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 · 2017

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
53
targets to trust the number
shrunkrawthe shrink90% CI
40%50%60%70%WR avg1Golden TateDET · 12071.1%2Ted GinnNO · 7068.3%3Adam HumphriesTB · 8367.7%4JuJu Smith-Schust…PIT · 7967.4%5Tyreek HillKC · 10567.1%6Michael ThomasNO · 14966.8%7Jarvis LandryMIA · 16166.8%8Randall CobbGB · 9366.5%9Jeremy KerleyNYJ · 2766.3%10Danny AmendolaNE · 8666.2%11Mohamed SanuATL · 9665.8%12Sterling ShepardNYG · 8465.7%13De'Anthony ThomasKC · 1665.2%14Larry FitzgeraldARI · 16265.1%15Trent TaylorSF · 6164.9%16Ryan GrantWAS · 6564.4%17Ty MontgomeryGB · 3164.3%18Stefon DiggsMIN · 9564.2%19Allen HurnsJAX · 5763.6%20Albert WilsonKC · 6263.5%21Cooper KuppLA · 9463.3%22Terrance WilliamsDAL · 8063.2%23Robert WoodsLA · 8563.0%24Nelson AgholorPHI · 9562.8%25Jarius WrightMIN · 2562.8%26Keenan AllenLAC · 15962.7%27Seth RobertsLV · 6562.7%28Mack HollinsPHI · 2262.7%29Adam ThielenMIN · 14262.6%30Eric DeckerTEN · 8362.5%31Michael CampanaroBAL · 2762.5%32Tyler BoydCIN · 3262.4%33Alex EricksonCIN · 1662.3%34Justin HardyATL · 2962.2%35Chris ConleyKC · 1660.9%36Phillip DorsettNE · 1860.5%37Tommylee LewisNO · 1560.3%

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 · 2017 · full board

Catch rate leaderboard for the 2017 NFL season, shrunk by empirical Bayes with 90% credible intervals, ranked by the shrunk estimate.
#PlayerTeamPostargetsRawShrunk90% interval
1Golden TateDETWR12076.7%71.1%65.3%76.6%
2Ted GinnNOWR7075.7%68.3%61.3%75.0%
3Adam HumphriesTBWR8373.5%67.7%60.9%74.1%
4JuJu Smith-SchusterPITWR7973.4%67.4%60.6%74.0%
5Tyreek HillKCWR10571.4%67.1%60.8%73.1%
6Michael ThomasNOWR14969.8%66.8%61.3%72.2%
7Jarvis LandryMIAWR16169.6%66.8%61.5%72.0%
8Randall CobbGBWR9371.0%66.5%59.9%72.7%
9Jeremy Kerley◦ provisionalNYJWR2781.5%66.3%57.4%74.7%
10Danny AmendolaNEWR8670.9%66.2%59.5%72.7%
11Mohamed SanuATLWR9669.8%65.8%59.3%72.0%
12Sterling ShepardNYGWR8470.2%65.7%58.9%72.2%
13De'Anthony Thomas◦ provisionalKCWR1687.5%65.2%55.6%74.4%
14Larry FitzgeraldARIWR16267.3%65.1%59.7%70.4%
15Trent TaylorSFWR6170.5%64.9%57.4%72.1%
16Ryan GrantWASWR6569.2%64.4%57.0%71.5%
17Ty Montgomery◦ provisionalGBWR3174.2%64.3%55.5%72.7%
18Stefon DiggsMINWR9567.4%64.2%57.6%70.5%
19Allen HurnsJAXWR5768.4%63.6%56.0%71.0%
20Albert WilsonKCWR6267.7%63.5%56.0%70.7%
21Cooper KuppLAWR9466.0%63.3%56.6%69.7%
22Terrance WilliamsDALWR8066.3%63.2%56.2%69.9%
23Robert WoodsLAWR8565.9%63.0%56.2%69.7%
24Nelson AgholorPHIWR9565.3%62.8%56.2%69.3%
25Jarius Wright◦ provisionalMINWR2572.0%62.8%53.6%71.6%
26Keenan AllenLACWR15964.1%62.7%57.2%68.1%
27Seth RobertsLVWR6566.1%62.7%55.3%69.9%
28Mack Hollins◦ provisionalPHIWR2272.7%62.7%53.3%71.6%
29Adam ThielenMINWR14264.1%62.6%56.8%68.2%
30Eric DeckerTENWR8365.1%62.5%55.6%69.2%
31Michael Campanaro◦ provisionalBALWR2770.4%62.5%53.4%71.2%
32Doug BaldwinSEAWR11764.1%62.4%56.2%68.4%
33Tyler Boyd◦ provisionalCINWR3268.8%62.4%53.6%70.8%
34Alex Erickson◦ provisionalCINWR1675.0%62.3%52.5%71.7%
35Justin Hardy◦ provisionalATLWR2969.0%62.2%53.2%70.8%
36Jamison CrowderWASWR10364.1%62.2%55.7%68.5%
37Kendall WrightCHIWR9264.1%62.1%55.4%68.6%
38Jermaine KearseNYJWR10363.1%61.5%55.0%67.8%
39Taylor Gabriel◦ provisionalATLWR5164.7%61.5%53.6%69.2%
40Davante AdamsGBWR11862.7%61.4%55.2%67.4%
41Tyler LockettSEAWR7163.4%61.3%54.0%68.4%
42Braxton Miller◦ provisionalHOUWR2965.5%61.0%52.0%69.7%
43Chris Conley◦ provisionalKCWR1668.8%60.9%51.0%70.3%
44Tyrell WilliamsLACWR6962.3%60.6%53.3%67.8%
45Antonio BrownPITWR16561.2%60.5%55.0%65.9%
46Phillip Dorsett◦ provisionalNEWR1866.7%60.5%50.9%69.9%
47Tommylee Lewis◦ provisionalNOWR1566.7%60.3%50.4%69.8%
48Jordan Taylor◦ provisionalDENWR2065.0%60.3%50.7%69.5%
49Chris Godwin Jr.TBWR5561.8%60.2%52.3%67.8%
50Rishard MatthewsTENWR8760.9%60.0%53.1%66.7%
51Kelvin BenjaminBUFWR7761.0%60.0%52.8%67.0%
52Brandon Coleman◦ provisionalNOWR3762.2%60.0%51.4%68.3%
53Chester Rogers◦ provisionalINDWR3762.2%60.0%51.4%68.3%
54Brice Butler◦ provisionalDALWR2462.5%59.7%50.4%68.7%
55T.J. Jones◦ provisionalDETWR5060.0%59.2%51.1%67.0%
56Julio JonesATLWR14859.5%59.2%53.4%64.8%
57Pierre GarconSFWR6759.7%59.2%51.7%66.4%
58Jordy NelsonGBWR8959.6%59.1%52.3%65.8%
59Amara Darboh◦ provisionalSEAWR1361.5%59.1%49.0%68.8%
60Cody Latimer◦ provisionalDENWR3259.4%58.8%49.9%67.4%
61Martavis BryantPITWR8558.8%58.7%51.7%65.5%
62Geronimo Allison◦ provisionalGBWR3959.0%58.7%50.1%67.0%
63DeVante ParkerMIAWR9758.8%58.7%52.0%65.2%
64Tavon Austin◦ provisionalLAWR2259.1%58.6%49.2%67.8%
65Jakeem Grant◦ provisionalMIAWR2259.1%58.6%49.2%67.8%
66Michael Floyd◦ provisionalMINWR1758.8%58.5%48.7%68.0%
67Damiere Byrd◦ provisionalCARWR1758.8%58.5%48.7%68.0%
68Demaryius ThomasDENWR14258.5%58.5%52.6%64.2%
69Brandon LaFellCINWR8958.4%58.4%51.6%65.2%
70Odell Beckham Jr.◦ provisionalNYGWR4358.1%58.3%50.0%66.5%
71Pharoh Cooper◦ provisionalLAWR1957.9%58.3%48.6%67.7%
72Curtis Samuel◦ provisionalCARWR2657.7%58.2%49.0%67.2%
73Brenton Bersin◦ provisionalCARWR1457.1%58.2%48.1%67.9%
74Chris HoganNEWR5957.6%58.0%50.3%65.6%
75Taywan Taylor◦ provisionalTENWR2857.1%58.0%48.9%66.9%
76Marqise LeeJAXWR9757.7%58.0%51.3%64.5%
77Laquon Treadwell◦ provisionalMINWR3557.1%57.9%49.2%66.5%
78Andre Holmes◦ provisionalBUFWR2356.5%57.9%48.4%67.0%
79Kenny Golladay◦ provisionalDETWR4957.1%57.8%49.7%65.8%
80Cole BeasleyDALWR6357.1%57.7%50.1%65.2%
81Dontrelle Inman◦ provisionalCHIWR4456.8%57.7%49.4%65.8%
82Tre McBride◦ provisionalCHIWR1553.3%57.3%47.3%67.0%
83Mike WallaceBALWR9256.5%57.2%50.4%63.9%
84Marvin JonesDETWR10856.5%57.1%50.7%63.5%
85Brandin CooksNEWR11556.5%57.1%50.8%63.3%
86Chad Hansen◦ provisionalNYJWR1752.9%57.1%47.3%66.7%
87Michael CrabtreeLVWR10356.3%57.0%50.5%63.5%
88Donte Moncrief◦ provisionalINDWR4755.3%57.0%48.8%65.0%
89Devin FunchessCARWR11256.3%57.0%50.6%63.2%
90Brandon Marshall◦ provisionalNYGWR3354.5%56.9%48.1%65.6%
91Sammy WatkinsLAWR7055.7%56.9%49.5%64.1%
92Jeremy MaclinBALWR7255.6%56.8%49.4%64.0%
93Will Fuller◦ provisionalHOUWR5154.9%56.7%48.6%64.6%
94Leonte Carroo◦ provisionalMIAWR1450.0%56.7%46.6%66.5%
95Terrelle Pryor◦ provisionalWASWR3754.0%56.6%48.0%65.1%
96Deonte ThompsonBUFWR6955.1%56.5%49.1%63.8%
97Demarcus Robinson◦ provisionalKCWR3953.8%56.5%47.9%64.9%
98Willie Snead◦ provisionalNOWR1650.0%56.5%46.6%66.1%
99Russell Shepard◦ provisionalCARWR3253.1%56.4%47.5%65.2%
100Kenny StillsMIAWR10555.2%56.3%49.8%62.7%
101Kasen Williams◦ provisionalCLEWR1850.0%56.3%46.5%65.8%
102Johnny Holton◦ provisionalLVWR1850.0%56.3%46.5%65.8%
103Rashard Higgins◦ provisionalCLEWR5054.0%56.3%48.2%64.2%
104ArDarius Stewart◦ provisionalNYJWR1346.2%56.0%45.9%65.9%
105Paul RichardsonSEAWR8154.3%55.9%48.9%62.9%
106Robbie ChosenNYJWR11554.8%55.9%49.6%62.2%
107DeSean JacksonTBWR9254.4%55.8%49.0%62.6%
108Dede Westbrook◦ provisionalJAXWR5152.9%55.7%47.7%63.7%
109DeAndre HopkinsHOUWR17554.9%55.7%50.2%61.1%
110Louis Murphy◦ provisionalSFWR1747.1%55.7%45.8%65.3%
111Josh Bellamy◦ provisionalCHIWR4652.2%55.5%47.3%63.6%
112Torrey SmithPHIWR6852.9%55.3%47.9%62.7%
113Mike Williams◦ provisionalLACWR2347.8%55.2%45.8%64.5%
114Brandon Tate◦ provisionalBUFWR1442.9%55.2%45.1%65.0%
115Travis BenjaminLACWR6552.3%55.0%47.5%62.5%
116Corey DavisTENWR6552.3%55.0%47.5%62.5%
117Marquise GoodwinSFWR10553.3%55.0%48.5%61.5%
118Bennie FowlerDENWR5651.8%55.0%47.1%62.8%
119Eli Rogers◦ provisionalPITWR3650.0%55.0%46.3%63.6%
120Josh Reynolds◦ provisionalLAWR2445.8%54.5%45.1%63.7%
121Tavarres King◦ provisionalNYGWR3748.6%54.4%45.7%63.0%
122T.Y. HiltonINDWR10952.3%54.3%47.8%60.7%
123A.J. GreenCINWR14352.4%54.1%48.2%59.9%
124Dez BryantDALWR13252.3%54.0%48.0%60.0%
125Bruce EllingtonHOUWR5850.0%54.0%46.2%61.7%
126Kendrick Bourne◦ provisionalSFWR3447.1%54.0%45.1%62.7%
127Chris Moore◦ provisionalBALWR3847.4%53.8%45.2%62.3%
128Emmanuel SandersDENWR9251.1%53.8%46.9%60.5%
129Keelan ColeJAXWR8350.6%53.6%46.6%60.6%
130Roger LewisNYGWR7250.0%53.6%46.2%60.9%
131Mike EvansTBWR13851.4%53.4%47.4%59.3%
132Kenny Britt◦ provisionalNEWR4346.5%53.1%44.7%61.4%
133Amari CooperLVWR9650.0%53.0%46.2%59.7%
134Isaiah McKenzie◦ provisionalDENWR1330.8%52.9%42.8%63.0%
135Josh Malone◦ provisionalCINWR1735.3%52.8%43.0%62.5%
136J.J. NelsonARIWR6147.5%52.6%44.9%60.2%
137Michael Clark◦ provisionalGBWR1428.6%52.1%42.1%62.1%
138Travis Rudolph◦ provisionalNYGWR2236.4%51.9%42.4%61.4%
139Josh Gordon◦ provisionalCLEWR4242.9%51.5%43.1%59.9%
140Ricardo LouisCLEWR6144.3%50.8%43.1%58.5%
141Jaron BrownARIWR6944.9%50.8%43.3%58.2%
142Bryce Treggs◦ provisionalCLEWR1827.8%50.6%40.9%60.3%
143Kaelin Clay◦ provisionalCARWR2030.0%50.6%41.0%60.2%
144Josh DoctsonWASWR7844.9%50.3%43.1%57.5%
145Alshon JefferyPHIWR12346.3%50.0%43.8%56.1%
146Aldrick Robinson◦ provisionalSFWR4839.6%49.4%41.3%57.6%
147Corey ColemanCLEWR5839.7%48.6%40.8%56.4%
148Markus Wheaton◦ provisionalCHIWR1717.6%48.5%38.7%58.3%
149John BrownARIWR5637.5%47.6%39.8%55.5%
150Kamar Aiken◦ provisionalINDWR4434.1%47.3%39.1%55.7%
151Breshad Perriman◦ provisionalBALWR3528.6%46.5%37.8%55.3%
152Zay JonesBUFWR7436.5%45.6%38.4%52.9%

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).