Accuracy
How accurate is the CrypToadz 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)
89.7%
Mean accuracy (100−APE)
84.1%
Within ±5%
29%
Within ±10%
48%
Within ±20%
73%
Bias
+7.9%
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 | 1360 | 90.7% | 52% | +5.7% |
| middle | 90 | 57.4% | 11% | +33.8% |
| grail | 175 | 87.4% | 41% | +12.3% |
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 |
|---|---|---|---|---|---|---|
| #2550 | floor · high | Aug 3, 2026 | 0.290Ξ | 0.368Ξ | +26.9% | 73.1% |
| #2580 | grail · high | Aug 3, 2026 | 0.306Ξ | 0.303Ξ | -0.9% | 99.1% |
| #2241 | floor · high | Aug 3, 2026 | 0.291Ξ | 0.322Ξ | +10.7% | 89.4% |
| #6297 | floor · high | Aug 3, 2026 | 0.295Ξ | 0.359Ξ | +21.7% | 78.3% |
| #2580 | grail · high | Aug 3, 2026 | 0.296Ξ | 0.404Ξ | +36.5% | 63.5% |
| #3689 | floor · high | Aug 3, 2026 | 0.302Ξ | 0.371Ξ | +22.9% | 77.2% |
| #342 | grail · high | Aug 3, 2026 | 0.303Ξ | 0.385Ξ | +27.1% | 72.9% |
| #2380 | floor · high | Aug 3, 2026 | 0.303Ξ | 0.380Ξ | +25.4% | 74.6% |
| #2371 | floor · high | Aug 2, 2026 | 0.307Ξ | 0.320Ξ | +4.2% | 95.8% |
| #5808 | floor · high | Aug 2, 2026 | 0.310Ξ | 0.418Ξ | +34.8% | 65.2% |
| #4319 | floor · high | Aug 2, 2026 | 0.327Ξ | 0.360Ξ | +10.1% | 89.9% |
| #4914 | floor · high | Aug 2, 2026 | 0.327Ξ | 0.377Ξ | +15.3% | 84.7% |
| #6718 | floor · high | Aug 2, 2026 | 0.371Ξ | 0.341Ξ | -8.2% | 91.8% |
| #2241 | floor · high | Aug 2, 2026 | 0.332Ξ | 0.313Ξ | -5.8% | 94.2% |
| #2371 | floor · high | Aug 2, 2026 | 0.320Ξ | 0.313Ξ | -2.2% | 97.8% |
| #3707 | floor · high | Aug 1, 2026 | 0.314Ξ | 0.315Ξ | +0.5% | 99.6% |
| #5312 | grail · medium | Aug 1, 2026 | 0.888Ξ | 1.065Ξ | +19.9% | 80.1% |
| #2520 | floor · high | Aug 1, 2026 | 0.314Ξ | 0.388Ξ | +23.6% | 76.4% |
| #6181 | middle · high | Jul 31, 2026 | 0.323Ξ | 0.602Ξ | +86.4% | 13.6% |
| #2371 | floor · high | Jul 31, 2026 | 0.315Ξ | 0.399Ξ | +26.7% | 73.3% |
| #6416 | floor · high | Jul 30, 2026 | 0.326Ξ | 0.318Ξ | -2.5% | 97.6% |
| #6713 | grail · high | Jul 30, 2026 | 0.332Ξ | 0.318Ξ | -4.3% | 95.7% |
| #4106 | floor · high | Jul 30, 2026 | 0.332Ξ | 0.318Ξ | -4.3% | 95.7% |
| #3410 | floor · high | Jul 30, 2026 | 0.320Ξ | 0.318Ξ | -0.7% | 99.3% |
| #6416 | floor · high | Jul 30, 2026 | 0.302Ξ | 0.362Ξ | +19.9% | 80.1% |
| #3990 | floor · medium | Jul 30, 2026 | 0.310Ξ | 0.428Ξ | +38.1% | 61.9% |
| #5229 | middle · high | Jul 29, 2026 | 0.400Ξ | 0.675Ξ | +68.8% | 31.3% |
| #4106 | floor · high | Jul 29, 2026 | 0.320Ξ | 0.431Ξ | +34.7% | 65.3% |
| #6713 | grail · high | Jul 29, 2026 | 0.321Ξ | 0.371Ξ | +15.4% | 84.6% |
| #3707 | floor · high | Jul 29, 2026 | 0.319Ξ | 0.441Ξ | +38.2% | 61.8% |
| #6096 | floor · high | Jul 29, 2026 | 0.323Ξ | 0.364Ξ | +12.8% | 87.2% |
| #3410 | floor · high | Jul 28, 2026 | 0.317Ξ | 0.378Ξ | +19.2% | 80.8% |
| #6951 | floor · high | Jul 28, 2026 | 0.334Ξ | 0.391Ξ | +17.1% | 82.9% |
| #2241 | floor · high | Jul 28, 2026 | 0.330Ξ | 0.373Ξ | +13.0% | 87.0% |
| #6718 | middle · high | Jul 28, 2026 | 0.358Ξ | 0.503Ξ | +40.5% | 59.5% |
| #2255 | floor · high | Jul 27, 2026 | 0.370Ξ | 0.385Ξ | +4.1% | 96.0% |
| #6096 | floor · high | Jul 27, 2026 | 0.370Ξ | 0.388Ξ | +4.9% | 95.1% |
| #6713 | grail · high | Jul 27, 2026 | 0.377Ξ | 0.521Ξ | +38.2% | 61.8% |
| #6061 | floor · high | Jul 27, 2026 | 0.400Ξ | 0.366Ξ | -8.6% | 91.4% |
| #373 | floor · high | Jul 27, 2026 | 0.388Ξ | 0.353Ξ | -9.0% | 91.0% |
Model cryptoadz-1785800663-2026-08-03.2 · evaluated through 8/3/2026. Method details on the methodology page.