Memecoin and new token risk radar
Every trending and new DEX pair gets an automated risk estimate, the checks behind it, and a JEV note.
How to read this
The label is the predicted chance that liquidity collapses or the token vanishes within an hour. The page shows a chance band, never an exact number.
- Low risk: under 10 percent. It does not mean safe.
- Elevated risk: 10 to 25 percent.
- High risk: 25 to 50 percent.
- Critical risk: 50 percent or more.
- Not checked: no safety source answered. Nothing was measured.
- Unscored: no risk model is adopted. The flags stay, and no label is claimed.
A flag is a checkable fact with a source, not a claim about people. The estimate predicts a collapse of liquidity or a vanished token. It does not detect fraud. New tokens carry risk even with a low label.
How often are our flags confirmed?
Each label is a band of the predicted chance of a collapse within one hour: low risk under 10 percent, elevated risk 10 to 25 percent, high risk 25 to 50 percent, critical risk 50 percent or more. After we list a token, we read its pair again 1 h and 24 h later. A token counts as collapsed when liquidity fell 80% or more, or the price fell 90% or more, and as vanished when the pair leaves the data. The tables count both together. If the labels are useful, a higher label collapses or vanishes more often than the average token, and the 95 percent interval shows that this is not luck. If the labels do not beat the average, these tables show it.
| Label | Tokens | Rate | What it means |
|---|---|---|---|
| Critical risk | 2,750 | 75.3% | Collapsed or vanished more often than average (n=2750). |
| High risk | 3,256 | 16.0% | Collapsed or vanished less often than average (n=3256). |
| Elevated risk | 13,333 | 20.3% | Collapsed or vanished less often than average (n=13333). |
| Low risk | 8,819 | 5.5% | Collapsed or vanished less often than average (n=8819). |
| Not checked | 2,176 | 66.4% | Collapsed or vanished more often than average (n=2176). |
| Unscored | 9,931 | 19.5% | Collapsed or vanished less often than average (n=9931). |
| All listed tokens | 42,340 | 23.0% | The average token: the base rate. |
| Label | Tokens | Rate | What it means |
|---|---|---|---|
| Critical risk | 1,981 | 77.3% | Collapsed or vanished more often than average (n=1981). |
| High risk | 2,708 | 29.8% | Collapsed or vanished less often than average (n=2708). |
| Elevated risk | 9,551 | 24.7% | Collapsed or vanished less often than average (n=9551). |
| Low risk | 4,059 | 19.3% | Collapsed or vanished less often than average (n=4059). |
| Not checked | 1,576 | 71.5% | Collapsed or vanished more often than average (n=1576). |
| Unscored | 9,912 | 44.0% | Collapsed or vanished more often than average (n=9912). |
| All listed tokens | 31,924 | 36.9% | The average token: the base rate. |
21.4% is the share of tokens that disappeared from the data within 24 h. A vanished token is counted in both tables above, because its market is gone.
Which number counts what. The tiles count pairs that the radar scanned or flagged in the window they name. These tables count tokens that have a reading at the horizon (1 hour or 24 hours). The list below counts the pairs on this page.
A rate shows a dash when the group has fewer than 10 tokens. A label whose interval reaches the base rate has no measured skill yet. The table does not prove that a label will keep working.
How the Rug Radar model is tested
- AUC is the chance that the model gives a higher number to a token that collapsed than to one that did not. 0.5 is a coin flip and 1.0 is perfect. The whole interval lies above 0.5, so the model ranks tokens better than a coin flip on this test set.
- Brier score is the mean squared error of the predicted chances. The model is closer to the outcomes than the base rate is.
- Spam cohort. 420 tokens, 4.2 percent of the tokens seen, look like copies made in bulk. They are part of every number on this card.
| Set | Time span | Tokens | Alive | Liquidity collapsed | Vanished |
|---|---|---|---|---|---|
| Train | Oct 01, 23:37 to Oct 02, 11:51 UTC | 5,305 | 4,361 | 801 | 143 |
| Test | Oct 02, 15:55 to Oct 02, 22:01 UTC | 2,601 | 2,005 | 320 | 276 |
The test tokens are all newer than the training tokens. 27 tokens were left out because they had no safety check. Alive means that the pair was there and kept its liquidity.
| Chain | AUC | Tokens |
|---|---|---|
| solana | 0.79 | 2,390 |
| robinhood | 0.67 | 87 |
| Chain only (baseline) | 0.56 | - |
A model that knows only the chain scores 0.56. The fitted model is 0.24 above it.
| Predicted bucket | Predicted average | Observed rate | Tokens | Match |
|---|---|---|---|---|
| under 10 | 2% | 6.6% | 1,104 | The predicted average is outside the interval. |
| 10 to 25 | 13% | 12.8% | 421 | The predicted average is inside the interval. |
| 25 to 50 | 30% | 34.8% | 838 | The predicted average is outside the interval. |
| 50 to 75 | 66% | 57.4% | 61 | The predicted average is inside the interval. |
| 75 or more | 86% | 80.2% | 177 | The predicted average is outside the interval. |
| Rule | Value | Needed | Result |
|---|---|---|---|
| auc pooled lower | 0.782 | 0.65 | Passed |
| auc solana lower | 0.765 | 0.6 | Passed |
| brier margin | 0.038 | 0.005 | Passed |
| test span hours | 6.1 | 6 | Passed |
| test tokens | 2601.0 | 300.0 | Passed |
| train span hours | 12.233 | 24 | Failed |
| train tokens | 5305.0 | 500.0 | Passed |
| When | State | Pooled AUC | Note |
|---|---|---|---|
| Sep 29, 15:35 | unscored | - | the candidate is unscored and may not label: only 35 native training tokens are left (need 150) out of 890 native labelled tokens after the spam cohort and two embargo gaps of 1 hour |
| Sep 29, 16:10 | unscored | - | the candidate is unscored and may not label: only 37 native calibration tokens are left (need 50) out of 1095 native labelled tokens after the spam cohort and two embargo gaps of 1 hour |
| Sep 30, 16:10 | preliminary | 0.86 | the current model is unscored and may not label; the candidate is preliminary |
| Oct 01, 23:02 | preliminary | 0.86 | the candidate Brier score 0.1049 is not worse than the current 0.1187 on the same 2702 recent test tokens (6.3 hours); the candidate is preliminary |
| Oct 02, 23:02 | preliminary | 0.80 | the candidate Brier score 0.1347 is not worse than the current 0.1368 on the same 2601 recent test tokens (6.1 hour); the candidate is preliminary |
The model was tested on tokens that it had not seen in training. The test set is small (2,601 tokens), so the intervals are wide and a result can change with new data. The model predicts that liquidity collapses or that a token vanishes. It does not detect fraud.