Matthew Chapman: 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
59.7%z 3.79 - 2
47.1%z 3.61 - 3
43.7%z 3.39 - 4
43.3%z 3.36 - 5
42.6%z 3.32 - 6
42.6%z 3.32 - 7
40.4%z 3.17 - 8
40.2%z 3.16 - 9
40%z 3.15 - 10
39.9%z 3.14 - 11
39.8%z 3.14 - 12
39.5%z 3.11 - 13
39.2%z 3.10 - 14
38.4%z 3.05 - 15
38.1%z 3.02 - 16
37.8%z 3.00 - 17
37.3%z 2.98 - 18
36.5%z 2.92 - 19
36.5%z 2.92 - 20
36.2%z 2.90 - 21
36%z 2.89 - 22
35.7%z 2.87 - 23
35.4%z 2.85 - 24
35.3%z 2.84 - 25
35.2%z 2.84 - 26
35%z 2.82 - 27
34.7%z 2.81 - 28
34.3%z 2.78 - 29
34.3%z 2.78 - 30
33.7%z 2.74 - 31
33.7%z 2.74 - 32
33.6%z 2.73 - 33
33.5%z 2.73 - 34
33.5%z 2.73 - 35
33.5%z 2.73 - 36
33.5%z 2.72 - 37
33.4%z 2.72 - 38
33.3%z 2.72 - 39
32.8%z 2.68 - 40
32.7%z 2.67 - 41
32.5%z 2.66 - 42
32.3%z 2.65 - 43
32.2%z 2.64 - 44
32.1%z 2.64 - 45
31.9%z 2.62 - 46
31.8%z 2.62 - 47
31.8%z 2.62 - 48
31.8%z 2.61 - 49
31.7%z 2.61 - 50
31.7%z 2.61