calibration
window: 24 h48 h168 h· machine readable: /api/calibration · how all of it is computed: methodology
71 of 399 setups were already expired when they were written down: the engine finds a cluster after the fact, and lifetime counts from the first buy. Correct as a record, useless as an offer — such a setup never reaches the executor at any score. Detection takes 30 min, at worst 120 min. States: expired: 299, rejected: 96, trade setup: 3, watch: 1.
what holds setups back
gate table, holds alone
Holding back means the condition was the only one not passed — one failed among five holds nothing. The last column decides what to do: measured and did not fit is about the threshold, not measured is about the data.
| condition | passed | failed | not measured | holds alone | of those, measured |
|---|---|---|---|---|---|
| market cap | 309 | 90 | 0 | 11 | 11 · not measured 0 |
| entry not late | 331 | 68 | 0 | 10 | 10 · not measured 0 |
| taxes | 365 | 1 | 33 | 5 | 0 · not measured 5 |
| 24h volume | 358 | 41 | 0 | 2 | 2 · not measured 0 |
| second source | 392 | 7 | 0 | 2 | 2 · not measured 0 |
| risk level | 352 | 3 | 44 | 1 | 0 · not measured 1 |
| liquidity | 342 | 53 | 4 | 0 | 0 · not measured 0 |
| audit | 382 | 1 | 16 | 0 | 0 · not measured 0 |
| consensus | 399 | 0 | 0 | 0 | 0 · not measured 0 |
| no honeypot | 382 | 0 | 17 | 0 | 0 · not measured 0 |
| market cap: why | $# against a minimum of $# × 90 | ||||
| entry not late: why | price moved #% against a limit of #% × 68 | ||||
| taxes: why | the audit did not name the taxes × 33buy #%, sell #% against a limit of #% × 1 | ||||
| 24h volume: why | $# against a minimum of $# × 41 | ||||
| second source: why | #% of wallets left the Binance signal against a limit of #% × 3Binance closed its own signal (exitRate), the move already happened × 2Binance closed its own signal (outDecline), the move already happened × 1tags against a buy: Smart Money Remove Holdings, Insider Wash Trading, Wash Trading × 1 | ||||
| risk level: why | the risk level did not arrive × 44# against a limit of # × 3 | ||||
| liquidity: why | $# against the $# needed (# positions of $#) × 53Token Info did not answer, pool depth unknown × 4 | ||||
| audit: why | the audit answered but gave no result × 16mechanisms found: #, Contract Not Renounced, Newly Created Token × 1 | ||||
| no honeypot: why | upstream did not declare the safety check ready, and there is no audit either × 17 | ||||
what the price did after the setup
three horizons, medians
Not profit and not trades: this is the token price movement from the moment the setup appeared — the price we could have entered at, not the price the wallet paid. A move counts as a win above 2.0% of costs. Three horizons because one always flatters: fifteen minutes says whether we were in time, four hours whether the token survived its own pump.
Read measured together with the median. The longer the horizon, the fewer tokens live to a second price point — a dead token has no price — so the four-hour median is computed over survivors and flatters. We cannot remove that bias, only name it.
| horizon 15m | setups | measured | median | above cost | best | worst |
|---|---|---|---|---|---|---|
| trade setup | 259 | 259 | -0.1% | 7% | +23.1% | -39.1% |
| rejected | 96 | 95 | -6.5% | 24% | +436.6% | -82.6% |
| watch | 44 | 44 | -1.9% | 20% | +220.4% | -88.5% |
| horizon 1h | setups | measured | median | above cost | best | worst |
|---|---|---|---|---|---|---|
| trade setup | 259 | 252 | -0.3% | 12% | +44.9% | -42.1% |
| rejected | 96 | 96 | -17.0% | 13% | +538.5% | -84.8% |
| watch | 44 | 43 | -7.1% | 23% | +188.7% | -89.7% |
| horizon 4h | setups | measured | median | above cost | best | worst |
|---|---|---|---|---|---|---|
| trade setup | 259 | 217 | -1.2% | 21% | +91.4% | -69.5% |
| rejected | 96 | 13 | -0.0% | 31% | +1235.6% | -86.1% |
| watch | 44 | 10 | -1.6% | 20% | +14.2% | -69.7% |
does the score rank
score in five parts
Setups with a measured outcome, split by score into five equal parts. If the formula works, the median grows from top to bottom — the main question to ask of a score.
| score | setups | median at one hour | above cost |
|---|---|---|---|
| 28 to 50 | 78 | -17.0% | 13% |
| 51 to 61 | 78 | -0.5% | 21% |
| 61 to 70 | 78 | -0.1% | 17% |
| 70 to 73 | 78 | +0.0% | 10% |
| 73 to 77 | 79 | -2.4% | 5% |
wallets in clusters
addresses and their clusters
One cluster has one outcome and several participants, so crediting it to each counts one event many times. The modest question here: are there addresses that turn up only in bad clusters. A reason to look, not to retire.
| address | score | clusters | with outcome | median at one hour |
|---|---|---|---|---|
| 0xc7f6…e3ad | 73.1 | 104 | 102 | -1.5% |
| 0x4873…4828 | 68.5 | 96 | 94 | -1.1% |
| 0x9df2…e831 | 72.3 | 94 | 94 | -1.4% |
| 0x9a2a…8def | 66.8 | 89 | 89 | -1.8% |
| 0xc8db…9361 | 67.9 | 89 | 89 | -1.1% |
| 0x95c4…a260 | 64.9 | 87 | 87 | -1.8% |
| 0x2d97…75c5 | 74.3 | 80 | 80 | -1.1% |
| 0x2f7a…bfa6 | 66.9 | 78 | 76 | -2.2% |
| 0x7590…7538 | 70.2 | 78 | 78 | -1.1% |
| 0x207b…3be2 | 68.1 | 77 | 77 | -1.9% |
| 0xdd6c…2a34 | 69.8 | 74 | 74 | -2.0% |
| 0x1046…2a25 | 73.0 | 73 | 70 | -1.0% |
| 0x14b9…97d3 | 72.3 | 72 | 72 | -2.0% |
| 0x487c…241d | 69.1 | 71 | 70 | -1.1% |
| 0x962c…fa25 | 69.2 | 71 | 71 | -0.4% |
| 0x4f98…2348 | 73.4 | 70 | 70 | -1.6% |
| 0xf092…61de | 68.8 | 70 | 70 | -1.1% |
| 0x183d…f1c9 | 72.9 | 69 | 69 | -1.1% |
| 0xe92d…79ee | 70.0 | 69 | 67 | -0.3% |
| 0x2285…0683 | 65.6 | 68 | 68 | -1.6% |
universe
subscription, bench, silence
A subscription socket holds a hundred addresses. An address that gave no trades takes a slot for nothing — but dropping it helps only if there is a replacement. Bench at zero means a cleanup would shrink the universe, not improve it.
silent: 0x7100…ad6e
what follows from this
conclusions and report time
- "market cap" holds 11 setups, and in 11 cases the value was measured and did not fit — here the threshold decides.
- "entry not late" holds 10 setups, and in 10 cases the value was measured and did not fit — here the threshold decides.
- "taxes" holds 5 setups, and in 5 cases the value was NOT MEASURED. That is about the data, not the threshold: lowering it would let the unchecked through.
- Most setups (97 of 399) are held by several conditions at once: tuning any single threshold will not give one new trade.
- 71 of 399 setups were born already expired: found later than their lifetime ran out, so they reach the executor at no score.
- 1 of 150 subscription addresses gave no trade at all over the window. There is a replacement: 1003 addresses beyond the subscription.
Report built 2026-10-16 20:04:06 UTC. No number here was picked by hand — this page and npm run calibrate call the same function. What it deliberately does not do is name the number you should set: that is a human choice, and advice fitted to one week is not a calculation.