When Does Yards After Catch Stabilize for Wide Receivers?

I’ve been watching the NFL for 30+ years, and there doesn’t seem to be a game that goes by without the commentary team citing a WR’s yards after catch (YAC) at some point during the broadcast. But should they? Today’s post seeks to answer that question.

Methods

They’re the same as always. To save some space, I’ll just remind everyone that this is a split-half reliability test, the applied-to-football methodological details of which you can read about here — and extrapolate to WR YAC — if you’re so inclined.

Results and Discussion

Below is the familiar stability table, this time for WR YAC:

GamesnrR2 = 0.50Avg YACObs 5.00 YAC
Wtd Average314.384.69
46250.09384.264.63
84290.20314.324.66
123080.32264.364.68
162200.36284.444.72
201640.44254.474.74
241140.47274.654.82
28790.52264.614.81
32570.62204.604.80
36490.64214.604.80

I haven’t done so in a while, so here’s how to read the table if you’re a newcomer to these posts. Each row represents a player-game group. The third row, for example, has “12” in the “Games” column because it includes all WRs that played at least 24 games with the same team since 2006, and I’ve both a) randomly selected 24 games if they played more than 24, and b) randomly split the 24 games into two sets of 12. From there,

  • the “n” column represents the number of WRs in each group,
  • the “r” column shows the correlation between each group’s two sets of games,
  • the “R2 = 0.50″ column mathematically converts the group’s correlation into the number of games it takes for WR YAC to reflect 50% skill and 50% luck (aka the stabilization point),
  • the “Avg YAC” column shows the group’s average YAC,
  • and the “Obs 5.00 YAC” column regresses a 5.00 Actual YAC toward’s the group’s average YAC based on how long it takes for that group’s YAC to stabilize.

The bottom row labeled “Wtd Average” reveals the table’s take-home finding. In this case, it shows us that WR YAC takes 31 games to stabilize. And given an average of 3.2 receptions, 5.3 targets, and 26.1 routes run per game in my sample, this translates to about 100 receptions, 163 targets, and 801 routes. In what may come as a surprise, this is nearly twice as long as it takes TE YAC to stabilize (22 games, 58 receptions, 84 targets, and 408 routes).

What’s more, if we compare these findings to those for WR stats I’ve analyzed previously, we can conclude that YAC is considerably less reliable than average depth of target (aDOT, aka air yards), targets per route run (TPRR), receptions per route run (RPRR), and yards per route run (YPRR), while only being slightly more reliable than touchdowns per route run (TDPRR) and, duh, yards per target (YPT).

Two other conclusions we can draw from today’s result respect the questions I posed at the beginning of this YAC series. First, focusing on the above comparison with WR aDOT, it turns out that YAC takes nearly eight times longer to stabilize (i.e., 31 games vs. 8 games). Therefore, the choice of which one to use for WR prediction is a no-brainer. Second, given that WR YAC takes longer to stabilize than TE YAC, whereas TE aDOT takes longer than WR aDOT, we can now answer my second question: The rank order of reliability between positions for YAC is not the same as the rank order of reliability for aDOT.

Finally, converting Actual YAC into True YAC, below are the standings for all WRs that played at least 25 percent of snaps in 2016 per Pro Football Focus:

PlayerTmGRecActual YACRkTrue YACRk
Golden TateDET16916.845.51
Jarvis LandryMIA16946.755.52
Adam HumphriesTB15557.125.33
Taylor GabrielATL12358.115.34
Tyrell WilliamsSD16696.565.25
Quincy EnunwaNYJ16586.285.06
Cordarrelle PattersonMIN16526.375.07
Randall CobbGB12606.195.08
Mike WallaceBLT16725.8144.99
Jamison CrowderWAS16675.7154.810
Odell Beckham Jr.NYG161015.3254.811
Marqise LeeJAX16635.6164.812
Amari CooperOAK16835.4184.813
Chris HoganNE15386.194.814
Davante AdamsGB16755.4184.815
Willie SneadNO15725.4184.716
Michael A. ThomasNO15925.2294.717
Brandon LaFellCIN16645.4184.718
Allen HurnsJAX11355.9124.719
Eddie RoyalCHI9335.9124.720
Doug BaldwinSEA16945.1314.721
Cody CoreCIN6176.934.722
Julian EdelmanNE16985.0364.623
Cole BeasleyDAL16755.1314.624
Julio JonesATL14835.0364.625
Travis BenjaminSD14475.3254.626
Tyler LockettSEA14415.4184.627
Seth RobertsOAK16385.4184.628
Mohamed SanuATL15595.1314.629
Quinton PattonSF13375.4184.630
Breshad PerrimanBLT15335.3254.531
Albert WilsonKC14315.3254.532
DeSean JacksonWAS15564.9404.533
Sammie CoatesPIT11215.5174.534
J.J. NelsonARZ14345.1314.535
Victor CruzNYG14395.0364.536
Malcolm MitchellNE11325.1314.537
Brandin CooksNO15784.7424.538
Bennie FowlerDEN10116.194.539
Tyreek HillKC16614.7424.440
Tavon AustinLA15584.7424.441
Ricardo LouisCLV7185.2294.442
Will FullerHST14474.7424.443
Pierre GarconWAS16794.5484.444
Cameron MeredithCHI14664.5484.445
Jordan TaylorDEN11164.9404.446
Phillip DorsettIND15334.6474.447
Devin FunchessCAR14234.7424.448
Jaron BrownARZ6115.0364.349
Kenny StillsMIA16424.5484.350
Adam ThielenMIN16694.4524.351
Kenny BrittLA15684.4524.352
DeVante ParkerMIA15564.4524.353
Paul RichardsonSEA11214.5484.354
Rod StreaterSF13184.4524.355
Marvin JonesDET15554.3564.356
Eli RogersPIT13484.3564.357
Corey BrownCAR15274.3564.358
Andre RobertsDET13144.3564.359
Kendall WrightTEN11294.2604.360
Andrew HawkinsCLV16334.2604.361
Brian QuickLA16414.2604.362
Aldrick RobinsonATL13204.1634.263
Dorial Green-BeckhamPHI14364.1634.264
Bryan WaltersJAX8244.0674.265
A.J. GreenCIN10664.1634.266
Steve L. SmithBLT14704.1634.267
Charone PeakeNYJ11193.8724.268
Jordan NorwoodDEN14213.7744.269
Braxton MillerHST9153.5844.270
Cody LatimerDEN782.91034.271
Sterling ShepardNYG16654.0674.272
Charles D. JohnsonMIN13203.6804.273
Ryan GrantWAS1192.91034.274
Jeff JanisGB10113.1964.275
Tyler BoydCIN16543.9704.176
Jordy NelsonGB16974.0674.177
James WrightCIN10132.91034.178
Russell ShepardTB12233.4854.179
Walt PowellBUF7142.91034.180
Dontrelle InmanSD16583.8724.181
Terrance WilliamsDAL15443.7744.182
Brittan GoldenARZ781.91234.183
Antonio BrownPIT151063.9704.184
Torrey SmithSF12203.1964.185
Kamar AikenBLT16293.4854.186
Alshon JefferyCHI12523.7744.187
Chris ConleyKC16443.6804.188
Kelvin BenjaminCAR16633.7744.189
Ted GinnCAR16543.6804.090
Danny AmendolaNE11233.0994.091
Jermaine KearseSEA16413.4854.092
Roger LewisNYG970.31304.093
Stefon DiggsMIN13843.7744.094
Andre HolmesOAK14142.11194.095
Josh BellamyCHI12192.61134.096
Demaryius ThomasDEN16903.7744.097
Chester RogersIND9192.51154.098
Jaelen StrongHST6141.91234.099
Corey ColemanCLV10333.0994.0100
T.Y. HiltonIND16913.6804.0101
Justin HardyATL11212.41184.0102
Nelson AgholorPHI15363.0993.9103
Jeremy KerleySF16643.4853.9104
Rishard MatthewsTEN16653.4853.9105
Brandon ColemanNO13262.61133.9106
Anquan BoldinDET16673.4853.9107
Marquise GoodwinBUF14292.71113.9108
Jeremy MaclinKC12443.0993.9109
Brandon MarshallNYJ15593.2933.9110
Vincent JacksonTB5151.21283.9111
Donte MoncriefIND9302.51153.9112
Deonte ThompsonCHI9222.01213.9113
John BrownARZ14392.81083.9114
Cobi HamiltonPIT11171.41273.9115
Brice ButlerDAL14161.21283.9116
Emmanuel SandersDEN15793.3923.8117
Robby AndersonNYJ14422.81083.8118
Larry FitzgeraldARZ161073.4853.8119
Jordan MatthewsPHI14733.2933.8120
Dez BryantDAL12502.91033.8121
DeAndre HopkinsHST16783.2933.8122
Michael CrabtreeOAK16893.1963.7123
Sammy WatkinsBUF8281.71253.7124
Robert WoodsBUF13512.51153.7125
Tajae SharpeTEN16412.11193.6126
Michael FloydARZ12331.71253.6127
Terrelle PryorCLV16772.81083.6128
Allen RobinsonJAX16732.71113.6129
Mike EvansTB16962.01213.2130
4.3

As seems to be the theme of this post, I think the most valuable information here comes from juxtaposing it with the True WR aDOT standings I posted previously. If we focus on the Top 10 above, we find (unsurprisingly) the usual suspects in a lineup based on “Who runs a bunch of short, zone-attacking routes?” To wit, Golden Tate ranks 1st in True YAC, but ranked 116th in True aDOT, Jarvis Landry ranked 127th in True aDOT, Adam Humphries ranked 129th, Quincy Enunwa ranked 102nd, Cordarelle Patterson ranked 130th, Randall Cobb ranked 125th, and Jamison Crowder ranked 117th. The predictive value, then, comes — as it usually does — from the outliers: Taylor Gabriel ranked a respectable 69th in True aDOT, Tyrell Williams ranked 48th, and Mike Wallace ranked 49th. It seems to me that these three are WRs who were targeted on deeper routes and made the most of it after catching said target. Of course, Gabriel has a new offensive coordinator and Williams has a new coaching staff altogether, so Wallace has the best combination of the three with respect to statistical reliability and structural continuity.

In this context, the flip side is also informative. Among the Top 10 in True aDOT, there also exist the usual suspects vis-a-vis YAC: Robby Anderson’s True aDOT ranked 3rd, but his True YAC ranks 118th. Similarly, Mike Evans True aDOT ranked 5th, but his True YAC ranks 130th, Marquise Goodwin’s rankings were 6th and 108th, Roger Lewis’ were 7th and 93rd, Josh Bellamy’s were 8th and 96th, and Terrelle Pryor’s were 10th and 128th. Analogous to the previous paragraph, this means that Sammie Coates (No. 1 True aDOT, No. 34 True YAC), J.J. Nelson (2nd, 35th), DeSean Jackson (4th, 33rd), and Will Fuller (9th, 43rd) are the outliers. But, as before, these four have to be pared down further based on their situations heading into 2017. Coates is affected by Martavis Bryant’s return from the cloud and Jackson is affected by switching teams (although his new offense also loves the deep ball). That leaves Nelson and Fuller as the True aDOT WRs that are also respectable with respect to True YAC.

DT : IR :: TL : DR

WR YAC takes eight times longer to stabilize than WR aDOT (aka air yards) and twice as long as TE YAC. This suggests that, when you hear YAC cited during an NFL broadcast, your statistical bullshit detector should hit DEFCON 1. Instead, our search for “true” WR skill should focus on players that bridge the gap between the two “true” stats. Heading into 2017, that means Mike Wallace, J.J. Nelson, and Will Fuller (and maybe DeSean Jackson).

Email to someoneShare on Facebook0Tweet about this on TwitterShare on Reddit0
Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *