AppraisrAppraisr
Accuracy

How accurate is the CryptoPunks V1 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)
88.6%
Mean accuracy (100−APE)
85.0%
Within ±5%
25%
Within ±10%
45%
Within ±20%
73%
Bias
-5.6%

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
floor127889.0%47%-6.3%
middle8474.3%21%+5.4%
grail2287.6%45%-3.2%

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

ItemBodyTier / evidenceDateSaleModelErrorAccuracy
#9975Malefloor · highAug 2, 20262.460Ξ2.234Ξ-9.2%90.8%
#8605Femalefloor · highAug 2, 20261.980Ξ2.453Ξ+23.9%76.1%
#1696Malefloor · highAug 1, 20262.400Ξ2.440Ξ+1.7%98.3%
#1896Malefloor · highAug 1, 20263.380Ξ2.382Ξ-29.5%70.5%
#2108Malefloor · highJul 31, 20262.500Ξ2.290Ξ-8.4%91.6%
#2756Femalefloor · highJul 30, 20262.000Ξ2.170Ξ+8.5%91.5%
#2970Malefloor · highJul 30, 20262.390Ξ2.322Ξ-2.8%97.2%
#2283Femalefloor · highJul 28, 20262.349Ξ2.204Ξ-6.2%93.8%
#6253Femalefloor · highJul 27, 20262.300Ξ2.188Ξ-4.9%95.1%
#3569Malemiddle · highJul 27, 20262.440Ξ3.582Ξ+46.8%53.2%
#1696Malemiddle · highJul 26, 20262.446Ξ3.222Ξ+31.7%68.3%
#3815Malefloor · highJul 26, 20263.000Ξ2.573Ξ-14.2%85.8%
#8966Femalefloor · highJul 26, 20262.500Ξ2.489Ξ-0.4%99.6%
#6731Femalefloor · highJul 26, 20263.500Ξ2.519Ξ-28.0%72.0%
#1570Malefloor · highJul 26, 20262.210Ξ2.967Ξ+34.3%65.8%
#7769Malefloor · highJul 25, 20262.550Ξ2.331Ξ-8.6%91.4%
#1855Femalefloor · highJul 25, 20262.450Ξ2.040Ξ-16.7%83.3%
#6394Femalefloor · highJul 25, 20262.436Ξ2.203Ξ-9.6%90.4%
#2145Femalefloor · highJul 25, 20262.250Ξ2.141Ξ-4.8%95.2%
#4707Malefloor · highJul 24, 20262.850Ξ2.535Ξ-11.1%89.0%
#1490Malefloor · highJul 24, 20262.960Ξ2.467Ξ-16.7%83.3%
#1307Malefloor · highJul 24, 20262.170Ξ2.260Ξ+4.2%95.9%
#3804Malefloor · highJul 23, 20262.420Ξ2.981Ξ+23.2%76.8%
#1490Malemiddle · highJul 23, 20262.468Ξ4.056Ξ+64.3%35.7%
#1005Malemiddle · highJul 23, 20263.000Ξ3.691Ξ+23.0%77.0%
#6253Femalefloor · highJul 23, 20262.185Ξ2.032Ξ-7.0%93.0%
#6253Femalefloor · highJul 22, 20262.010Ξ2.099Ξ+4.4%95.6%
#9747Femalefloor · highJul 21, 20262.040Ξ2.943Ξ+44.3%55.7%
#1551Malefloor · highJul 21, 20262.399Ξ2.569Ξ+7.1%92.9%
#6841Malefloor · highJul 21, 20262.650Ξ2.226Ξ-16.0%84.0%
#3254Femalefloor · highJul 20, 20262.270Ξ2.094Ξ-7.8%92.3%
#7769Malefloor · highJul 20, 20262.350Ξ2.635Ξ+12.1%87.9%
#6297Zombiegrail · lowJul 18, 202633.00Ξ34.36Ξ+4.1%95.9%
#8259Malefloor · highJul 16, 20262.850Ξ2.597Ξ-8.9%91.1%
#1141Femalefloor · highJul 16, 20262.000Ξ2.356Ξ+17.8%82.2%
#2970Malefloor · highJul 14, 20262.260Ξ2.110Ξ-6.6%93.4%
#7769Malefloor · highJul 14, 20262.650Ξ2.292Ξ-13.5%86.5%
#7721Malefloor · highJul 13, 20262.900Ξ2.678Ξ-7.7%92.3%
#2002Malefloor · highJul 12, 20265.250Ξ2.451Ξ-53.3%46.7%
#6869Malemiddle · highJul 11, 20262.800Ξ3.912Ξ+39.7%60.3%

Model v1-punks-1785682583-2026-08-03.2 · evaluated through 8/2/2026. Method details on the methodology page.