NFL · Empirical Bayes

True-Talent Leaderboards

A raw single-season rate is a noisy guess at a player’s real ability, and the noise is worst for the smallest samples — so a naïve leaderboard is topped by whoever got lucky in the fewest tries. This one shrinks every rate toward its position-group prior by how much the sample can be trusted, and shows a 90% credible interval instead of a point. Toggle raw vs. shrunk to watch the flukes fall back to the pack.

Catch rate · 2018

Receptions per target, shrunk within position (WR vs TE separately). Depth-of-target confound; catchable-target rate not available here.

Beats raw by
18.3%
lower out-of-sample error
RMSE raw → shrunk
11.1% → 9.1%
odd vs. even weeks
Split-half reliability
0.35
how repeatable the raw stat is
Stabilizes at
48
targets to trust the number
shrunkrawthe shrink90% CI
40%50%60%70%80%WR avg1Michael ThomasNO · 14879.0%2Tyler LockettSEA · 7173.0%3Ryan SwitzerPIT · 4471.6%4DeAndre CarterHOU · 2571.0%5Adam ThielenMIN · 15470.7%6Cole BeasleyDAL · 8770.3%7Danny AmendolaMIA · 7970.0%8Emmanuel SandersDEN · 9869.1%9Adam HumphriesTB · 10668.7%10Phillip DorsettNE · 4268.7%11Taylor GabrielCHI · 9368.7%12DeAndre HopkinsHOU · 16368.7%13Keenan AllenLAC · 13768.6%14Rashard HigginsCLE · 5368.2%15Chester RogersIND · 7467.9%16Tyler BoydCIN · 10867.9%17Sammy WatkinsKC · 5567.8%18Cooper KuppLA · 5567.8%19Jordy NelsonLV · 8967.8%20Alshon JefferyPHI · 9267.8%21Amari CooperDAL · 10767.7%22Bruce EllingtonDET · 4267.6%23Albert WilsonMIA · 3567.3%24Calvin RidleyATL · 9267.0%25Chris MooreBAL · 2566.9%26Jarius WrightCAR · 6166.8%27Seth RobertsLV · 6466.8%28Stefon DiggsMIN · 15066.6%29Brandin CooksLA · 11766.6%30Julian EdelmanNE · 10866.6%31Will FullerHOU · 4566.5%32Dontrelle InmanIND · 3966.5%33Eli RogersPIT · 1566.4%34Jaron BrownSEA · 1965.5%35Brandon LaFellLV · 1665.4%36T.J. JonesDET · 2765.1%37Darius JenningsTEN · 1564.8%

Each row is a player: the solid dot is the shrunk estimate, the hollow dot the raw rate, joined by the red pull of regression; the grey bar is the 90% credible interval. A hollow shrunk dot means the sample is below the stabilization line — the number is mostly the position prior. Switch to Raw rank and watch the small-sample names climb.

Catch rate · 2018 · full board

Catch rate leaderboard for the 2018 NFL season, shrunk by empirical Bayes with 90% credible intervals, ranked by the shrunk estimate.
#PlayerTeamPostargetsRawShrunk90% interval
1Michael ThomasNOWR14884.5%79.0%74.1%83.6%
2Tyler LockettSEAWR7180.3%73.0%66.1%79.5%
3Ryan Switzer◦ provisionalPITWR4481.8%71.6%63.6%79.0%
4DeAndre Carter◦ provisionalHOUWR2588.0%71.0%62.0%79.4%
5Adam ThielenMINWR15473.4%70.7%65.3%75.9%
6Cole BeasleyDALWR8774.7%70.3%63.6%76.5%
7Danny AmendolaMIAWR7974.7%70.0%63.1%76.5%
8Emmanuel SandersDENWR9872.5%69.1%62.7%75.2%
9Adam HumphriesTBWR10671.7%68.7%62.5%74.7%
10Phillip Dorsett◦ provisionalNEWR4276.2%68.7%60.5%76.5%
11Taylor GabrielCHIWR9372.0%68.7%62.1%75.0%
12DeAndre HopkinsHOUWR16370.5%68.7%63.3%73.8%
13Keenan AllenLACWR13770.8%68.6%62.9%74.1%
14Rashard HigginsCLEWR5373.6%68.2%60.4%75.6%
15Chester RogersINDWR7471.6%67.9%60.8%74.7%
16Tyler BoydCINWR10870.4%67.9%61.6%73.9%
17Sammy WatkinsKCWR5572.7%67.8%60.1%75.2%
18Cooper KuppLAWR5572.7%67.8%60.1%75.2%
19Jordy NelsonLVWR8970.8%67.8%61.1%74.2%
20Alshon JefferyPHIWR9270.7%67.8%61.1%74.1%
21Amari CooperDALWR10770.1%67.7%61.4%73.7%
22Bruce Ellington◦ provisionalDETWR4273.8%67.6%59.3%75.5%
23Albert Wilson◦ provisionalMIAWR3574.3%67.3%58.6%75.5%
24Calvin RidleyATLWR9269.6%67.0%60.4%73.4%
25Chris Moore◦ provisionalBALWR2576.0%66.9%57.6%75.7%
26Jarius WrightCARWR6170.5%66.8%59.3%74.1%
27Seth RobertsLVWR6470.3%66.8%59.4%74.0%
28Stefon DiggsMINWR15068.0%66.6%61.0%72.0%
29Brandin CooksLAWR11768.4%66.6%60.4%72.5%
30Julian EdelmanNEWR10868.5%66.6%60.3%72.7%
31Mohamed SanuATLWR9668.8%66.6%60.0%72.9%
32Will Fuller◦ provisionalHOUWR4571.1%66.5%58.3%74.3%
33Dontrelle Inman◦ provisionalINDWR3971.8%66.5%58.0%74.6%
34Eli Rogers◦ provisionalPITWR1580.0%66.4%56.4%75.9%
35Doug BaldwinSEAWR7368.5%66.0%58.8%72.9%
36Julio JonesATLWR17066.5%65.5%60.2%70.7%
37JuJu Smith-SchusterPITWR16766.5%65.5%60.1%70.8%
38Jaron Brown◦ provisionalSEAWR1973.7%65.5%55.7%74.7%
39Brandon LaFell◦ provisionalLVWR1675.0%65.4%55.4%74.8%
40DJ MooreCARWR8267.1%65.3%58.3%72.0%
41T.J. Jones◦ provisionalDETWR2770.4%65.1%55.9%73.9%
42Martavis Bryant◦ provisionalLVWR2770.4%65.1%55.9%73.9%
43Robert WoodsLAWR13066.1%65.1%59.1%70.9%
44Keke Coutee◦ provisionalHOUWR4168.3%65.0%56.5%73.1%
45Ryan GrantINDWR5267.3%64.8%56.9%72.5%
46Darius Jennings◦ provisionalTENWR1573.3%64.8%54.7%74.4%
47Alex Erickson◦ provisionalCINWR2969.0%64.7%55.6%73.4%
48Nelson AgholorPHIWR9766.0%64.7%58.1%71.1%
49Davante AdamsGBWR17065.3%64.6%59.2%69.8%
50Demaryius ThomasHOUWR9065.6%64.4%57.6%71.0%
51Trent Sherfield◦ provisionalARIWR2867.9%64.3%55.1%73.1%
52Willie SneadBALWR9565.3%64.2%57.5%70.7%
53Laquon TreadwellMINWR5366.0%64.2%56.2%71.9%
54Golden TatePHIWR11464.9%64.1%57.8%70.2%
55Demarcus Robinson◦ provisionalKCWR3366.7%64.0%55.1%72.5%
56Geronimo Allison◦ provisionalGBWR3066.7%63.9%54.8%72.6%
57Cody Latimer◦ provisionalNYGWR1668.8%63.8%53.8%73.4%
58Taywan TaylorTENWR5764.9%63.7%55.8%71.2%
59DaeSean Hamilton◦ provisionalDENWR4665.2%63.7%55.4%71.6%
60Dede WestbrookJAXWR10364.1%63.5%56.9%69.8%
61Deontay Burnett◦ provisionalNYJWR1566.7%63.2%53.1%72.9%
62Tyreek HillKCWR13763.5%63.2%57.3%68.9%
63T.Y. HiltonINDWR12063.3%63.0%56.8%69.0%
64Chris HoganNEWR5563.6%63.0%55.0%70.6%
65Tre'Quan Smith◦ provisionalNOWR4463.6%62.9%54.5%71.0%
66Brandon Powell◦ provisionalDETWR1764.7%62.8%52.8%72.4%
67Breshad Perriman◦ provisionalCLEWR2564.0%62.8%53.3%71.9%
68Mike WilliamsLACWR6863.2%62.8%55.3%70.0%
69Trent Taylor◦ provisionalSFWR4163.4%62.7%54.2%71.0%
70Tyrell WilliamsLACWR6563.1%62.7%55.1%70.0%
71Austin Carr◦ provisionalNOWR1464.3%62.7%52.4%72.5%
72Richie James◦ provisionalSFWR1464.3%62.7%52.4%72.5%
73Justin Hardy◦ provisionalATLWR2263.6%62.6%53.0%71.9%
74Kendrick BourneSFWR6762.7%62.5%54.9%69.8%
75Ty Montgomery◦ provisionalBALWR4062.5%62.3%53.7%70.6%
76Christian KirkARIWR6962.3%62.3%54.8%69.5%
77Randall CobbGBWR6162.3%62.2%54.5%69.7%
78Damion Ratley◦ provisionalCLEWR2161.9%62.1%52.3%71.5%
79Keith Kirkwood◦ provisionalNOWR2161.9%62.1%52.3%71.5%
80Tavon Austin◦ provisionalDALWR1361.5%62.0%51.6%72.0%
81Jakeem Grant◦ provisionalMIAWR3461.8%62.0%53.1%70.6%
82Antonio BrownPITWR16861.9%62.0%56.5%67.3%
83Mike EvansTBWR13961.9%62.0%56.0%67.7%
84Chris ConleyKCWR5261.5%61.8%53.8%69.7%
85Robert Foster◦ provisionalBUFWR4461.4%61.8%53.3%69.9%
86Odell Beckham Jr.NYGWR12561.6%61.8%55.6%67.8%
87Anthony MillerCHIWR5461.1%61.6%53.6%69.4%
88Sterling ShepardNYGWR10861.1%61.4%55.0%67.8%
89Larry FitzgeraldARIWR11361.1%61.4%55.0%67.6%
90Isaiah McKenzie◦ provisionalBUFWR3060.0%61.3%52.1%70.2%
91Chris Godwin Jr.TBWR9760.8%61.3%54.5%67.8%
92Bennie Fowler◦ provisionalNYGWR2759.3%61.1%51.7%70.2%
93Dante Pettis◦ provisionalSFWR4560.0%61.1%52.7%69.3%
94Curtis SamuelCARWR6560.0%60.9%53.3%68.3%
95Maurice Harris◦ provisionalWASWR4759.6%60.9%52.5%69.0%
96Jamison CrowderWASWR4959.2%60.7%52.4%68.7%
97Equanimeous St. Brown◦ provisionalGBWR3658.3%60.5%51.6%69.1%
98Zach Pascal◦ provisionalINDWR4658.7%60.5%52.1%68.6%
99Andre Roberts◦ provisionalNYJWR1855.6%60.4%50.3%70.0%
100A.J. GreenCINWR7859.0%60.2%52.9%67.3%
101Deonte Thompson◦ provisionalBUFWR3056.7%60.1%50.8%69.0%
102Allen Hurns◦ provisionalDALWR3557.1%60.1%51.1%68.7%
103Paul Richardson◦ provisionalWASWR3557.1%60.1%51.1%68.7%
104Josh GordonNEWR7157.8%59.5%52.1%66.8%
105Andre Holmes◦ provisionalDENWR2454.2%59.5%49.9%68.8%
106Russell Shepard◦ provisionalNYGWR1952.6%59.5%49.5%69.1%
107Tim Patrick◦ provisionalDENWR4156.1%59.4%50.7%67.8%
108Allen RobinsonCHIWR9557.9%59.3%52.5%66.0%
109Ted Ginn◦ provisionalNOWR3154.8%59.3%50.1%68.2%
110Corey DavisTENWR11258.0%59.3%52.8%65.6%
111Josh Bellamy◦ provisionalCHIWR2653.8%59.2%49.7%68.5%
112Kenny StillsMIAWR6556.9%59.2%51.5%66.6%
113Marvin JonesDETWR6256.5%58.9%51.2%66.5%
114Tajae Sharpe◦ provisionalTENWR4755.3%58.8%50.4%67.0%
115Kenny GolladayDETWR12257.4%58.7%52.5%64.9%
116Josh DoctsonWASWR7856.4%58.6%51.3%65.7%
117Marvin Hall◦ provisionalATLWR2050.0%58.6%48.7%68.2%
118Torrey Smith◦ provisionalCARWR3253.1%58.6%49.4%67.5%
119Josh ReynoldsLAWR5354.7%58.3%50.1%66.2%
120Devin FunchessCARWR7955.7%58.1%50.9%65.3%
121Travis Benjamin◦ provisionalLACWR2450.0%58.1%48.5%67.5%
122DeSean JacksonTBWR7455.4%58.1%50.7%65.3%
123Terrelle Pryor◦ provisionalBUFWR3151.6%58.0%48.8%67.0%
124Quincy EnunwaNYJWR6955.1%58.0%50.4%65.4%
125Brandon Marshall◦ provisionalSEAWR2347.8%57.5%47.8%67.0%
126Keelan ColeJAXWR7054.3%57.5%50.0%64.9%
127Marquise Goodwin◦ provisionalSFWR4452.3%57.4%48.9%65.8%
128Zay JonesBUFWR10354.4%56.8%50.2%63.4%
129Donte MoncriefJAXWR8953.9%56.8%49.8%63.7%
130Marcell Ateman◦ provisionalLVWR3148.4%56.8%47.5%65.8%
131Andy Jones◦ provisionalDETWR2445.8%56.7%47.0%66.2%
132Pierre Garcon◦ provisionalSFWR4751.1%56.7%48.3%64.9%
133DeVante Parker◦ provisionalMIAWR4751.1%56.7%48.3%64.9%
134Michael CrabtreeBALWR10054.0%56.6%49.9%63.3%
135Aldrick Robinson◦ provisionalMINWR3548.6%56.4%47.4%65.3%
136Auden Tate◦ provisionalCINWR1233.3%56.4%45.8%66.7%
137Robbie ChosenNYJWR9453.2%56.2%49.3%63.0%
138Antonio CallawayCLEWR8252.4%56.0%48.8%63.1%
139Jarvis LandryCLEWR15054.0%56.0%50.1%61.7%
140Cody Core◦ provisionalCINWR2944.8%55.6%46.3%64.8%
141Marquez Valdes-ScantlingGBWR7451.3%55.6%48.2%62.9%
142David MooreSEAWR5349.1%55.3%47.1%63.3%
143J.J. Nelson◦ provisionalARIWR1936.8%55.0%44.9%64.8%
144DJ Chark◦ provisionalJAXWR3243.8%54.8%45.6%63.8%
145Michael Floyd◦ provisionalWASWR2540.0%54.6%44.9%64.0%
146Courtland SuttonDENWR8549.4%54.0%46.9%61.1%
147Jermaine KearseNYJWR7648.7%53.9%46.5%61.2%
148Justin Hunter◦ provisionalPITWR1323.1%53.8%43.3%64.2%
149Michael GallupDALWR6947.8%53.7%46.1%61.2%
150James Washington◦ provisionalPITWR3842.1%53.3%44.4%62.1%
151Chad Williams◦ provisionalARIWR4637.0%49.8%41.3%58.3%
152John BrownBALWR9842.9%49.2%42.4%56.0%
153John RossCINWR5836.2%47.9%40.0%55.9%
154Kelvin BenjaminKCWR6737.3%47.7%40.1%55.3%

How the shrinkage works

Two estimators

Rate stats (completion %, success rate, catch rate) use a beta-binomial model: a Beta(α, β) prior fit by marginal likelihood over each position group, then a Beta posterior per player. Per-play averages (EPA, CPOE, yards) use a normal-normal model with DerSimonian–Laird between-player variance. Both pull each player toward their group by exactly how thin their sample is.

Does it help? & the fine print

The trust panel’s numbers come from a leakage-free odd/even-week holdout: fit on odd weeks, predict even-week raw. Shrinkage lowers out-of-sample error for every stat. Priors are fit per season and per position (WR and TE separately), so a TE’s baseline isn’t a skill. Regular season only, 2016–2025. Rushing and receiving efficiency are heavily scheme-driven — read the wide bands as the honesty they are.

Source: nflverse play-by-play. Counted and computed deterministically — never modeled by a language model. Built by build_nfl_leaderboards.py (byte-reproducible).