## Who Gets Credit for YAC?

In the last post, I discussed Yards After Catch (YAC) and introduced its complement, Air Yards. In this post, I'll take a closer look at YAC and how much credit a QB should get for YAC compared to his receivers. Ultimately, these results will contribute to an improved passer rating system, an effort I began here.

When we break down pass yards and how credit should be shared by the passer and receiver, we could start by saying that the receiver should get credit for the Yards After Catch and the QB should get credit for the Air Yards.

But a plausible alternative is that the QB's accuracy contributes to YAC. His accuracy, and his ability to read defenses and find open receivers, would logically allow him to contribute significantly to YAC. An accurate and aware QB would be able to lead his receivers, hitting them in stride and steering them to where they can best evade a tackle.

If a QB is responsible for some of his team's YAC, his accuracy and ability to read open receivers should correlate well with YAC. To test the theory that a QB's abilities contribute to YAC, I conducted a simple regression. The correlation of accuracy and YAC should be positive, strong, and significant. Additionally, the r-squared of the model should be relatively high.

As measures of a QB's accuracy and ability to read open receivers, I used pass completion percentage and interception rate. The dependent variable was YAC/completion. The data comes from every team between the 2002 and 2006 regular seasons (n=160). The results of the regression are listed below.

 VARIABLE COEFFICIENT STDERROR T STAT P-VALUE Comp % 0.01 0.01 0.66 0.51 Int Rate -4.69 6.02 -0.78 0.44 r-squared 0.01

Neither completion percentage (p=0.51) nor interception rate (p=0.44) was statistically significant. The r-squared for the model was nearly zero. These results yield evidence that, for passes as a whole, the QB makes little contribution to YAC. Yards after Catch, therefore, belong mostly to the receiver.

This result is important because it means a valid QB passer rating should exclude YAC. In the next post, I'll look at QB performances from 2006 and who were the downfield threats and who were the "dink and dunk" throwers. Ultimately, I'll apply what I learned about Air Yards and YAC to an improved passer rating formula.

### 9 Responses to “Who Gets Credit for YAC?”

1. Tarr says:

I think a really useful angle to further investigate would be year-to-year correlations of YAC. Specifically:

1) Do certain receivers have consistently good YAC (e.g. TO)? My guess is yes.

2) Can we largely explain year-to-year correlations of a QB's YAC by what receivers they complete passes to? If you could create a model that shows that QBs have almost no year-to-year correlation after correcting for their receivers, then that really drives the point home.

I think this material would make for a great guest column on FO.

2. Anonymous says:

Another thing to consider is possibly the QB's arm strength. I had done a breakdown on the Eagles Rac with McNabb and with Garcia, and the differences were massive. someone had thought that Garcia was more accurate, thus the Rac should increase. But when looking at the breakdown, Donte Stallworth I believe was the only receiver with anywhere near a close RAC with both QB's. The rest had massively higher rac yards from McNabb. Westbrook's I believe dropped in half. Maybe looking at WR's numbers in the same season with different QB's and what those QB's strengths and weaknesses are could help determine how much is QB related (like if the strong arm holds true) and how much is receiver related.

3. Brian Burke says:

Slash-Since I made this post I've learned a lot about YAC. QBs that tend to check down a lot, or rely on RB screens/flares get lots of YAC. Those types of plays do not require a very skilled QB.

The QBs who don't get a lot of YAC are those with the stronger arms--Manning, Palmer, Romo, etc. They throw downfield passes and out routes near the sidelines that don't tend to generate YAC.

Garcia was throwing downfield, but McNabb was dumping off.

One other note. Super mobile guys like Vince Young and M. Vick don't rely on checkdown/outlet receivers. Their 3rd option is to run, so they tend to generate very little YAC.

4. Anonymous says:

Actually just watching the games, the thing that changed when McNabb went out was the long ball. The Air yards were higher for Reggie Brown with McNabb than with Garcia (also the plays of 20+ and 40+ were higher), same with Donte Stallworth, Greg Lewis, Hank Baskett, LJ Smith, Westbrooks air yards went up slightly (from -.5 to 1.9 per catch. In the end with the receivers from what I saw in the game was the underneath stuff that Garcia threw would allow usually a relatively low yac, though possibly higher than McNabbs, but the deep balls that went over the D that only really came from McNabb would give those plays that were 30-40 yards in the air, and another 30 or so yac that put the receivers yac from him over that of Garcia. Also just from watching the games, while Garcia was throwing to the backs just as much as McNabb, I believe the reason they were getting a much higher Yac because the secondary was forced to play off more due to the threat of the long ball with McNabb that was no longer there with Garcia (also evidenced by his air yards being higher along with his yards per completion).

Just a question here. Wondering if you have found any corrollary between a vertical passing offense and a horizontal one. Granted they are mixed up pretty well now with the WCO's ideas slipping into about any offense run in the league, but a comparison possibly between the run&shoot or Dallas of the 90's type teams vs. the Packers and 49ers of earlier.

5. Anonymous says:

I guess maybe another thing I would be curious about (wouldn't mind finding it out myself but not sure where to find these numbers) is the rac differences based on air yard differences.

For example the average Yac on a pass that is thrown 10 yards downfield vs. that of one thrown 20 yards. I guess things like the deep outs would have an affect here, with a strong armed QB as you mentioned. But it would help to see if guys like Andre Johnson who went from one of the least accurate QB's in the deep ball to a strong armed accurate one will continue to have a career high in Yac or not.

6. Brian Burke says:

Slash-To be honest I try hard not to watch the Eagles. So I'll defer to your first hand experience.

The McNabb/Garcia issue is probably one where you need to document every play.

That's a good point about the bomb passes that get behind the defense, they have potential for a lot of YAC. But RBs all have a lot more YAC than WRs as a rule. My theory is that WRs get a lot of catches and get knocked out of bounds right away, or they catch passes over the middle in front of the safeties who knock them down pretty quickly.

You would love KC Joyner's "Scientific Football." Do a search for him. I've got his book and it is full of charts of routes, pass depths, all that stuff. It might have the McNabb/Garcia data we'd need to draw some conclusions.

Interestingly, McNabb and Garcia are right next to each other in my QB rating so far this year.

7. Anonymous says:

Your model is pretty pathetic. The dependent and independent variable seem irrelevant in the method they are being tested.

Common sense indicates the QB affects YAC

8. Brian Burke says:

Tool-I didn't say the QB gets zero credit, but that the receiver is far more critical. Learn to read.

9. Anonymous says:

Very interesting. I wonder if it would make a difference if you used a completion percentage stat that takes into account the number of times the QB intentionally threw the ball away, spiked it, was hit while throwing, or had his passes dropped if it would show anything. Pro Football Focus has gathered interesting stats on this. Probably not much based on your results. I understand your argument, but wonder if adjusting QB rating based on it might be pretty tricky. It isn't inherently bad to be a dink and dunk QB if it is successful, so even though QB's don't have much control over YAC is it fair to punish them for it? For instance, clearly when a QB throws a screen that gets a lot of YAC he sees that the runner has open field ahead of him, so it doesn't seem right to count this intelligent decision as less valueble than instead throwing a pass for equal yardage that is downfield.

### Leave a Reply

Note: Only a member of this blog may post a comment.