Marcus Gilbert: 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
78.7%z 4.00 - 2
75.5%z 3.97 - 3
72.5%z 3.93 - 4
66.5%z 3.87 - 5
60%z 3.80 - 6
47.3%z 3.62 - 7
46.8%z 3.59 - 8
46.8%z 3.59 - 9
46.2%z 3.55 - 10
44.2%z 3.42 - 11
43.8%z 3.40 - 12
42.5%z 3.31 - 13
42.4%z 3.30 - 14
42.4%z 3.30 - 15
41.8%z 3.26 - 16
41.6%z 3.25 - 17
41.4%z 3.24 - 18
41.2%z 3.22 - 19
41.1%z 3.22 - 20
40.8%z 3.20 - 21
40.8%z 3.20 - 22
40.7%z 3.20 - 23
40.6%z 3.19 - 24
40.6%z 3.19 - 25
40.5%z 3.18 - 26
40.4%z 3.18 - 27
40.3%z 3.17 - 28
40%z 3.15 - 29
39.9%z 3.14 - 30
39.4%z 3.11 - 31
39.4%z 3.11 - 32
39%z 3.09 - 33
38.9%z 3.08 - 34
38.2%z 3.03 - 35
38.2%z 3.03 - 36
38.2%z 3.03 - 37
38.1%z 3.03 - 38
38.1%z 3.02 - 39
38%z 3.02 - 40
37.8%z 3.01 - 41
37.8%z 3.01 - 42
37.7%z 3.00 - 43
37.6%z 2.99 - 44
37.5%z 2.99 - 45
37.4%z 2.98 - 46
37.3%z 2.97 - 47
37%z 2.95 - 48
36.8%z 2.94 - 49
36.8%z 2.94 - 50
36.8%z 2.94