Does a higher Belphor score mean a better outcome?
Updated 3 October 2026 · by AtroX · Belphor Research · the data behind this guide
Does a high Belphor score mean a coin will do well?
Not reliably. In Belphor's ignition record, signals scoring under 65 were down 80% or more at their latest reading far more often than signals at 65 and above, but above 65 a higher score did not mean a better outcome. The association is retrospective and is not a probability of making money.
Belphor score: a number from 0 to 100 that Belphor gives each signal when it detects it. The live ignition table groups signals by the v11 score formula, and since 2 September 2026 the scanner emits an ignition only at a v11 score of 65 or above.
From the live data page, updated 2026-10-11 (12 Aug 2026 to 11 Oct 2026), the share down 80% or more by v11 score: under 65, 81% (n 6,275); 65 to 69, 15% (n 1,186); 70 to 79, 14% (n 1,100); 80 and up, 23% (n 145). The tables further down are this guide's dated snapshot of 3 October 2026.
Read next: the table by Belphor v11 ignition score, on the live data page.
Partly, as an association in a retrospective record. As of 3 October 2026, the ignition score's line at 65 separates the share of large declines: 79% of ignitions scoring under 65 were down 80% or more at their latest reading (n 7,322) against 15% of those at 65 and above (n 2,188). Above that line a higher score did not mean a better outcome: the 65 to 69 and 70 to 79 buckets reached +50% at 20% and 19%, and the 80-and-up bucket was more often down 80% or more (24%, n 136).
- In this retrospective ignition record, the 65 cutoff separates the observed frequency of large declines, but this does not validate a prospective filter: 79% against 15% were down 80% or more; above it, the buckets do not rank on reaching +50%.
- The top ignition bucket, 80 and up, swung both ways on 136 signals: more often +50% (24%) and more often down 80% or more (24%) than 65 to 79.
- The takeover score did not rank either: 40 to 59 reached +50% more often (55%, n 248) than 60 to 79 (47%, n 57).
- The score is published so it can be judged; this page is that judgment on the record as it stood on 3 October 2026.
Two scores, two tables
Belphor scores every signal 0 to 100 at the moment of detection and publishes the score beside the outcome, so the score can be judged rather than taken on trust. Two families have enough history to judge: momentum ignitions, scored by the v11 formula, and pump.fun community takeovers, scored at the takeover. Since 2 September 2026 the ignition scanner emits a signal only at 65 or above; signals recorded before that date are re-scored with the v11 formula from their stored metrics, so the table can show what the line keeps out.
Ignitions by v11 score
| v11 score | Signals | Down 80% or more | Reached +50% | Reached +200% | Above entry now | Median now |
|---|---|---|---|---|---|---|
| under 65 (not a signal since v11) | 7,322 | 79% | 19% | 6.8% | 14% | 0.00x |
| 65 to 69 | 1,046 | 15% | 20% | 4.3% | 31% | 0.84x |
| 70 to 79 | 1,006 | 14% | 19% | 5.0% | 31% | 0.84x |
| 80 and up | 136 | 24% | 24% | 3.7% | 25% | 0.65x |
| 65 and up together | 2,188 | 15% | 20% | 4.6% | 30% | 0.84x |
Fetched from the public JSON on 3 October 2026. The live table refreshes daily; this article is a dated snapshot of it. How each column is measured: how Belphor measures outcomes.
In this retrospective record, the line at 65 is associated with one clear difference. Below it, 79% of signals were down 80% or more at their latest reading (n 7,322); at 65 and above, 15% (n 2,188). Signals above the line also ended above their detection price twice as often (30% against 14%).
The line does not pick runners. The record does not separate the under-65 row from the 65 to 69 or 70 to 79 rows on reaching +50% (19%, 20%, 19%).
Where the buckets overlap
Above the line, a higher score did not mean a steadier outcome. The record does not separate the 65 to 69 and 70 to 79 buckets on any column: down 80% or more at 15% and 14%, reaching +50% at 20% and 19%. The 80-and-up bucket swung both ways: it reached +50% more often (24%) and was down 80% or more more often (24%) than either, and ended above its detection price less often (25% against 31%). That bucket holds 136 signals, past the 30 this record needs to compare rows, and well short of the thousand or more behind each bucket below it.
Takeovers by score
| Score | Takeovers | Reached +30% | Reached +50% | Reached +200% | With a later reading | Down 80% or more | Median peak |
|---|---|---|---|---|---|---|---|
| under 40 | 47 | 57% | 49% | 19% | 1 | too few readings | 1.48x |
| 40 to 59 | 248 | 68% | 55% | 20% | 17 | 35% | 1.59x |
| 60 to 79 | 57 | 63% | 47% | 16% | 4 | too few readings | 1.46x |
Fetched from the public JSON on 3 October 2026. The live table refreshes daily; this article is a dated snapshot of it. How each column is measured: how Belphor measures outcomes.
The takeover score did not rank either. Takeovers scoring 40 to 59 reached +50% more often (55%, n 248) than those scoring 60 to 79 (47%, n 57) or under 40 (49%, n 47). The share down 80% or more cannot be compared across these rows: two of the three have too few later readings, and the third rests on 17, under the 30 this record needs to compare rows.
Why a score might separate collapse without ranking
One reading of the tables, offered as reasoning rather than measurement: a score that adds up the measurable signs of a fragile coin is good at saying which coins are fragile. Which of the sturdier coins will run is a different question, decided mostly by attention that arrives after the signal, and the inputs that answer the first question do not answer the second. That reading fits the ignition table; the table does not prove it.
How to use the score
The retrospective record associates scores below 65 with more large declines. It does not isolate the score from pool age, market period or observation age, and does not establish a prospective trading edge. Nothing here supports a bigger position on a higher number either, which is why sizing stays fixed in position sizing and why the math of a rule set is judged as a whole in expectancy in trading.
How often do Solana memecoin momentum signals survive? The ignition score buckets, recomputed daily. The takeover score buckets are on how often pump.fun community takeovers succeed. Both carry JSON and CSV twins.
Frequently asked
Does a higher Belphor score mean a better outcome?
Partly, as an association in a retrospective record. As of 3 October 2026, ignitions scoring 65 or above were down 80% or more far less often than those under 65 (15% against 79%), but above 65 the buckets did not line up in score order, and takeovers scoring 60 to 79 reached +50% less often than those scoring 40 to 59.
What does the 65 line mean for ignitions?
Since 2 September 2026 the scanner emits an ignition only at a v11 score of 65 or above. Older signals are re-scored with the same formula, so the under-65 row shows what the scanner no longer sends: 79% of those 7,322 signals were down 80% or more at their latest reading.
Should you size a trade by the score?
Not on this record. Between 65 and 79 the buckets did not separate on reaching +50%, and the 80-and-up bucket was down 80% or more more often (24%) than 65 to 79. The retrospective tables associate low scores with more large declines; they do not validate the score as a gate or as a measure of conviction.
Belphor records pump.fun takeovers on-chain, scores each signal 0 to 100 and publishes measured outcomes with their sample sizes, windows and missing readings. Reflex Bot is coming soon and priced separately. The tables in this guide come from that record.
Explore Belphor →