Revoked Authorities and 24-Hour Outcomes: A Cohort Study

Correction note (September 2, 2026): We revised the article's wording and methodology disclosure to describe this as an observational, non-random cohort analysis rather than a causal or universal claim. The reported cohort figures are unchanged.

What this cohort can—and cannot—show

Revoked mint and freeze authorities are specific contract facts: at analysis time, the identified authorities could not use those controls. They are not a complete measure of market, holder, creator or liquidity risk.

This article describes outcomes in 3,744 Solana tokens submitted by users or discovered by MemeAssist from July 22 through August 24, 2026. This was a convenience cohort, not a random sample of Solana tokens. All included records had analysis-time signals, no mint-authority flag, no freeze-authority flag, no dangerous-extension flag, an assigned composite rug-risk level and a resolved 24-hour outcome. The comparison asks whether observed outcomes differed across rating groups inside this exact cohort. It does not estimate the causal effect of revoking authorities or of receiving a rating.

Observed outcomes by composite rating

All four groups met the same contract-flag inclusion rule, while their observed death rates ranged from 3.6% to 48.7%. The pattern is non-monotonic: the moderate and high groups had lower observed death rates than the low group, while the extreme group had the highest rate. That pattern should not be simplified into a universal ranking or a claim that one input explains the differences.

The rating is composite and incorporates liquidity, holder distribution, creator behaviour, wallet activity, trading patterns and historical indicators. Group differences are therefore partly constructed from multiple correlated inputs. Token age, liquidity, creator identity, market regime and repeated analyses of the same token or related tokens may also confound the comparison. We did not isolate the contribution of any one factor.

Returns among tokens classified as survivors

In the low-rated group, 7.7% were classified dead within 24 hours. Among records classified as survivors, the median 24-hour return was −7.1%, and 35.6% had a positive return. The return statistic excludes dead outcomes, so it must not be read as the return of the whole group.

These figures support a limited practical distinction: contract-authority checks and composite risk classifications are not demonstrated buy signals. Price outcomes can vary after a token passes contract checks. The data here do not establish why a token rose, fell or died. Timing, momentum and liquidity require separate evaluation, as discussed in should I buy this token?

How to interpret clean-contract and behavioural signals

A token can lack authority flags while still having concentrated ownership, connected wallets, creator-history concerns, shallow liquidity or unusual trading. Those features are relevant to due diligence, but this grouped analysis does not identify which feature accounted for an outcome or establish universal holder cutoffs.

• Interpret revoked authorities narrowly. They address specific authority-based actions, not every way a token can lose liquidity or value.
• Review multiple dimensions. Holder distribution, creator history, wallet links, liquidity and trading conditions can add context, without guaranteeing an outcome.
• Avoid causal interpretations. The composite rating did not cause the observed outcomes, and this analysis does not prove that its components independently predicted them.

Methods, status handling and exclusions

Period and sampling: July 22 through August 24, 2026. Tokens entered the tracker through user submission or system discovery. Selection was not random, so submission and discovery mechanisms may overrepresent particular token types, activity levels or market conditions.

Inclusion: a record needed captured analysis-time signals; no mint-authority, freeze-authority or dangerous-extension flag; an analysis-time composite rug-risk rating of low, moderate, high or extreme; and a resolved 24-hour status. The included counts were low 2,022, moderate 569, high 337 and extreme 816, totaling 3,744.

Outcome handling: resolved records were those marked done (treated as survivors) or dead. Death rate was dead / (done + dead) within each rating group. Median 24-hour returns and the positive-return share were calculated among done survivors only. Records marked no_data, pending or otherwise unresolved were excluded, as were records missing any required inclusion field.

Missingness: counts for excluded no_data, pending, otherwise unresolved and missing-field records were not retained with this aggregate extract, so we cannot report them. This matters because missing outcomes may be informative rather than random—for example, illiquid, newly observed or difficult-to-quote tokens may be more likely to lack a resolved outcome. Complete-case rates could therefore differ from rates in all submitted or discovered tokens.

Verification and reproducibility: this article reports an internal aggregate extract. No public row-level export accompanies it, and the underlying token records and outcome classifications were not manually verified against independent sources for this article. Readers cannot independently reproduce the table from the published materials.

Uncertainty and limits: we did not calculate confidence intervals or conduct significance tests. The percentages and medians are sample estimates subject to sampling variability, classification error and measurement error; differences, especially between groups, should not be assumed statistically meaningful. The composite rating's construction, token age, liquidity, creator, market regime and repeated analyses may confound results. The cohort does not establish causation and should not be generalized to all Solana tokens, other periods or other discovery channels. See how the engine builds its ratings for the signal categories involved.

Use contract checks as one part of review

Paste a Solana mint into MemeAssist to review contract checks alongside the Overall Health Score, composite Rug Risk Rating, AI Verdict and risk breakdown. These outputs organize available signals; they do not guarantee safety, survival or returns.