Accuracy
How accurate is the CryptoDickbutts model?
Every score comes from the newest chronological hold-out; the target sale is excluded from training, comparables, and its causal floor proxy.
Median accuracy (100−APE)
90.9%
Mean accuracy (100−APE)
88.5%
Within ±5%
26%
Within ±10%
54%
Within ±20%
86%
Bias
-0.4%
Accuracy by market tier
Floor, middle-market, and grail sales are scored separately so aggregate accuracy cannot hide a weak segment
| Tier | Sales | Median accuracy | Within ±10% | Bias |
|---|---|---|---|---|
| floor | 457 | 91.4% | 55% | -0.5% |
| middle | 6 | 76.1% | 17% | -27.8% |
| grail | 140 | 90.4% | 51% | +1.1% |
Weekly median accuracy
Chronological hold-out accuracy by sale week
Error distribution
Signed hold-out error; positive means over-predicted
Scored hold-out sales
60 out-of-sample predictions frozen at evaluation time
| Item | Tier / evidence | Date | Sale | Model | Error | Accuracy |
|---|---|---|---|---|---|---|
| #4353 | floor · high | Aug 3, 2026 | 0.633Ξ | 0.579Ξ | -8.5% | 91.5% |
| #4353 | floor · high | Aug 3, 2026 | 0.553Ξ | 0.646Ξ | +16.8% | 83.2% |
| #1095 | floor · high | Jul 31, 2026 | 0.658Ξ | 0.618Ξ | -6.0% | 94.0% |
| #1988 | grail · high | Jul 29, 2026 | 0.740Ξ | 0.610Ξ | -17.6% | 82.4% |
| #866 | floor · high | Jul 29, 2026 | 0.583Ξ | 0.566Ξ | -2.9% | 97.1% |
| #208 | floor · high | Jul 28, 2026 | 0.586Ξ | 0.590Ξ | +0.7% | 99.3% |
| #1095 | floor · high | Jul 28, 2026 | 0.594Ξ | 0.642Ξ | +8.1% | 91.9% |
| #3199 | grail · high | Jul 27, 2026 | 0.621Ξ | 0.556Ξ | -10.5% | 89.5% |
| #4028 | floor · high | Jul 27, 2026 | 0.651Ξ | 0.545Ξ | -16.3% | 83.7% |
| #1318 | floor · high | Jul 27, 2026 | 0.790Ξ | 0.726Ξ | -8.1% | 91.9% |
| #5012 | floor · high | Jul 27, 2026 | 0.630Ξ | 0.573Ξ | -9.1% | 91.0% |
| #3605 | floor · high | Jul 27, 2026 | 0.630Ξ | 0.591Ξ | -6.2% | 93.8% |
| #1644 | grail · high | Jul 27, 2026 | 0.630Ξ | 0.590Ξ | -6.4% | 93.7% |
| #1421 | floor · high | Jul 27, 2026 | 0.675Ξ | 0.562Ξ | -16.7% | 83.3% |
| #3757 | grail · high | Jul 27, 2026 | 0.650Ξ | 0.583Ξ | -10.3% | 89.7% |
| #1864 | floor · high | Jul 27, 2026 | 0.630Ξ | 0.567Ξ | -10.0% | 90.0% |
| #1612 | floor · high | Jul 26, 2026 | 0.630Ξ | 0.575Ξ | -8.7% | 91.3% |
| #2305 | floor · high | Jul 26, 2026 | 0.630Ξ | 0.582Ξ | -7.6% | 92.4% |
| #1864 | floor · high | Jul 26, 2026 | 0.556Ξ | 0.612Ξ | +10.1% | 89.9% |
| #1252 | floor · high | Jul 25, 2026 | 0.634Ξ | 0.585Ξ | -7.7% | 92.3% |
| #208 | floor · high | Jul 25, 2026 | 0.590Ξ | 0.542Ξ | -8.1% | 91.9% |
| #3613 | grail · high | Jul 25, 2026 | 0.596Ξ | 0.649Ξ | +8.9% | 91.1% |
| #2305 | floor · high | Jul 25, 2026 | 0.576Ξ | 0.606Ξ | +5.2% | 94.8% |
| #4585 | floor · high | Jul 25, 2026 | 0.625Ξ | 0.553Ξ | -11.5% | 88.5% |
| #2371 | floor · high | Jul 25, 2026 | 0.611Ξ | 0.587Ξ | -3.9% | 96.1% |
| #1612 | floor · high | Jul 25, 2026 | 0.565Ξ | 0.614Ξ | +8.7% | 91.3% |
| #3870 | floor · high | Jul 25, 2026 | 0.611Ξ | 0.607Ξ | -0.7% | 99.4% |
| #1617 | floor · high | Jul 25, 2026 | 0.650Ξ | 0.562Ξ | -13.5% | 86.5% |
| #1095 | floor · high | Jul 25, 2026 | 0.640Ξ | 0.577Ξ | -9.8% | 90.2% |
| #4585 | floor · high | Jul 23, 2026 | 0.520Ξ | 0.585Ξ | +12.5% | 87.5% |
| #208 | floor · high | Jul 23, 2026 | 0.520Ξ | 0.597Ξ | +14.8% | 85.2% |
| #4028 | floor · high | Jul 23, 2026 | 0.540Ξ | 0.597Ξ | +10.6% | 89.4% |
| #4028 | floor · high | Jul 22, 2026 | 0.597Ξ | 0.548Ξ | -8.2% | 91.8% |
| #208 | floor · high | Jul 22, 2026 | 0.597Ξ | 0.556Ξ | -6.9% | 93.1% |
| #1421 | floor · high | Jul 22, 2026 | 0.540Ξ | 0.604Ξ | +11.9% | 88.2% |
| #924 | floor · high | Jul 19, 2026 | 0.540Ξ | 0.592Ξ | +9.6% | 90.4% |
| #3757 | grail · high | Jul 19, 2026 | 0.542Ξ | 0.634Ξ | +17.0% | 83.0% |
| #3638 | floor · high | Jul 17, 2026 | 1.800Ξ | 0.697Ξ | -61.3% | 38.7% |
| #2092 | floor · high | Jul 15, 2026 | 0.690Ξ | 0.577Ξ | -16.4% | 83.6% |
| #2832 | floor · high | Jul 15, 2026 | 0.545Ξ | 0.558Ξ | +2.4% | 97.6% |
Model cryptodickbutts-1785785351-2026-08-03.2 · evaluated through 8/3/2026. Method details on the methodology page.