AppraisrAppraisr
Accuracy

How accurate is the CrypToadz 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)
89.3%
Mean accuracy (100−APE)
84.0%
Within ±5%
28%
Within ±10%
48%
Within ±20%
72%
Bias
+7.7%

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
floor137990.6%52%+5.0%
middle9355.5%11%+38.6%
grail17587.2%40%+12.3%

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
#1981floor · highSep 13, 20260.500Ξ0.324Ξ-35.2%64.8%
#1800grail · highSep 12, 20260.267Ξ0.360Ξ+34.8%65.2%
#2300grail · highSep 12, 20260.270Ξ0.378Ξ+40.0%60.0%
#2497floor · highSep 12, 20260.277Ξ0.360Ξ+30.0%70.0%
#755floor · highSep 12, 20260.279Ξ0.270Ξ-3.3%96.7%
#755floor · highSep 12, 20260.270Ξ0.280Ξ+3.7%96.3%
#755floor · highSep 11, 20260.280Ξ0.344Ξ+22.9%77.1%
#3794floor · highSep 11, 20260.318Ξ0.305Ξ-4.1%95.9%
#4805floor · highSep 11, 20260.292Ξ0.269Ξ-7.7%92.3%
#3482middle · mediumSep 11, 20260.420Ξ0.644Ξ+53.3%46.7%
#4087floor · highSep 10, 20260.279Ξ0.277Ξ-0.8%99.2%
#5701floor · mediumSep 10, 20260.256Ξ0.334Ξ+30.5%69.5%
#1628floor · highSep 10, 20260.270Ξ0.342Ξ+26.7%73.3%
#4087floor · highSep 9, 20260.278Ξ0.302Ξ+8.6%91.4%
#4852grail · highSep 9, 20260.275Ξ0.354Ξ+28.7%71.3%
#6339floor · highSep 9, 20260.279Ξ0.332Ξ+19.0%81.0%
#2849floor · highSep 9, 20260.285Ξ0.290Ξ+1.8%98.3%
#877floor · highSep 8, 20260.304Ξ0.303Ξ-0.5%99.5%
#4135grail · highSep 7, 20260.800Ξ0.959Ξ+19.9%80.1%
#6947floor · highSep 6, 20260.278Ξ0.371Ξ+33.5%66.6%
#1853floor · highSep 6, 20260.371Ξ0.281Ξ-24.3%75.7%
#3132floor · highSep 6, 20260.362Ξ0.276Ξ-23.8%76.2%
#877floor · highSep 5, 20260.300Ξ0.382Ξ+27.3%72.7%
#5660floor · highSep 5, 20260.274Ξ0.270Ξ-1.4%98.6%
#6921floor · highSep 4, 20260.267Ξ0.364Ξ+36.3%63.7%
#5660floor · highSep 4, 20260.269Ξ0.273Ξ+1.5%98.5%
#5365middle · mediumSep 4, 20260.269Ξ0.700Ξ+160.2%0.0%
#1551floor · highSep 3, 20260.271Ξ0.360Ξ+32.8%67.2%
#365floor · highSep 3, 20260.272Ξ0.339Ξ+24.6%75.4%
#3717floor · highSep 2, 20260.284Ξ0.331Ξ+16.6%83.5%
#2740floor · highSep 1, 20260.289Ξ0.280Ξ-3.0%97.0%
#4805floor · highAug 31, 20260.276Ξ0.345Ξ+25.0%75.0%
#2321floor · highAug 31, 20260.272Ξ0.343Ξ+26.1%73.9%
#1726floor · highAug 31, 20260.310Ξ0.376Ξ+21.3%78.7%
#5293floor · highAug 31, 20260.280Ξ0.305Ξ+8.9%91.1%
#2016floor · highAug 30, 20260.282Ξ0.283Ξ+0.4%99.7%
#4661floor · highAug 30, 20260.300Ξ0.340Ξ+13.3%86.7%
#5541floor · mediumAug 30, 20260.283Ξ0.365Ξ+29.0%71.0%
#830floor · highAug 29, 20260.283Ξ0.352Ξ+24.4%75.6%
#229floor · highAug 29, 20260.283Ξ0.361Ξ+27.6%72.4%

Model cryptoadz-1789305347-2026-08-03.2 · evaluated through 9/13/2026. Method details on the methodology page.