ODDPUCK
Between seasonsProjections liveLatest sheet Sat, Apr 11, 2026

Projections

Every headline stat per skater — over a full regular season, ranked league-wide by assists.

Each player's recency-weighted per-game rate — from 5 seasons of real games, recent games weighted more (the weighting is backtested) — × their team's scheduled games in the window. Computed overnight, then only read — this page never recalculates a projection. Availability is not projected: Proj GP is the team's schedule, not a forecast of how many games the player dresses for, and injuries, scratches and trades are not modelled.

How well does projection ordering hold? View the ordering study

897 skaters · sorted by assists

#
Skater
PTS
G
A
SOG
PPP
HIT
BLK
PIM
Proj GP
501
24.3
8.9
15.4
84.8
1.9
28.5
37.9
15.5
82
502
23.3
7.9
15.4
63.6
2.8
91.2
99.9
43.7
82
503
29.5
14.2
15.4
133.4
2.4
171.4
56.0
42.2
82
504
19.8
4.4
15.4
81.5
0.1
114.9
151.4
69.3
82
505
30.6
15.2
15.3
121.1
5.5
126.6
41.8
47.5
82
506
19.6
4.3
15.3
123.9
1.4
84.9
76.4
97.5
82
507
18.9
3.6
15.3
69.9
0.9
127.3
109.9
54.6
82
508
24.5
9.3
15.2
102.3
1.3
107.2
36.5
43.0
82
509
18.9
3.7
15.2
96.6
0.1
157.8
139.7
24.0
82
510
20.4
5.2
15.2
117.2
0.2
51.2
40.6
34.9
82
511
20.9
5.9
15.1
94.2
0.1
142.9
153.8
39.8
82
512
22.9
7.8
15.1
141.6
7.0
43.7
21.3
40.8
82
513
29.3
14.2
15.1
116.5
1.7
90.5
44.3
33.8
82
514
30.6
15.5
15.1
113.8
5.0
65.3
58.8
23.5
82
515
33.9
18.8
15.1
118.7
8.9
60.4
29.3
16.9
82
516
16.7
1.6
15.1
77.2
2.6
109.7
106.6
31.6
82
517
21.9
6.9
15.0
87.6
2.2
65.3
35.4
24.9
82
518
21.1
6.1
14.9
92.1
8.8
26.1
27.8
12.7
82
519
29.7
14.7
14.9
111.9
1.8
111.3
27.0
13.6
82
520
16.2
1.3
14.9
82.7
0.1
84.5
79.6
25.3
82
521
15.2
0.3
14.9
50.1
0.2
59.7
112.8
12.8
82
522
18.4
3.6
14.9
86.2
0.9
47.9
72.5
28.8
82
523
18.6
3.8
14.8
64.6
1.3
118.5
114.3
54.9
82
524
24.7
9.9
14.8
99.3
3.2
133.8
65.5
18.2
82
525
18.2
3.4
14.8
102.3
0.1
40.6
114.0
33.2
82
526
32.0
17.3
14.7
133.5
3.6
80.2
38.6
14.0
82
527
26.4
11.7
14.7
68.3
1.2
64.5
45.5
19.6
82
528
39.1
24.5
14.7
118.2
0.3
89.6
40.4
8.5
82
529
23.2
8.5
14.6
92.4
3.1
57.7
32.2
34.8
82
530
24.3
9.7
14.6
88.5
2.5
70.0
38.9
17.9
82
531
19.5
5.0
14.5
103.5
0.7
60.8
96.1
33.1
82
532
36.9
22.4
14.5
109.6
1.7
153.0
42.5
48.4
82
533
31.3
16.9
14.5
140.1
3.4
131.0
18.7
32.2
82
534
15.6
1.2
14.4
70.5
0.1
103.9
138.3
50.6
82
535
19.4
5.0
14.4
68.5
0.4
54.1
101.2
21.2
82
536
19.1
4.7
14.4
106.1
1.6
136.0
89.1
39.3
82
537
28.5
14.2
14.3
92.6
3.3
31.7
15.2
13.7
82
538
17.8
3.5
14.3
60.3
0.1
155.0
72.8
93.7
82
539
20.7
6.4
14.3
97.3
6.5
101.3
62.4
37.8
82
540
26.6
12.3
14.3
134.7
3.2
53.6
17.1
23.9
82
541
16.7
2.5
14.2
60.2
2.4
93.6
82.3
32.1
82
542
24.2
10.0
14.2
80.2
0.1
227.4
44.7
48.7
82
543
28.6
14.4
14.2
154.7
1.9
106.5
42.3
22.9
82
544
19.9
5.8
14.2
47.7
0.1
129.2
25.3
65.4
82
545
17.8
3.7
14.2
129.9
0.3
130.4
93.2
35.9
82
546
25.5
11.4
14.1
106.0
2.1
59.0
30.4
14.9
82
547
17.3
3.3
14.1
74.7
0.1
137.1
136.0
38.0
82
548
24.0
9.9
14.1
103.8
2.0
64.2
25.1
20.9
82
549
18.4
4.3
14.0
101.0
3.4
127.3
67.4
43.5
82
550
22.6
8.6
14.0
92.5
0.2
160.6
40.4
78.7
82

No consistency figure for assists: it did not clear the nightly split-half-vs-Poisson gate. The gate withholds a figure both when a stat is measurably no better than chance and when the evidence is too split to call, and those are not separable from here — so read this as no published measure, not as a finding that assists is inconsistent. Sort by another stat to see it where it holds. Measured 2026-08-21. The nightly job has not refreshed this since — it is the last completed measurement, not a current one.