Writers of Pro Football Prospectus 2008

22 Aug 2008

FO on ESPN: Catch This

This week's ESPN column is about Wes Welker and his sublime catch rate from last year.

Posted by: Bill Barnwell on 22 Aug 2008

14 comments, Last at 25 Aug 2008, 7:04pm by Matt Saracen - QB1 - Dillon Panthers

Comments

1
by Josh (not verified) :: Fri, 08/22/2008 - 5:01pm

I think it would be nice to have a "yards per pass" stat for WR's, like we do with QB's.

2
by Reinhard (not verified) :: Fri, 08/22/2008 - 5:10pm

One thing that stuck out about Welker last year was how comfortable he seemd with the ball in his hands. He quickly decided whether he should initiate contact for extra yards or avoid contact for extra yards; he rarely seemed to make a mistake and leave yards on the field.

3
by Temo (not verified) :: Fri, 08/22/2008 - 5:13pm

The stats for WRs on the site lists intended passes only, so you can divide yards by that number to get your number.

4
by sophandros (not verified) :: Fri, 08/22/2008 - 7:28pm

2: I think a lot of that has to do with the system that he played in at Texas Tech. Mike Leach drills the fundamentals of catching and advancing the ball with his WRs. There was an article about him written by Michael Lewis a couple of years ago. Interesting stuff.

5
by Joe T. (not verified) :: Fri, 08/22/2008 - 9:08pm

But just what would a "yards per pass attempt" stat reveal about a WR's performance?

It seems to me that there are too many variables contributing to an average of yards per pass attempt. Even though yards per completion is a flawed stat (because I believe it takes into consideration YAC) yards per pass attempt includes all the uncatchable junk and knuckleballs that a QB would chuck at the receiver.

The most useful stats are those that isolate as many variables as possible...I can't see that yards per intended pass for WRs would be useful.

6
by MC2 (not verified) :: Sat, 08/23/2008 - 12:43am

"The most useful stats are those that isolate as many variables as possible…I can’t see that yards per intended pass for WRs would be useful."

It probably wouldn't be worth much, but it would still be better than catch rate, IMO.

7
by remiel (not verified) :: Sat, 08/23/2008 - 3:05am

"One thing that stuck out about Welker last year was how comfortable he seemd with the ball in his hands. He quickly decided whether he should initiate contact for extra yards or avoid contact for extra yards; he rarely seemed to make a mistake and leave yards on the field."

Ironically i was considering that very aspect of his play a few hours ago when he got absolutely creamed after ill-advisedly choosing to run back a punt with an Eagles defender pelting towards him and no space to manoeuvre because he was next to the sideline. Took him out of the game.

8
by JMM (not verified) :: Sat, 08/23/2008 - 10:16am

"...as a player's yards per reception go up, catch rate tends to go down."...

Can you explore this a bit more? What is the (math) relationship? Who are the outliers?

9
by Josh (not verified) :: Sun, 08/24/2008 - 2:19am

Out of Welker's 112 receptions last year, 21 were at or behind the line of scrimmage, and another 79 were 10 yards or less beyong the line of scrimmage. Only 12 receptions made 11 or more yards from the line of scrimmage. Does FO have catch rates broken down by where the ball was caught?
Considering where Welker caught his passes, I think that sort of statistic is needed if you want to show that Welker did something special last season.

10
by Israel (not verified) :: Sun, 08/24/2008 - 3:57am

The guy after Welker on that list looks pretty impressive as well, especially when you consider that he had a rookie throwing to him.

11
by The McNabb Bowl Game Anomaly (aka SJM) (not verified) :: Sun, 08/24/2008 - 10:36am

Re: 9

FO doesn't quite break it down that far (unless you buy the game-charting data). Their categories in Pro Football Prospectus are short (5 yards or less in the air), mid (6-15), deep (16-25), and bomb (26+).

According to PFP 08, Welker was targeted 65% short, 29% mid, and 6% deep. This only strengthens your point, because 65% of passes traveling 5 yards or less is extraordinarily high.

12
by Rich Conley (not verified) :: Mon, 08/25/2008 - 11:03am

"Exactly 120 receivers besides Welker had 50 attempts thrown to them in consecutive seasons for different teams (since 1995). Sixty-one of them saw their catch rate decline, and 59 saw it improve. The average change was -0.5 percent; in other words, basically nothing."

This totally disregards the biggest point of contention we had with your original analysis: QBs matter. Doing the analysis without considering the quality of the QBs involved is useless. We expected Welker to improve because he went from a shitty QB to an excellent one. Probably half of 120 receivers mentioned went from a decent QB to a bad one.

13
by Matt Weiner (not verified) :: Mon, 08/25/2008 - 1:24pm

What I'd want to know is, what was the average of the absolute value of the change, and how did it compare with the average absolute value of the change of receivers who stayed on the same team?

14
by Matt Saracen - QB1 - Dillon Panthers (not verified) :: Mon, 08/25/2008 - 7:04pm

Joe T - #5

A yards per attempt figure would then make the WR stats look a bit more like the CB stats - where you can see success rate AND yards per attempt - as well as the average depth of the passes thrown to them too.

I'm sure it wouldn't be too hard to add those stats to WRs as well, but for now the depth breakdown is a good indicator of a receivers usage.