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)
90.6%
Mean accuracy (100−APE)
88.3%
Within ±5%
27%
Within ±10%
52%
Within ±20%
86%
Bias
-1.0%
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 | 458 | 91.0% | 53% | -1.7% |
| middle | 6 | 83.3% | 33% | -15.8% |
| grail | 153 | 89.9% | 50% | +1.6% |
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 |
|---|---|---|---|---|---|---|
| #2680 | grail · high | Sep 12, 2026 | 0.501Ξ | 0.506Ξ | +1.0% | 99.0% |
| #3248 | floor · high | Sep 9, 2026 | 0.499Ξ | 0.485Ξ | -2.8% | 97.2% |
| #4609 | floor · high | Sep 8, 2026 | 0.465Ξ | 0.479Ξ | +3.0% | 97.0% |
| #3009 | floor · high | Sep 7, 2026 | 0.500Ξ | 0.439Ξ | -12.2% | 87.8% |
| #4265 | floor · high | Sep 7, 2026 | 0.529Ξ | 0.448Ξ | -15.3% | 84.7% |
| #844 | grail · high | Sep 7, 2026 | 0.500Ξ | 0.472Ξ | -5.6% | 94.4% |
| #4926 | grail · high | Sep 7, 2026 | 0.524Ξ | 0.472Ξ | -9.9% | 90.1% |
| #4926 | grail · high | Sep 7, 2026 | 0.462Ξ | 0.487Ξ | +5.4% | 94.6% |
| #844 | grail · high | Sep 7, 2026 | 0.439Ξ | 0.520Ξ | +18.5% | 81.6% |
| #3907 | grail · high | Sep 6, 2026 | 0.527Ξ | 0.459Ξ | -12.9% | 87.1% |
| #3907 | grail · high | Sep 6, 2026 | 0.439Ξ | 0.506Ξ | +15.3% | 84.7% |
| #3699 | floor · high | Sep 6, 2026 | 0.445Ξ | 0.594Ξ | +33.5% | 66.5% |
| #3874 | floor · high | Sep 5, 2026 | 0.520Ξ | 0.471Ξ | -9.4% | 90.6% |
| #4430 | grail · high | Sep 5, 2026 | 0.520Ξ | 0.488Ξ | -6.2% | 93.9% |
| #4834 | floor · high | Sep 5, 2026 | 0.530Ξ | 0.430Ξ | -18.9% | 81.1% |
| #3700 | middle · low | Sep 4, 2026 | 0.991Ξ | 1.017Ξ | +2.6% | 97.4% |
| #3009 | floor · high | Sep 4, 2026 | 0.427Ξ | 0.615Ξ | +44.0% | 56.0% |
| #4016 | grail · high | Sep 4, 2026 | 0.428Ξ | 0.670Ξ | +56.5% | 43.5% |
| #4834 | floor · high | Sep 4, 2026 | 0.430Ξ | 0.627Ξ | +45.8% | 54.2% |
| #4899 | floor · high | Sep 3, 2026 | 0.601Ξ | 0.652Ξ | +8.5% | 91.5% |
| #2756 | floor · high | Sep 2, 2026 | 0.643Ξ | 0.619Ξ | -3.7% | 96.3% |
| #633 | floor · high | Sep 1, 2026 | 1.020Ξ | 0.748Ξ | -26.7% | 73.3% |
| #2371 | floor · high | Sep 1, 2026 | 0.612Ξ | 0.586Ξ | -4.3% | 95.7% |
| #4265 | floor · high | Aug 30, 2026 | 0.540Ξ | 0.682Ξ | +26.3% | 73.7% |
| #2136 | grail · high | Aug 30, 2026 | 0.700Ξ | 0.771Ξ | +10.1% | 89.9% |
| #187 | grail · high | Aug 30, 2026 | 0.636Ξ | 0.595Ξ | -6.5% | 93.5% |
| #4364 | floor · high | Aug 28, 2026 | 0.560Ξ | 0.622Ξ | +11.1% | 88.9% |
| #4028 | floor · high | Aug 28, 2026 | 0.570Ξ | 0.629Ξ | +10.4% | 89.7% |
| #3804 | floor · high | Aug 28, 2026 | 0.575Ξ | 0.631Ξ | +9.7% | 90.3% |
| #2371 | floor · high | Aug 28, 2026 | 0.575Ξ | 0.634Ξ | +10.3% | 89.7% |
| #1029 | floor · high | Aug 28, 2026 | 0.576Ξ | 0.631Ξ | +9.6% | 90.5% |
| #3115 | floor · high | Aug 28, 2026 | 0.687Ξ | 0.596Ξ | -13.2% | 86.8% |
| #3115 | floor · high | Aug 27, 2026 | 0.592Ξ | 0.626Ξ | +5.7% | 94.3% |
| #4734 | floor · high | Aug 26, 2026 | 0.687Ξ | 0.643Ξ | -6.4% | 93.6% |
| #755 | floor · high | Aug 25, 2026 | 0.594Ξ | 0.621Ξ | +4.6% | 95.5% |
| #2905 | floor · high | Aug 25, 2026 | 0.595Ξ | 0.619Ξ | +4.0% | 96.0% |
| #4734 | floor · high | Aug 24, 2026 | 0.613Ξ | 0.700Ξ | +14.2% | 85.8% |
| #2372 | grail · high | Aug 24, 2026 | 0.629Ξ | 0.620Ξ | -1.4% | 98.6% |
| #1029 | floor · high | Aug 24, 2026 | 0.629Ξ | 0.595Ξ | -5.4% | 94.6% |
| #2371 | floor · high | Aug 24, 2026 | 0.629Ξ | 0.632Ξ | +0.5% | 99.5% |
Model cryptodickbutts-1789249235-2026-08-03.2 · evaluated through 9/12/2026. Method details on the methodology page.