AppraisrAppraisr
Accuracy

How accurate is the Sappy Seals 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)
93.0%
Mean accuracy (100−APE)
90.0%
Within ±5%
37%
Within ±10%
66%
Within ±20%
88%
Bias
-1.0%

Accuracy by market tier

Floor, middle-market, and grail sales are scored separately so aggregate accuracy cannot hide a weak segment

TierSalesMedian accuracyWithin ±10%Bias
floor181693.1%67%-1.0%
middle3681.3%19%-4.0%
grail1158.0%18%+3.4%

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

ItemTier / evidenceDateSaleModelErrorAccuracy
#5581floor · highSep 13, 20260.061Ξ0.065Ξ+5.9%94.1%
#1959floor · highSep 13, 20260.061Ξ0.064Ξ+4.1%95.9%
#7840floor · highSep 13, 20260.061Ξ0.067Ξ+8.9%91.1%
#5788floor · highSep 13, 20260.061Ξ0.067Ξ+8.9%91.1%
#6731floor · highSep 12, 20260.071Ξ0.063Ξ-11.1%88.9%
#4800floor · highSep 12, 20260.065Ξ0.069Ξ+6.7%93.4%
#2980floor · highSep 12, 20260.079Ξ0.072Ξ-8.8%91.3%
#1779floor · highSep 12, 20260.063Ξ0.072Ξ+14.8%85.2%
#3573floor · highSep 12, 20260.063Ξ0.069Ξ+10.2%89.8%
#431floor · highSep 12, 20260.062Ξ0.070Ξ+12.5%87.5%
#1602floor · highSep 12, 20260.065Ξ0.068Ξ+4.3%95.7%
#2139middle · highSep 11, 20260.088Ξ0.102Ξ+15.9%84.1%
#348floor · highSep 11, 20260.073Ξ0.075Ξ+3.0%97.0%
#116floor · highSep 11, 20260.070Ξ0.066Ξ-6.2%93.8%
#7727floor · highSep 11, 20260.071Ξ0.063Ξ-10.7%89.3%
#7727floor · highSep 11, 20260.063Ξ0.063Ξ+0.6%99.4%
#2980floor · highSep 11, 20260.072Ξ0.068Ξ-5.0%95.0%
#7297floor · highSep 11, 20260.066Ξ0.065Ξ-1.2%98.8%
#116floor · highSep 11, 20260.066Ξ0.066Ξ+0.2%99.9%
#7066floor · highSep 11, 20260.072Ξ0.066Ξ-8.2%91.8%
#6523floor · highSep 11, 20260.069Ξ0.068Ξ-1.3%98.7%
#6011floor · highSep 11, 20260.072Ξ0.071Ξ-1.3%98.8%
#777floor · highSep 11, 20260.069Ξ0.064Ξ-6.7%93.3%
#5502floor · highSep 11, 20260.069Ξ0.064Ξ-6.7%93.3%
#4718floor · highSep 11, 20260.069Ξ0.066Ξ-3.8%96.3%
#4552floor · highSep 11, 20260.070Ξ0.061Ξ-12.7%87.3%
#6037floor · highSep 10, 20260.078Ξ0.071Ξ-8.6%91.4%
#1131floor · highSep 10, 20260.104Ξ0.078Ξ-25.0%75.0%
#5391floor · highSep 10, 20260.068Ξ0.065Ξ-4.5%95.5%
#4381floor · highSep 10, 20260.068Ξ0.063Ξ-7.9%92.2%
#5677floor · highSep 10, 20260.075Ξ0.068Ξ-9.1%90.9%
#4381floor · highSep 9, 20260.063Ξ0.064Ξ+2.4%97.6%
#5502floor · highSep 9, 20260.063Ξ0.065Ξ+2.9%97.2%
#8465floor · highSep 9, 20260.200Ξ0.072Ξ-64.0%36.0%
#8238floor · highSep 9, 20260.066Ξ0.067Ξ+1.8%98.2%
#777floor · highSep 9, 20260.063Ξ0.064Ξ+2.2%97.8%
#287floor · highSep 9, 20260.062Ξ0.076Ξ+22.2%77.8%
#4871middle · highSep 9, 20260.119Ξ0.094Ξ-20.9%79.1%
#8238floor · highSep 9, 20260.063Ξ0.072Ξ+13.9%86.1%
#287floor · highSep 9, 20260.076Ξ0.063Ξ-17.1%82.9%

Model sappy-seals-1789261451-2026-08-03.2 · evaluated through 9/13/2026. Method details on the methodology page.