AppraisrAppraisr
Accuracy

How accurate is the Otherdeed for Otherside 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)
94.6%
Mean accuracy (100−APE)
88.0%
Within ±5%
49%
Within ±10%
65%
Within ±20%
84%
Bias
-0.5%

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
floor164195.7%69%-1.0%
middle4060.5%10%-5.9%
grail26789.9%49%+3.0%

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
#89071floor · highAug 3, 20260.100Ξ0.104Ξ+4.1%95.9%
#93174floor · highAug 3, 20260.100Ξ0.108Ξ+8.1%91.9%
#70271floor · highAug 3, 20260.100Ξ0.113Ξ+13.1%86.9%
#47375floor · highAug 2, 20260.093Ξ0.100Ξ+8.0%92.0%
#59803floor · highAug 2, 20260.093Ξ0.102Ξ+10.0%90.0%
#56056grail · highAug 1, 20260.099Ξ0.099Ξ+0.3%99.8%
#84896floor · highAug 1, 20260.099Ξ0.102Ξ+3.2%96.8%
#53452floor · highAug 1, 20260.099Ξ0.098Ξ-1.2%98.8%
#56056grail · highJul 31, 20260.096Ξ0.099Ξ+2.9%97.1%
#53452floor · highJul 31, 20260.098Ξ0.096Ξ-1.8%98.2%
#77110floor · highJul 31, 20260.125Ξ0.102Ξ-18.4%81.6%
#88336floor · highJul 31, 20260.109Ξ0.103Ξ-5.3%94.7%
#88336floor · highJul 31, 20260.100Ξ0.111Ξ+11.2%88.8%
#84896floor · highJul 31, 20260.100Ξ0.103Ξ+3.1%96.9%
#3172floor · highJul 31, 20260.135Ξ0.116Ξ-14.2%85.8%
#3172floor · highJul 30, 20260.111Ξ0.110Ξ-0.7%99.3%
#3172floor · highJul 30, 20260.102Ξ0.121Ξ+18.6%81.4%
#51990grail · highJul 29, 20260.113Ξ0.098Ξ-13.5%86.5%
#88086floor · highJul 29, 20260.091Ξ0.101Ξ+10.6%89.4%
#96922grail · highJul 29, 20260.099Ξ0.096Ξ-2.8%97.3%
#81134grail · highJul 29, 20260.210Ξ0.131Ξ-37.6%62.4%
#66972floor · highJul 28, 20260.099Ξ0.099Ξ+0.2%99.8%
#88073floor · highJul 28, 20260.199Ξ0.119Ξ-40.2%59.8%
#51446floor · highJul 28, 20260.104Ξ0.101Ξ-2.5%97.5%
#66972floor · highJul 28, 20260.095Ξ0.107Ξ+12.2%87.8%
#53floor · highJul 28, 20263.750Ξ0.125Ξ-96.7%3.3%
#91060floor · highJul 28, 20260.110Ξ0.109Ξ-0.9%99.1%
#84743floor · highJul 28, 20260.212Ξ0.123Ξ-42.0%58.0%
#96922grail · highJul 28, 20260.095Ξ0.103Ξ+8.5%91.5%
#94111grail · insufficientJul 28, 20260.700Ξ0.875Ξ+25.0%75.0%
#69951floor · highJul 28, 20260.114Ξ0.105Ξ-7.9%92.1%
#15336grail · lowJul 28, 20261.360Ξ0.977Ξ-28.2%71.8%
#86851grail · highJul 28, 20260.165Ξ0.105Ξ-36.3%63.7%
#50909floor · highJul 27, 20260.105Ξ0.099Ξ-5.7%94.3%
#25134floor · highJul 27, 20260.097Ξ0.104Ξ+7.1%92.9%
#28537floor · highJul 27, 20260.116Ξ0.117Ξ+0.5%99.5%
#30593floor · highJul 27, 20260.097Ξ0.085Ξ-11.9%88.1%
#88275floor · highJul 27, 20260.100Ξ0.106Ξ+6.3%93.7%
#93174floor · highJul 27, 20260.100Ξ0.106Ξ+6.3%93.7%
#51990grail · highJul 27, 20260.104Ξ0.106Ξ+1.9%98.1%

Model otherdeed-1785727811-2026-08-03.2 · evaluated through 8/3/2026. Method details on the methodology page.