How much of a portfolio should be in memecoins?

An amount you can lose in full without anything in your life changing. That is not a hedge or a disclaimer — it is the only variable the arithmetic actually depends on, because total loss is the modal outcome here rather than the tail. No percentage can be given as advice on this page, and the percentages circulating elsewhere are conventions, not findings.

Why a percentage is the wrong output

Portfolio allocation frameworks — mean-variance optimisation, Kelly sizing, risk parity, the standard "satellite" sleeve — all take the same inputs: an expected return, a variance, and a correlation. Feed them a memecoin and every input is either unknown or undefined.

Expected return requires a distribution you can estimate. The best-documented dataset in the category says the distribution is not one these models can digest: across 18.67 million pump.fun tokens launched between 14 January 2024 and 18 June 2026, 68.67% recorded their last trade on the day they were created, roughly 80.37% were dead within two days, and only 4.55% survived past 90 days (CoinGecko, updated 23 June 2026, which notes its 90-day figure is understated because it tracks bonding-curve trades only). That is not a bell curve with fat tails. It is a spike at total loss with a thin smear of survivors.

The payoff side is no better behaved. Of roughly 11.9 million tokens launched on pump.fun since January 2024, 18 have ever exceeded a $10M market capitalisation and 96 have exceeded $1M (10 June 2026). Any model that needs the mean of that distribution is estimating the mean of a set dominated by a handful of observations, which is a statistical way of saying it is estimating noise.

Kelly sizing fails for a specific and instructive reason: it requires a win probability and payoff odds you can estimate before the bet. For a single memecoin you have neither. You are not making repeated draws from a stable urn — the composition of the urn changes weekly with which venue is dominant, which chain has attention, and how much bot capital is competing with you.

So the honest position is not that these models say to allocate a small percentage. It is that they do not apply, and using one anyway produces a number with a false provenance. The underlying reasons the distribution looks like this — bonding-curve mechanics, thin liquidity, attention decay — are set out in why most memecoins go to zero.

Size to total loss instead

If the models are out, what is left is a solvency question rather than an optimisation question. It has one form: if this goes to zero tomorrow, what in my life changes?

If the answer includes rent, a mortgage payment, tuition, debt service, an emergency fund, or a bill with a date on it, that money has no allocation here at all. Not a small one. The reason is mechanical rather than moral: money with a job attached has a deadline, and a deadline forces you to sell at whatever the pool will pay on the day you need it, which is the single worst way to interact with an asset whose liquidity is thin and whose price is set by attention.

If the answer is "nothing changes", you have found your ceiling. The number you write down is the largest amount for which that sentence stays true, and it is denominated in currency, not in percent of net worth. Two people with identical net worth and different obligations have different correct answers, which is exactly why a headline percentage cannot be right for both.

This framing is also the one that survives being wrong. A percentage-of-portfolio rule quietly scales your memecoin exposure up as the rest of your portfolio grows, including when the growth is unrealised and the rest of your holdings are correlated with the same risk appetite that made you buy memecoins in the first place.

Diversification inside memecoins buys less than you think

The standard fix for a high-variance asset is to hold more of them. It works less well here, for a reason worth stating plainly.

Diversification reduces risk when holdings fail independently. Memecoins largely do not. They share a dependence on the same attention cycle, the same venues, the same handful of terminals for execution, and the same underlying gas asset — SOL or ETH — whose price moves against you in exactly the conditions where memecoin bids disappear. Twenty positions in twenty launchpad tokens is closer to one leveraged bet on retail risk appetite than to twenty independent bets.

The measured market backs this up. CoinGecko's State of Memecoins work put total memecoin market capitalisation at a peak of $150.6bn in December 2024, against $88.0bn in October 2021, falling to $47.2bn in November 2025 — roughly a 69% drawdown across the whole category at once. Within that, DOGE alone accounted for 47.3% of memecoin market cap and dog-themed tokens excluding DOGE for another 39.5%. Two things follow. The category moves together, and "the memecoin market" as measured is mostly one very old token — so a portfolio of new launchpad tokens is not exposure to that index in any meaningful sense. Launchpad memecoins were 1.5% of memecoin market cap in July 2024 and peaked at 20.5% in January 2025.

What diversification does still buy you is protection against the idiosyncratic failure of any single token: the rug, the honeypot, the dev who sells. That is a real benefit, and it is smaller than the benefit diversification usually provides.

Fixed-fractional sizing: the part that matters more than the headline number

Once the sleeve is set, the useful decision is how much of it goes into any one position — and here there is an actual mechanic worth adopting.

Fixed-fractional sizing means each position is a constant fraction of the sleeve's current value, recalculated as it changes, rather than a constant absolute amount. The property that matters is automatic: after losses, the fraction is applied to a smaller sleeve, so positions shrink on their own. You cannot ladder yourself into ruin by repeating the same absolute bet size against a declining balance, which is the specific failure mode that turns a bad month into a wiped account.

The mirror-image discipline is on the upside. A position that outruns the rest of the sleeve is no longer sized the way you sized it; it is now a concentrated bet you never consciously placed. Deciding in advance what happens when that occurs is the same problem as deciding when to sell, which we treat separately in when to take profit on a memecoin.

One constraint overrides both: pool depth. A position sized as a percentage of your sleeve can still be too large to exit, because what you can realise depends on your size relative to the pool reserve, not relative to your own balance — the arithmetic is in how memecoin liquidity pools actually work. A position you cannot sell is not sized correctly no matter what the spreadsheet says.

What this page does not tell you

It does not give you a number, and that is deliberate rather than coy. It does not know your income stability, your obligations, your jurisdiction's tax treatment, or whether the rest of your holdings are already correlated with crypto risk appetite.

It does not account for tax drag on an active strategy. In several jurisdictions, a year of frequent trading that ends net down can still produce a tax bill, because winning disposals are taxed and losing ones may not be deductible against them — the shape of that problem is in memecoin taxes explained, and it is general information rather than advice.

The correlation claim above is reasoning from observed category-wide drawdowns and shared infrastructure, not from a published correlation matrix for launchpad tokens. We are not aware of one, and treat it as an inference rather than a measurement.

And none of this addresses whether you should hold any at all. That question is closer to the one we take up in is trading memecoins gambling, and the answer is yours to reach.

Frequently asked questions

Is 5% of a portfolio a reasonable memecoin allocation?

Figures in the low single digits to around ten per cent circulate widely, and none of them derives from anything specific to this asset class — they are inherited from conventional alternative-asset sleeves, where the underlying assets do not go to zero at the observed rate. We cannot tell you whether a percentage is reasonable for you. The question that has an answer is what amount you can lose entirely without anything changing.

Does diversifying across many memecoins reduce the risk?

It reduces the risk of a single rug or honeypot ending your sleeve, which is real. It does much less against the category risk, because memecoins share an attention cycle, a small set of venues, and the same underlying gas asset. Total memecoin market cap fell from a peak of $150.6bn in December 2024 to $47.2bn in November 2025 — everything drew down together.

Should I size positions in dollars or as a percentage?

As a fraction of the sleeve, recalculated as the sleeve changes, with an absolute cap set by pool depth. Fixed-fractional sizing shrinks your positions automatically after losses, which is the failure mode that does the most damage. But a position too large relative to a pool's reserve cannot be exited near the quoted price regardless of how it looks against your balance.

What if I only trade with money I have already made?

Realised gains from earlier trades are your money, and treating them as a separate pot of house money is a framing rather than a fact — losing them costs exactly what losing any other money costs. The test does not change: what happens if this goes to zero tomorrow.

---

Whatever the size, check what is actually committed on-chain

Position sizing is a defence against loss you cannot avoid; verifying the token is a defence against loss you can. The one commitment a buyer can check without trusting anyone is whether the liquidity is locked, and a fixed-term LP lock from Team Finance — built by TrustSwap, which also builds Meme Central — is visible as a verified badge on that token's page in the Meme Central launch feed. It covers Ethereum, Robinhood Chain, Polygon, Base and BNB. It does not stop a dev selling their own allocation, and it says nothing about whether a token is worth holding at any size.

---

Nothing here is financial, legal or tax advice. Memecoins are extremely high-risk: most lose most of their value, and the majority of tokens launched never reach a decentralised exchange at all. Never spend money you cannot afford to lose entirely. Meme Central does not recommend any specific token. Data described as Meme Central's own reflects tokens indexed by Meme Central and is not whole-market data.