Brad Lander: raw computed results
The raw ranking from the production similarity calculation (face-embedding centroid cosine → gender-cohort z-normalization). The web-opinion crown, curation, and fusion re-rank are removed.
- 1
82.2%z 4.20 - 2
81.1%z 4.11 - 3
78.4%z 4.00 - 4
68.9%z 3.90 - 5
59.2%z 3.79 - 6
58.2%z 3.78 - 7
52.8%z 3.72 - 8
47.2%z 3.61 - 9
47.2%z 3.61 - 10
46.9%z 3.60 - 11
46.4%z 3.56 - 12
45.9%z 3.53 - 13
45.8%z 3.53 - 14
45.8%z 3.53 - 15
45.5%z 3.51 - 16
45.5%z 3.51 - 17
45%z 3.47 - 18
44.1%z 3.41 - 19
43.9%z 3.40 - 20
43.5%z 3.38 - 21
43.5%z 3.37 - 22
43.3%z 3.36 - 23
43.2%z 3.36 - 24
43.1%z 3.35 - 25
42.9%z 3.34 - 26
42.6%z 3.32 - 27
41.7%z 3.26 - 28
41.5%z 3.25 - 29
40.9%z 3.20 - 30
40.7%z 3.19 - 31
40.2%z 3.16 - 32
40.1%z 3.16 - 33
39.8%z 3.14 - 34
39.7%z 3.13 - 35
39.6%z 3.12 - 36
39.4%z 3.11 - 37
39.1%z 3.09 - 38
39%z 3.08 - 39
38.8%z 3.07 - 40
38.7%z 3.06 - 41
38.4%z 3.05 - 42
38.4%z 3.04 - 43
38.3%z 3.04 - 44
38.2%z 3.03 - 45
38.2%z 3.03 - 46
38%z 3.02 - 47
37.8%z 3.01 - 48
37.8%z 3.00 - 49
37.3%z 2.98 - 50
37.3%z 2.97