Accuracy
How accurate is the BEANZ Official 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)
92.2%
Mean accuracy (100−APE)
87.4%
Within ±5%
33%
Within ±10%
58%
Within ±20%
81%
Bias
-6.1%
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 | 2106 | 92.6% | 61% | -6.0% |
| middle | 68 | 78.1% | 21% | +0.9% |
| grail | 81 | 78.6% | 21% | -15.7% |
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 |
|---|---|---|---|---|---|---|
| #7726 | floor · high | Nov 2, 2026 | 0.027Ξ | 0.027Ξ | -0.7% | 99.3% |
| #4349 | floor · high | Nov 2, 2026 | 0.028Ξ | 0.027Ξ | -2.5% | 97.5% |
| #9861 | floor · high | Nov 2, 2026 | 0.027Ξ | 0.028Ξ | +2.6% | 97.4% |
| #8098 | floor · high | Nov 2, 2026 | 0.027Ξ | 0.028Ξ | +2.9% | 97.1% |
| #6969 | floor · high | Nov 2, 2026 | 0.027Ξ | 0.028Ξ | +2.2% | 97.8% |
| #11818 | floor · high | Nov 2, 2026 | 0.069Ξ | 0.031Ξ | -55.1% | 44.9% |
| #3564 | grail · medium | Nov 2, 2026 | 0.200Ξ | 0.110Ξ | -45.0% | 55.0% |
| #7877 | floor · high | Nov 2, 2026 | 0.030Ξ | 0.028Ξ | -6.7% | 93.3% |
| #17289 | floor · high | Nov 1, 2026 | 0.028Ξ | 0.028Ξ | +1.8% | 98.2% |
| #15256 | floor · high | Nov 1, 2026 | 0.030Ξ | 0.028Ξ | -6.7% | 93.3% |
| #13988 | floor · high | Nov 1, 2026 | 0.030Ξ | 0.028Ξ | -6.7% | 93.3% |
| #1581 | floor · high | Nov 1, 2026 | 0.050Ξ | 0.035Ξ | -30.1% | 69.9% |
| #17289 | floor · high | Nov 1, 2026 | 0.027Ξ | 0.028Ξ | +2.6% | 97.4% |
| #5432 | floor · high | Nov 1, 2026 | 0.037Ξ | 0.029Ξ | -21.8% | 78.2% |
| #9173 | floor · high | Nov 1, 2026 | 0.029Ξ | 0.028Ξ | -4.4% | 95.6% |
| #17184 | floor · high | Nov 1, 2026 | 0.029Ξ | 0.029Ξ | -0.7% | 99.3% |
| #10391 | floor · high | Nov 1, 2026 | 0.027Ξ | 0.028Ξ | +3.3% | 96.7% |
| #10001 | floor · high | Nov 1, 2026 | 0.028Ξ | 0.030Ξ | +8.3% | 91.7% |
| #15420 | floor · high | Nov 1, 2026 | 0.028Ξ | 0.030Ξ | +8.7% | 91.3% |
| #4323 | floor · high | Nov 1, 2026 | 0.028Ξ | 0.029Ξ | +5.5% | 94.6% |
| #1126 | floor · high | Nov 1, 2026 | 0.028Ξ | 0.029Ξ | +5.1% | 94.9% |
| #11388 | floor · high | Nov 1, 2026 | 0.028Ξ | 0.029Ξ | +5.1% | 94.9% |
| #12698 | floor · high | Nov 1, 2026 | 0.028Ξ | 0.029Ξ | +3.9% | 96.1% |
| #13988 | floor · high | Nov 1, 2026 | 0.028Ξ | 0.029Ξ | +3.6% | 96.4% |
| #15256 | floor · high | Nov 1, 2026 | 0.029Ξ | 0.030Ξ | +5.3% | 94.7% |
| #447 | floor · high | Nov 1, 2026 | 0.030Ξ | 0.029Ξ | -4.6% | 95.4% |
| #10933 | floor · high | Nov 1, 2026 | 0.060Ξ | 0.036Ξ | -40.0% | 60.0% |
| #6687 | floor · high | Nov 1, 2026 | 0.029Ξ | 0.030Ξ | +5.3% | 94.7% |
| #6366 | floor · high | Oct 31, 2026 | 0.029Ξ | 0.029Ξ | +1.1% | 99.0% |
| #5551 | floor · high | Oct 31, 2026 | 0.029Ξ | 0.029Ξ | +0.7% | 99.3% |
| #8265 | floor · high | Oct 31, 2026 | 0.029Ξ | 0.032Ξ | +11.1% | 88.9% |
| #7856 | floor · high | Oct 30, 2026 | 0.030Ξ | 0.029Ξ | -4.9% | 95.1% |
| #16326 | floor · high | Oct 30, 2026 | 0.027Ξ | 0.030Ξ | +10.3% | 89.7% |
| #11174 | floor · high | Oct 30, 2026 | 0.028Ξ | 0.038Ξ | +38.2% | 61.8% |
| #8003 | floor · high | Oct 30, 2026 | 0.028Ξ | 0.030Ξ | +8.3% | 91.7% |
| #8003 | floor · high | Oct 30, 2026 | 0.030Ξ | 0.030Ξ | -0.2% | 99.8% |
| #7877 | floor · high | Oct 30, 2026 | 0.028Ξ | 0.032Ξ | +15.1% | 84.9% |
| #19773 | floor · high | Oct 30, 2026 | 0.028Ξ | 0.032Ξ | +15.1% | 84.9% |
| #4089 | floor · high | Oct 30, 2026 | 0.029Ξ | 0.031Ξ | +6.2% | 93.8% |
| #3109 | floor · high | Oct 30, 2026 | 0.029Ξ | 0.032Ξ | +9.6% | 90.4% |
Model beanz-1793595851-2026-08-03.2 · evaluated through 11/2/2026. Method details on the methodology page.