rare: Bitcoin, Ethereum, Solana, Hyperliquid, BNB, and a very few other projects.
This is the reality of this damn market.
Why are there so many coins?
In short: creating tokens has never been so easy and cheap.
Pump.fun launched on Solana in January 2024, allowing anyone to create a token in 60 seconds — no programming required, costs nearly zero. Pick a name, upload an image, click create, and you’re done. You now have a cryptocurrency.
A research paper published in February 2026 analyzed a month's worth of data from Pump.fun: In September 2025, 243,123 different wallet addresses created a total of 655,770 tokens, of which only 4,338 "graduated" to decentralized exchanges, with a graduation rate of 0.63%.
By mid-2025, Pump.fun deployed over 80% of Solana tokens, while Solana accounted for about 64-70% of all tokens ever created on-chain.
In other words, a platform deployed on one chain has created the majority of the total tokens in the crypto market. Among them, 99.37% failed before reaching a market cap of $90,000.
Where is the money?
This is the most important part for your portfolio.
Bitcoin accounts for about 56 cents of every dollar in the crypto market. With Ethereum and stablecoins, that share rises to 79%. The top 10 tokens together account for about 90% of the total market cap. The remaining approximately $230 billion is scattered across tens of millions of other tokens.
The average data for non-top 100 tokens is shocking: even if $230 billion is evenly distributed among about 17,000 tracked tokens, the average market cap is only about $13 million. And this average is significantly inflated by a few hundred mid-cap tokens; the median is much lower, much lower. The market cap of millions of tokens is effectively zero.
Survivorship funnel
A picture is worth a thousand words; it basically summarizes the entire content of this letter:
[Survivorship funnel illustration]
Over time, this differentiation will only become more extreme.
Actual insights
The market follows a power law distribution
The top ten tokens hold over 90% of the total crypto market cap. Holding Bitcoin and Ethereum means you hold the most important assets by market cap weight. This pattern has persisted for years.
Token creation ≠ value creation
Tens of millions of tokens exist, but the vast majority were created to make money for the creators, not to benefit the buyers.
The haystack is getting bigger, but the needles aren't increasing.
Finding legitimate projects with real use is becoming increasingly difficult every month. The signal-to-noise ratio is worse than ever. More tokens do not equal more opportunities; on the contrary — it means more noise and harder-to-grasp opportunities.
Survivorship bias is everywhere
You’ve only heard about that meme coin that went up 1000 times, but you haven’t heard about those that went to zero among the 655,000 tokens issued in the same month. Success stories fill the feeds on X (Twitter), while failures go unnoticed.
Liquidity is the real filter
CoinGecko tracks about 17,000 tokens, Binance lists 415. The gap between "existing" and "having substantial liquidity" is vast.
What I want to say finally
The current number of existing tokens ranges from 37 million to 120 million, and the exact number is not important. What matters is the distribution pattern.
Over 99.99% of tokens have effectively failed.
Among all the tokens created, about 500 have a market cap exceeding $10 million. 99.37% of PumpFun tokens failed to "graduate". 85% of TGE projects issued in 2025 traded below their initial price.
Data from all sources are highly consistent.
I write this not to scare you away from the crypto market, but to help you understand that this background can make you wiser in allocating capital and attention. Opportunities in the crypto market are still real, but they exist in a few assets and protocols, not in the millions of tokens created to siphon wealth from careless buyers.
I’ve always said that for almost everyone, the best strategy is to dollar-cost average (DCA) into Bitcoin, perhaps along with a very few other tokens, and then wait.
Indeed, there are stages when altcoins can significantly outperform, but the vast majority will lose money in the process of sifting through the haystack, so it’s better to stick with mainstream assets.
I hope this data can support this position and perhaps help some reader realize that the treasure hunt is not suitable for them.
One last point: even if you do want to find those gems, those needles, this data should at least reinforce the understanding that you should typically only use a smaller proportion of your portfolio for such pursuits, while putting the majority (over 80%) of your funds into assets like BTC, ETH, SOL, and HYPE. If you’re a bit bolder like me, you might also consider adding ZEC and TAO.