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)
92.8%
Mean accuracy (100−APE)
90.1%
Within ±5%
35%
Within ±10%
66%
Within ±20%
89%
Bias
-1.9%
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 | 1610 | 92.9% | 68% | -2.1% |
| middle | 35 | 67.5% | 14% | +4.2% |
| grail | 5 | 58.6% | 0% | +17.2% |
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 |
|---|---|---|---|---|---|---|
| #2982 | floor · high | Sep 18, 2026 | 0.130Ξ | 0.170Ξ | +30.8% | 69.2% |
| #6205 | floor · high | Sep 18, 2026 | 0.165Ξ | 0.145Ξ | -12.1% | 87.9% |
| #6970 | floor · high | Sep 18, 2026 | 0.139Ξ | 0.139Ξ | 0.0% | 100.0% |
| #3035 | floor · high | Sep 17, 2026 | 0.133Ξ | 0.138Ξ | +3.8% | 96.2% |
| #6237 | floor · high | Sep 17, 2026 | 0.139Ξ | 0.141Ξ | +1.6% | 98.4% |
| #6970 | floor · high | Sep 17, 2026 | 0.134Ξ | 0.144Ξ | +7.5% | 92.5% |
| #6970 | floor · high | Sep 17, 2026 | 0.143Ξ | 0.140Ξ | -2.0% | 98.0% |
| #1347 | floor · high | Sep 17, 2026 | 0.135Ξ | 0.140Ξ | +3.7% | 96.3% |
| #8143 | floor · high | Sep 17, 2026 | 0.135Ξ | 0.146Ξ | +8.2% | 91.9% |
| #3035 | floor · high | Sep 17, 2026 | 0.137Ξ | 0.145Ξ | +5.7% | 94.3% |
| #4907 | floor · high | Sep 17, 2026 | 0.137Ξ | 0.135Ξ | -1.6% | 98.4% |
| #6997 | floor · high | Sep 17, 2026 | 0.158Ξ | 0.135Ξ | -14.5% | 85.5% |
| #8100 | floor · high | Sep 16, 2026 | 0.135Ξ | 0.141Ξ | +4.4% | 95.6% |
| #760 | floor · medium | Sep 16, 2026 | 0.157Ξ | 0.165Ξ | +5.1% | 94.9% |
| #4907 | floor · high | Sep 16, 2026 | 0.134Ξ | 0.147Ξ | +9.7% | 90.3% |
| #6237 | floor · high | Sep 16, 2026 | 0.137Ξ | 0.153Ξ | +11.7% | 88.3% |
| #6970 | floor · high | Sep 16, 2026 | 0.136Ξ | 0.142Ξ | +4.4% | 95.6% |
| #2289 | floor · high | Sep 16, 2026 | 0.147Ξ | 0.133Ξ | -9.5% | 90.5% |
| #4907 | floor · high | Sep 16, 2026 | 0.147Ξ | 0.137Ξ | -6.8% | 93.2% |
| #4907 | floor · high | Sep 16, 2026 | 0.137Ξ | 0.152Ξ | +11.0% | 89.1% |
| #4907 | floor · high | Sep 16, 2026 | 0.147Ξ | 0.128Ξ | -12.9% | 87.1% |
| #5562 | floor · high | Sep 16, 2026 | 0.137Ξ | 0.160Ξ | +16.8% | 83.2% |
| #2431 | floor · high | Sep 16, 2026 | 0.170Ξ | 0.150Ξ | -11.7% | 88.3% |
| #7041 | floor · high | Sep 16, 2026 | 0.243Ξ | 0.136Ξ | -44.0% | 56.0% |
| #6357 | middle · high | Sep 16, 2026 | 0.270Ξ | 0.216Ξ | -20.0% | 80.0% |
| #7147 | floor · high | Sep 16, 2026 | 0.158Ξ | 0.125Ξ | -20.9% | 79.1% |
| #6254 | floor · high | Sep 16, 2026 | 0.147Ξ | 0.132Ξ | -10.2% | 89.8% |
| #7888 | floor · high | Sep 16, 2026 | 0.170Ξ | 0.149Ξ | -12.3% | 87.7% |
| #7760 | floor · high | Sep 16, 2026 | 0.199Ξ | 0.132Ξ | -33.7% | 66.3% |
| #7593 | floor · high | Sep 16, 2026 | 0.159Ξ | 0.125Ξ | -21.5% | 78.5% |
| #3035 | floor · high | Sep 16, 2026 | 0.139Ξ | 0.134Ξ | -3.6% | 96.4% |
| #6237 | floor · high | Sep 16, 2026 | 0.148Ξ | 0.132Ξ | -10.8% | 89.2% |
| #7098 | floor · high | Sep 16, 2026 | 0.170Ξ | 0.151Ξ | -11.1% | 88.9% |
| #3035 | floor · high | Sep 16, 2026 | 0.134Ξ | 0.130Ξ | -2.7% | 97.3% |
| #7760 | floor · medium | Sep 16, 2026 | 0.126Ξ | 0.140Ξ | +11.1% | 88.9% |
| #7768 | floor · high | Sep 15, 2026 | 0.134Ξ | 0.125Ξ | -6.4% | 93.6% |
| #5308 | floor · high | Sep 15, 2026 | 0.134Ξ | 0.126Ξ | -5.9% | 94.1% |
| #7031 | floor · high | Sep 15, 2026 | 0.134Ξ | 0.125Ξ | -6.7% | 93.4% |
| #2349 | floor · high | Sep 15, 2026 | 0.134Ξ | 0.125Ξ | -6.4% | 93.6% |
| #1347 | floor · high | Sep 15, 2026 | 0.134Ξ | 0.125Ξ | -6.4% | 93.6% |
Model rektguy-1789708535-2026-08-03.2 · evaluated through 9/18/2026. Method details on the methodology page.