AppraisrAppraisr
Accuracy

How accurate is the mfers 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)
92.6%
Mean accuracy (100−APE)
90.1%
Within ±5%
36%
Within ±10%
63%
Within ±20%
89%
Bias
+2.6%

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
floor149892.6%63%+2.8%
middle2472.8%25%+0.6%
grail1780.4%29%-17.9%

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
#6643floor · highSep 19, 20260.141Ξ0.157Ξ+11.4%88.7%
#4311floor · highSep 19, 20260.144Ξ0.154Ξ+6.9%93.1%
#5211floor · highSep 19, 20260.144Ξ0.167Ξ+16.0%84.0%
#5870floor · highSep 19, 20260.151Ξ0.162Ξ+7.3%92.7%
#1727floor · highSep 19, 20260.165Ξ0.155Ξ-5.9%94.1%
#5756floor · highSep 19, 20260.150Ξ0.160Ξ+6.7%93.3%
#883floor · highSep 19, 20260.165Ξ0.152Ξ-7.9%92.1%
#9254floor · highSep 19, 20260.149Ξ0.146Ξ-2.0%98.0%
#883floor · highSep 18, 20260.152Ξ0.158Ξ+4.0%96.1%
#1583floor · highSep 18, 20260.150Ξ0.157Ξ+4.7%95.3%
#1056floor · highSep 18, 20260.150Ξ0.158Ξ+5.3%94.7%
#8109floor · highSep 18, 20260.151Ξ0.159Ξ+5.3%94.7%
#1727floor · highSep 18, 20260.155Ξ0.151Ξ-2.6%97.4%
#9597floor · highSep 18, 20260.155Ξ0.164Ξ+5.8%94.2%
#2294floor · highSep 18, 20260.157Ξ0.148Ξ-5.7%94.3%
#8126floor · highSep 18, 20260.158Ξ0.154Ξ-2.5%97.5%
#883floor · highSep 18, 20260.158Ξ0.150Ξ-5.0%95.0%
#435floor · highSep 18, 20260.165Ξ0.153Ξ-7.5%92.5%
#8889floor · highSep 18, 20260.164Ξ0.149Ξ-9.2%90.8%
#2959floor · highSep 18, 20260.158Ξ0.149Ξ-5.7%94.4%
#8652floor · highSep 18, 20260.165Ξ0.151Ξ-8.7%91.3%
#9597floor · highSep 18, 20260.164Ξ0.159Ξ-3.1%96.9%
#6403floor · highSep 18, 20260.158Ξ0.155Ξ-1.9%98.1%
#8049floor · highSep 18, 20260.159Ξ0.154Ξ-3.1%96.9%
#2959floor · highSep 18, 20260.148Ξ0.151Ξ+2.0%98.0%
#883floor · highSep 18, 20260.149Ξ0.157Ξ+5.4%94.6%
#8109floor · highSep 18, 20260.156Ξ0.148Ξ-5.2%94.8%
#883floor · highSep 18, 20260.155Ξ0.148Ξ-4.4%95.6%
#1583floor · highSep 18, 20260.154Ξ0.147Ξ-4.6%95.5%
#6643floor · highSep 18, 20260.154Ξ0.152Ξ-1.3%98.7%
#5870floor · highSep 18, 20260.156Ξ0.144Ξ-7.7%92.3%
#1080floor · highSep 18, 20260.154Ξ0.148Ξ-3.9%96.1%
#8858floor · highSep 18, 20260.156Ξ0.147Ξ-5.8%94.2%
#6416floor · highSep 18, 20260.154Ξ0.150Ξ-2.6%97.4%
#2239floor · highSep 18, 20260.154Ξ0.148Ξ-3.9%96.1%
#6416floor · highSep 18, 20260.146Ξ0.157Ξ+7.5%92.5%
#8858floor · highSep 18, 20260.145Ξ0.144Ξ-0.7%99.3%
#3365floor · highSep 18, 20260.142Ξ0.152Ξ+6.7%93.3%
#1727floor · highSep 18, 20260.149Ξ0.159Ξ+6.7%93.3%
#2436floor · highSep 18, 20260.149Ξ0.152Ξ+2.0%98.0%

Model mfers-1789796135-2026-08-03.2 · evaluated through 9/19/2026. Method details on the methodology page.