Matt Taibbi: 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
82.6%z 4.23 - 2
81.4%z 4.13 - 3
76.2%z 3.97 - 4
62.3%z 3.82 - 5
46.8%z 3.59 - 6
45.4%z 3.50 - 7
44.6%z 3.45 - 8
43.7%z 3.38 - 9
43.1%z 3.35 - 10
42.1%z 3.28 - 11
41.7%z 3.26 - 12
41.3%z 3.24 - 13
41%z 3.22 - 14
40.2%z 3.16 - 15
39.9%z 3.14 - 16
39.4%z 3.11 - 17
39.2%z 3.10 - 18
39%z 3.09 - 19
38.9%z 3.07 - 20
38.8%z 3.07 - 21
38.7%z 3.06 - 22
38.6%z 3.06 - 23
38.5%z 3.05 - 24
38%z 3.02 - 25
37.8%z 3.00 - 26
37.5%z 2.99 - 27
37.5%z 2.98 - 28
37.4%z 2.98 - 29
37.4%z 2.98 - 30
37.2%z 2.96 - 31
36.8%z 2.94 - 32
36.7%z 2.94 - 33
36.6%z 2.93 - 34
36.6%z 2.93 - 35
36.5%z 2.92 - 36
36.5%z 2.92 - 37
36.3%z 2.91 - 38
36.1%z 2.89 - 39
35.8%z 2.88 - 40
35.8%z 2.87 - 41
35.7%z 2.87 - 42
35.3%z 2.84 - 43
35.3%z 2.84 - 44
35.2%z 2.84 - 45
35%z 2.82 - 46
35%z 2.82 - 47
34.9%z 2.82 - 48
34.8%z 2.81 - 49
34.8%z 2.81 - 50
34.7%z 2.80