AppraisrAppraisr
Accuracy

How accurate is the DeGods 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)
93.6%
Mean accuracy (100−APE)
92.0%
Within ±5%
41%
Within ±10%
72%
Within ±20%
94%
Bias
-2.8%

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
floor52593.6%72%-2.9%
middle151.4%0%+48.6%

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

ItemTier / evidenceDateSaleModelErrorAccuracy
#1505floor · highSep 18, 20260.212Ξ0.160Ξ-24.5%75.5%
#6434floor · highSep 18, 20260.287Ξ0.179Ξ-37.6%62.4%
#7039floor · highSep 16, 20260.165Ξ0.152Ξ-7.7%92.3%
#8770floor · highSep 16, 20260.125Ξ0.164Ξ+31.2%68.8%
#1625floor · highSep 15, 20260.150Ξ0.148Ξ-1.3%98.7%
#3517floor · highSep 14, 20260.148Ξ0.168Ξ+13.5%86.5%
#1625floor · highSep 14, 20260.139Ξ0.166Ξ+19.4%80.6%
#1251floor · highSep 14, 20260.148Ξ0.168Ξ+13.5%86.5%
#4549floor · highSep 12, 20260.167Ξ0.170Ξ+1.8%98.2%
#7494floor · highSep 12, 20260.168Ξ0.172Ξ+2.4%97.6%
#7039floor · highSep 10, 20260.164Ξ0.195Ξ+18.9%81.1%
#7494floor · highSep 10, 20260.170Ξ0.195Ξ+14.7%85.3%
#277floor · highSep 10, 20260.163Ξ0.176Ξ+8.0%92.0%
#4549floor · highSep 10, 20260.166Ξ0.200Ξ+20.5%79.5%
#2540floor · highSep 10, 20260.161Ξ0.176Ξ+9.3%90.7%
#4549floor · highSep 10, 20260.200Ξ0.199Ξ-0.5%99.5%
#7039floor · highSep 10, 20260.195Ξ0.171Ξ-12.3%87.7%
#7494floor · highSep 10, 20260.195Ξ0.161Ξ-17.4%82.6%
#277floor · highSep 10, 20260.176Ξ0.161Ξ-8.6%91.4%
#2540floor · highSep 10, 20260.176Ξ0.162Ξ-8.0%92.0%
#5887floor · highSep 7, 20260.158Ξ0.204Ξ+29.1%70.9%
#7039floor · highSep 5, 20260.171Ξ0.164Ξ-4.1%95.9%
#5887floor · highSep 5, 20260.200Ξ0.174Ξ-13.0%87.0%
#2540floor · highSep 5, 20260.160Ξ0.193Ξ+20.6%79.4%
#7494floor · highSep 5, 20260.160Ξ0.234Ξ+46.3%53.8%
#277floor · highSep 5, 20260.170Ξ0.192Ξ+12.9%87.1%
#7124floor · highSep 4, 20260.273Ξ0.195Ξ-28.6%71.4%
#9482floor · highSep 2, 20260.190Ξ0.196Ξ+3.2%96.8%
#7204floor · highSep 1, 20260.190Ξ0.192Ξ+1.1%99.0%
#5887middle · highSep 1, 20260.181Ξ0.269Ξ+48.6%51.4%
#5887floor · highAug 31, 20260.269Ξ0.190Ξ-29.4%70.6%
#277floor · highAug 30, 20260.210Ξ0.187Ξ-11.0%89.1%
#7494floor · highAug 30, 20260.236Ξ0.184Ξ-22.2%77.8%
#2540floor · highAug 30, 20260.196Ξ0.181Ξ-7.7%92.4%
#1872floor · highAug 29, 20260.202Ξ0.181Ξ-10.4%89.6%
#7111floor · highAug 29, 20260.191Ξ0.169Ξ-11.5%88.5%
#190floor · highAug 24, 20260.200Ξ0.193Ξ-3.5%96.5%
#4058floor · highAug 24, 20260.167Ξ0.202Ξ+21.0%79.0%
#7111floor · highAug 22, 20260.170Ξ0.206Ξ+21.2%78.8%
#7113floor · highAug 19, 20260.191Ξ0.206Ξ+7.9%92.2%

Model degods-1789712387-2026-08-03.2 · evaluated through 9/18/2026. Method details on the methodology page.