AppraisrAppraisr
CLONE X

CLONE X

19,764 items · DNA types

OpenSea ↗
Accuracy

How accurate is the CLONE X 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)
93.0%
Mean accuracy (100−APE)
89.1%
Within ±5%
38%
Within ±10%
63%
Within ±20%
87%
Bias
-3.4%

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
floor185693.7%68%-3.3%
middle15173.3%15%-3.5%
grail3677.4%36%-6.8%

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

ItemDNATier / evidenceDateSaleModelErrorAccuracy
#4900Robotfloor · lowAug 3, 20260.275Ξ0.275Ξ0.0%100.0%
#10645Angelfloor · insufficientAug 3, 20260.317Ξ0.282Ξ-11.0%89.0%
#7044Robotfloor · insufficientAug 3, 20260.277Ξ0.268Ξ-3.3%96.8%
#17903Demonfloor · insufficientAug 3, 20260.326Ξ0.301Ξ-7.7%92.3%
#2371Demonfloor · lowAug 3, 20260.320Ξ0.293Ξ-8.4%91.6%
#3011Robotfloor · lowAug 3, 20260.300Ξ0.285Ξ-5.0%95.0%
#19627Humanfloor · highAug 2, 20260.283Ξ0.285Ξ+0.7%99.3%
#4099Humanfloor · highAug 2, 20260.282Ξ0.267Ξ-5.3%94.7%
#4287Humanfloor · highAug 2, 20260.281Ξ0.282Ξ+0.4%99.6%
#9239Humanfloor · highAug 2, 20260.281Ξ0.276Ξ-1.8%98.2%
#3004Humanfloor · highAug 2, 20260.286Ξ0.265Ξ-7.3%92.7%
#19680Angelfloor · lowAug 2, 20260.330Ξ0.292Ξ-11.5%88.5%
#5184Humanfloor · highAug 2, 20260.287Ξ0.280Ξ-2.4%97.6%
#19627Humanfloor · highAug 2, 20260.277Ξ0.298Ξ+7.6%92.4%
#2672Robotfloor · lowAug 2, 20260.275Ξ0.289Ξ+5.1%94.9%
#4343Robotfloor · lowAug 2, 20260.285Ξ0.265Ξ-7.0%93.0%
#11886Humanfloor · highAug 2, 20260.286Ξ0.269Ξ-5.9%94.1%
#16456Humanfloor · highAug 2, 20260.286Ξ0.283Ξ-1.1%99.0%
#2290Humanfloor · highAug 2, 20260.309Ξ0.263Ξ-15.0%85.0%
#13623Humanfloor · highAug 2, 20260.296Ξ0.272Ξ-8.1%91.9%
#69Humanfloor · highAug 2, 20260.299Ξ0.287Ξ-4.0%96.0%
#493Robotmiddle · lowAug 2, 20260.550Ξ0.475Ξ-13.6%86.4%
#800Humanfloor · highAug 2, 20260.286Ξ0.273Ξ-4.6%95.5%
#11824Angelfloor · lowAug 2, 20260.330Ξ0.312Ξ-5.5%94.6%
#4853Angelfloor · lowAug 2, 20260.399Ξ0.328Ξ-17.8%82.2%
#3818Robotfloor · lowAug 2, 20260.286Ξ0.260Ξ-9.1%90.9%
#16903Humanfloor · highAug 2, 20260.287Ξ0.289Ξ+0.7%99.3%
#15173Robotfloor · lowAug 2, 20260.297Ξ0.285Ξ-4.2%95.8%
#19507Humanfloor · highAug 2, 20260.291Ξ0.277Ξ-4.8%95.2%
#15173Robotfloor · lowAug 2, 20260.285Ξ0.275Ξ-3.5%96.5%
#69Humanfloor · highAug 2, 20260.286Ξ0.283Ξ-1.1%99.0%
#2463Humanfloor · highAug 2, 20260.297Ξ0.271Ξ-8.7%91.3%
#79Robotfloor · lowAug 2, 20260.296Ξ0.267Ξ-9.8%90.2%
#4605Robotfloor · lowAug 2, 20260.290Ξ0.259Ξ-10.7%89.3%
#4626Humanfloor · highAug 1, 20260.290Ξ0.280Ξ-3.4%96.6%
#13822Humanfloor · highAug 1, 20260.299Ξ0.266Ξ-11.0%89.0%
#8073Humanfloor · highAug 1, 20260.400Ξ0.279Ξ-30.3%69.8%
#8147Humanfloor · highAug 1, 20260.307Ξ0.314Ξ+2.3%97.7%
#4626Humanfloor · highAug 1, 20260.274Ξ0.282Ξ+2.9%97.1%
#11774Demonfloor · lowAug 1, 20260.322Ξ0.276Ξ-14.3%85.7%

Model clone-x-1785734015-2026-08-03.2 · evaluated through 8/3/2026. Method details on the methodology page.