Erik Friedlander: 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
52%z 3.71 - 2
47.6%z 3.64 - 3
47.3%z 3.62 - 4
46.9%z 3.60 - 5
44.7%z 3.45 - 6
44.6%z 3.45 - 7
44.3%z 3.43 - 8
44.3%z 3.43 - 9
44.1%z 3.42 - 10
44.1%z 3.41 - 11
43.9%z 3.40 - 12
43.8%z 3.39 - 13
42.8%z 3.33 - 14
42.7%z 3.32 - 15
42.4%z 3.30 - 16
41.9%z 3.27 - 17
41.5%z 3.25 - 18
41.3%z 3.23 - 19
40.7%z 3.19 - 20
40.5%z 3.18 - 21
39.7%z 3.13 - 22
39.3%z 3.10 - 23
38.9%z 3.08 - 24
38.9%z 3.08 - 25
38.7%z 3.06 - 26
38.6%z 3.06 - 27
38.5%z 3.05 - 28
38.4%z 3.05 - 29
38.4%z 3.04 - 30
38.1%z 3.03 - 31
37.8%z 3.01 - 32
37.7%z 3.00 - 33
37.5%z 2.99 - 34
37.3%z 2.97 - 35
37.1%z 2.96 - 36
37.1%z 2.96 - 37
36.8%z 2.94 - 38
36.8%z 2.94 - 39
36.6%z 2.93 - 40
36.5%z 2.92 - 41
36.3%z 2.91 - 42
36.2%z 2.90 - 43
36.1%z 2.90 - 44
36%z 2.89 - 45
35.9%z 2.88 - 46
35.9%z 2.88 - 47
35.4%z 2.85 - 48
35.4%z 2.85 - 49
35.4%z 2.85 - 50
35.3%z 2.85