We Tracked 1,464 Scored Solana Tokens for 24 Hours: The Results

By the MemeAssist Research Desk · Published 2026-08-05 · Updated 2026-08-05 · 7 min read

We scored 1,464 Solana tokens with our screening engine and recorded what their price actually did over the next 24 hours. Tokens with a health score of 60+ averaged +5.2%; tokens under 40 averaged −5.0% and were 5× more likely to lose half their value. The score didn't predict winners — it reliably sorted survival odds.

Why we ran this 24-hour health score study

Every token checker claims its score "works". Almost none of them show you what happened to the tokens they scored. MemeAssist automatically re-checks every token its screening engine scores at 1, 6, 12 and 24 hours after analysis and writes the outcome to a database — the wins, the losses, and the tokens that went to zero. This is the first study published from that outcome layer: 1,464 tokens scored and then tracked for a full 24 hours between August 3–5, 2026.

Nothing here is back-fitted. The score was computed first, in public, at analysis time; the outcome arrived on its own 24 hours later. We publish the misses along with the hits because that's the entire point of tracking outcomes — see how our engine builds the health score if you want the mechanics.

The headline result: the score sorts survival odds

We bucketed the 1,464 outcomes by the health score the token received at analysis time:

Health score at analysis	Tokens	Avg 24h return	Lost ≥50% of value	Finished up	Gained ≥50%
60+ (strong)	261	+5.2%	1.5%	45.6%	3.8%
40–59 (middling)	939	−3.5%	3.6%	36.2%	0.9%
Under 40 (weak)	264	−5.0%	7.6%	45.1%	1.1%

Two things stand out. First, the catastrophic-loss rate: a token that scored under 40 was five times more likely to lose half its value within a day than a token that scored 60+. That is exactly what a risk score is supposed to do — it can't tell you which token moons, but it tells you which ones are structurally more likely to collapse.

Second, the upside sorted too, which we did not take for granted going in: strong scorers produced 50%+ gainers at roughly 3.5× the rate of weak scorers. Healthy structure — distributed holders, honest volume, a creator without a rap sheet — doesn't just avoid disasters; it's also where the sustained runs happened.

What a "typical" scored token did in 24 hours

Zooming out from the score bands, the base rates across all 1,464 tokens are a useful reality check for anyone trading this market:

The median token drifted −2.2% — most tokens simply bleed slowly.
~40% finished up on the day at all.
58 tokens (4.0%) lost more than half their value; the worst fell essentially to zero (−100%).
Only 13 tokens (0.9%) doubled; the best did +1,373%.

That last pair of numbers is the memecoin market in miniature: the payoff distribution is a long, thin right tail attached to a fat left one. Position sizing for the left tail — not chasing the right one — is what keeps accounts alive, a theme we measured from the trading side in what are the chances a memecoin rugs?

An honest caveat: what we can't conclude yet

Where the cohort is big enough to trust, we published it. Where it isn't, we're saying so:

Interesting bands are still small. Only 29 tokens carried the very highest rug-risk flags in this window — too few to publish a reliable rate (our internal rule is a minimum of 30 outcomes per claim). Those cuts will come in a follow-up study as the tracker accumulates data every day.
24 hours is one horizon. A token that survived day one can still die on day three. We also record 1h, 6h and 12h outcomes and will extend the window in future studies.
Selection effect. These are tokens that reached our screening engine at all — the very worst launches often die before anyone would bother scoring them. The true base rate for random fresh launches is uglier than anything in this table.

Methodology

Cohort: every token scored by the MemeAssist screening engine whose 24-hour outcome check completed between August 3–5, 2026 with a valid 24-hour price (n = 1,464). The health score is computed at analysis time from live on-chain data — holder concentration, bundle detection, wash-trade ratios, liquidity, creator history. Outcomes are the token's price 24 hours after analysis versus the price at analysis, from the same market-data pipeline that feeds our reports. No token was excluded for performing badly — the −100% outcomes are in the table. One honest limitation: 9 tokens (0.6% of the cohort) finished tracking without a usable 24-hour price — the market stopped quoting them — and are excluded rather than assumed dead. Since vanishing from the market is rarely good news, our loss rates are, if anything, slightly understated. The outcome tracker runs continuously, so every future study draws from a strictly larger cohort.

Run the health score on your own Solana token

The score in this study is the same one you get when you paste any Solana mint address into MemeAssist — an Overall Health Score, a Rug Risk Rating, an AI Verdict in plain English, and a Detailed Risk Breakdown. And your token joins the same outcome tracker, because we keep grading our own homework.

Frequently asked questions

Does a high MemeAssist health score guarantee a token will go up?

No. In our 1,464-token study, strong scorers (60+) averaged +5.2% over 24 hours and 45.6% finished up — meaning most still didn't moon and many fell. The score's proven job is sorting survival odds: weak scorers were 5× more likely to lose half their value.

How were the 24-hour outcomes measured?

Every token our screening engine scores is automatically re-checked 1, 6, 12 and 24 hours later, and the price change is written to a database before anyone knows the result. This study reads those records directly — nothing is reconstructed after the fact, and losing tokens are included. Only 9 tokens (0.6%) whose market stopped quoting a price were excluded.

What happened to the average scored Solana token in 24 hours?

The median token in our 1,464-token cohort drifted −2.2%. About 40% finished up, 4% lost more than half their value, and only 0.9% doubled. Slow bleed is the default outcome; big wins are rare and big losses are several times more common.

Will this study be updated with more data?

Yes. The outcome tracker records new results continuously, so follow-up studies will cover larger cohorts, longer horizons than 24 hours, and the rug-risk cuts that were still too small (under 30 outcomes) to publish reliably in this edition.