Accuracy
How accurate is the Remilio Babies 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)
93.5%
Mean accuracy (100−APE)
89.8%
Within ±5%
41%
Within ±10%
66%
Within ±20%
88%
Bias
+2.0%
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 | 1353 | 94.0% | 68% | +1.8% |
| middle | 52 | 73.5% | 29% | +18.4% |
| grail | 74 | 89.4% | 49% | -6.6% |
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 |
|---|---|---|---|---|---|---|
| #3071 | floor · high | Aug 3, 2026 | 0.162Ξ | 0.178Ξ | +10.0% | 90.0% |
| #6421 | floor · high | Aug 2, 2026 | 0.151Ξ | 0.181Ξ | +19.9% | 80.1% |
| #9624 | floor · high | Aug 2, 2026 | 0.152Ξ | 0.164Ξ | +7.9% | 92.1% |
| #2731 | floor · high | Aug 2, 2026 | 0.157Ξ | 0.181Ξ | +15.3% | 84.7% |
| #6435 | floor · high | Aug 1, 2026 | 0.161Ξ | 0.182Ξ | +12.8% | 87.2% |
| #8167 | floor · high | Aug 1, 2026 | 0.155Ξ | 0.167Ξ | +7.7% | 92.3% |
| #3075 | floor · high | Aug 1, 2026 | 0.169Ξ | 0.183Ξ | +8.4% | 91.6% |
| #2980 | floor · high | Aug 1, 2026 | 0.162Ξ | 0.170Ξ | +5.0% | 95.0% |
| #4995 | grail · high | Jul 31, 2026 | 0.770Ξ | 0.698Ξ | -9.4% | 90.7% |
| #7077 | floor · high | Jul 31, 2026 | 0.259Ξ | 0.245Ξ | -5.4% | 94.6% |
| #2897 | middle · high | Jul 31, 2026 | 0.245Ξ | 0.256Ξ | +4.5% | 95.5% |
| #8167 | floor · high | Jul 31, 2026 | 0.154Ξ | 0.184Ξ | +19.5% | 80.5% |
| #6314 | floor · high | Jul 31, 2026 | 0.154Ξ | 0.190Ξ | +23.4% | 76.6% |
| #9961 | floor · high | Jul 31, 2026 | 0.205Ξ | 0.174Ξ | -15.1% | 84.9% |
| #5455 | floor · high | Jul 31, 2026 | 0.190Ξ | 0.185Ξ | -2.6% | 97.4% |
| #8126 | middle · medium | Jul 31, 2026 | 0.223Ξ | 0.281Ξ | +26.0% | 74.0% |
| #7077 | middle · high | Jul 30, 2026 | 0.231Ξ | 0.272Ξ | +17.8% | 82.3% |
| #1069 | floor · high | Jul 30, 2026 | 0.179Ξ | 0.189Ξ | +5.6% | 94.4% |
| #2980 | floor · high | Jul 29, 2026 | 0.165Ξ | 0.193Ξ | +17.0% | 83.0% |
| #3170 | floor · high | Jul 29, 2026 | 0.166Ξ | 0.173Ξ | +4.2% | 95.8% |
| #8612 | floor · high | Jul 29, 2026 | 0.168Ξ | 0.207Ξ | +23.2% | 76.8% |
| #5380 | floor · high | Jul 27, 2026 | 0.188Ξ | 0.198Ξ | +5.1% | 94.9% |
| #1351 | floor · high | Jul 27, 2026 | 0.176Ξ | 0.188Ξ | +6.8% | 93.2% |
| #7309 | grail · medium | Jul 27, 2026 | 0.211Ξ | 0.220Ξ | +4.3% | 95.7% |
| #3075 | floor · high | Jul 27, 2026 | 0.177Ξ | 0.185Ξ | +4.5% | 95.5% |
| #1237 | floor · high | Jul 27, 2026 | 0.225Ξ | 0.189Ξ | -16.0% | 84.0% |
| #6435 | floor · high | Jul 27, 2026 | 0.178Ξ | 0.186Ξ | +4.5% | 95.5% |
| #3073 | floor · high | Jul 27, 2026 | 0.175Ξ | 0.179Ξ | +2.3% | 97.7% |
| #3071 | floor · high | Jul 27, 2026 | 0.181Ξ | 0.188Ξ | +3.9% | 96.1% |
| #5380 | floor · high | Jul 27, 2026 | 0.182Ξ | 0.203Ξ | +11.5% | 88.5% |
| #4816 | floor · high | Jul 27, 2026 | 0.190Ξ | 0.178Ξ | -6.3% | 93.7% |
| #2519 | floor · high | Jul 27, 2026 | 0.183Ξ | 0.179Ξ | -2.2% | 97.8% |
| #1237 | floor · high | Jul 27, 2026 | 0.183Ξ | 0.182Ξ | -0.6% | 99.5% |
| #1062 | floor · high | Jul 27, 2026 | 0.184Ξ | 0.180Ξ | -2.4% | 97.6% |
| #4816 | floor · high | Jul 26, 2026 | 0.174Ξ | 0.183Ξ | +5.2% | 94.8% |
| #1237 | floor · high | Jul 26, 2026 | 0.176Ξ | 0.180Ξ | +2.3% | 97.7% |
| #873 | floor · high | Jul 26, 2026 | 0.259Ξ | 0.248Ξ | -4.3% | 95.8% |
| #7872 | floor · high | Jul 26, 2026 | 0.176Ξ | 0.177Ξ | +0.6% | 99.4% |
| #3077 | floor · high | Jul 26, 2026 | 0.188Ξ | 0.178Ξ | -5.3% | 94.7% |
| #3477 | floor · high | Jul 25, 2026 | 0.178Ξ | 0.173Ξ | -2.8% | 97.2% |
Model remilio-1785724895-2026-08-03.2 · evaluated through 8/3/2026. Method details on the methodology page.