Accuracy
How accurate is the rektguy 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)
92.5%
Mean accuracy (100−APE)
89.7%
Within ±5%
34%
Within ±10%
64%
Within ±20%
88%
Bias
-0.6%
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 | 1640 | 92.8% | 65% | -0.9% |
| middle | 49 | 74.4% | 20% | +8.8% |
| grail | 5 | 58.6% | 0% | +18.4% |
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 |
|---|---|---|---|---|---|---|
| #4280 | floor · high | Nov 2, 2026 | 0.136Ξ | 0.124Ξ | -8.8% | 91.2% |
| #5886 | floor · high | Nov 1, 2026 | 0.173Ξ | 0.145Ξ | -16.2% | 83.8% |
| #4611 | floor · medium | Nov 1, 2026 | 0.121Ξ | 0.123Ξ | +1.7% | 98.4% |
| #6462 | floor · high | Nov 1, 2026 | 0.167Ξ | 0.140Ξ | -16.2% | 83.8% |
| #5678 | floor · high | Nov 1, 2026 | 0.134Ξ | 0.128Ξ | -4.3% | 95.7% |
| #6462 | floor · medium | Nov 1, 2026 | 0.135Ξ | 0.148Ξ | +9.6% | 90.4% |
| #8619 | floor · high | Nov 1, 2026 | 0.125Ξ | 0.123Ξ | -1.6% | 98.4% |
| #5452 | floor · high | Nov 1, 2026 | 0.121Ξ | 0.120Ξ | -0.8% | 99.2% |
| #8619 | floor · high | Nov 1, 2026 | 0.123Ξ | 0.119Ξ | -3.3% | 96.8% |
| #573 | floor · medium | Oct 31, 2026 | 0.174Ξ | 0.122Ξ | -29.9% | 70.1% |
| #4430 | floor · medium | Oct 31, 2026 | 0.180Ξ | 0.160Ξ | -11.1% | 88.9% |
| #2354 | floor · high | Oct 31, 2026 | 0.179Ξ | 0.168Ξ | -6.2% | 93.9% |
| #6126 | floor · high | Oct 31, 2026 | 0.122Ξ | 0.130Ξ | +6.6% | 93.4% |
| #389 | floor · medium | Oct 31, 2026 | 0.179Ξ | 0.138Ξ | -22.9% | 77.1% |
| #1573 | floor · high | Oct 31, 2026 | 0.129Ξ | 0.124Ξ | -3.8% | 96.2% |
| #5678 | floor · high | Oct 31, 2026 | 0.123Ξ | 0.113Ξ | -8.1% | 91.9% |
| #5019 | floor · high | Oct 31, 2026 | 0.124Ξ | 0.117Ξ | -5.7% | 94.4% |
| #2914 | floor · high | Oct 31, 2026 | 0.128Ξ | 0.148Ξ | +15.2% | 84.8% |
| #6126 | floor · high | Oct 31, 2026 | 0.129Ξ | 0.113Ξ | -12.1% | 87.9% |
| #384 | floor · high | Oct 31, 2026 | 0.131Ξ | 0.121Ξ | -7.6% | 92.4% |
| #1717 | floor · high | Oct 31, 2026 | 0.128Ξ | 0.114Ξ | -11.3% | 88.7% |
| #8568 | floor · high | Oct 31, 2026 | 0.119Ξ | 0.120Ξ | +0.8% | 99.2% |
| #528 | floor · high | Oct 31, 2026 | 0.120Ξ | 0.112Ξ | -6.7% | 93.3% |
| #407 | floor · high | Oct 31, 2026 | 0.120Ξ | 0.112Ξ | -6.7% | 93.3% |
| #975 | floor · medium | Oct 31, 2026 | 0.120Ξ | 0.115Ξ | -4.2% | 95.8% |
| #1063 | floor · medium | Oct 31, 2026 | 0.120Ξ | 0.113Ξ | -5.8% | 94.3% |
| #3589 | floor · medium | Oct 31, 2026 | 0.120Ξ | 0.123Ξ | +2.6% | 97.4% |
| #8054 | floor · high | Oct 31, 2026 | 0.120Ξ | 0.111Ξ | -7.4% | 92.6% |
| #3669 | floor · high | Oct 31, 2026 | 0.120Ξ | 0.125Ξ | +4.4% | 95.6% |
| #8568 | floor · high | Oct 31, 2026 | 0.120Ξ | 0.114Ξ | -4.7% | 95.3% |
| #3169 | floor · high | Oct 31, 2026 | 0.120Ξ | 0.124Ξ | +3.6% | 96.4% |
| #631 | floor · high | Oct 31, 2026 | 0.120Ξ | 0.112Ξ | -6.4% | 93.6% |
| #2082 | floor · medium | Oct 30, 2026 | 0.120Ξ | 0.119Ξ | -0.8% | 99.3% |
| #7005 | floor · high | Oct 30, 2026 | 0.119Ξ | 0.128Ξ | +7.4% | 92.6% |
| #3769 | floor · medium | Oct 29, 2026 | 0.179Ξ | 0.126Ξ | -29.6% | 70.4% |
| #7862 | floor · high | Oct 29, 2026 | 0.166Ξ | 0.157Ξ | -5.4% | 94.6% |
| #8568 | floor · high | Oct 29, 2026 | 0.112Ξ | 0.118Ξ | +5.4% | 94.6% |
| #528 | floor · high | Oct 29, 2026 | 0.111Ξ | 0.117Ξ | +5.4% | 94.6% |
| #3558 | floor · high | Oct 29, 2026 | 0.113Ξ | 0.117Ξ | +3.5% | 96.5% |
| #7862 | floor · high | Oct 29, 2026 | 0.158Ξ | 0.163Ξ | +3.2% | 96.8% |
Model rektguy-1793595959-2026-08-03.2 · evaluated through 11/2/2026. Method details on the methodology page.