Projecting 1st Round WRs: Completion Percentage and YPA

by John Morgan on Mar 11, 2008 1:12 PM PDT in 2008 NFL Draft Comment 10 comments

I have to do this freehand, working through the play by play and tallying everything in a notebook. I guess that would make these stats "exclusive". Anyway, here's the probable first round wide receivers. All stats reflect the last two seasons. The results might be surprising.

Devin Thomas
Games Against Top 25 Opponents: 4
Completion Percentage: 56%
Yard Per Attempt: 8.08

Malcolm Kelly
Games: 1
%: N/A
YPA: N/A

Early Doucet
Games: 9
%: 76%
YPA: 9.6

DeSean Jackson
Games: 8
%: 63%
YPA: 9.5

Limas Sweed
Games: 4
%: 55%
YPA: 6.3

Kelly faced one top 25 defense and recorded an incompletion. His schedule was by far the easiest, full of patsies. Doucet and Jackson are the clear leaders. Doucet is the only player to never get "shutdown". Jackson's otherwise sparkling record is pocked by his performance against Oregon State: 4/11, 5 yards. Remember, this isn't a foolproof way to tell who will succeed in the pros, but if you can't get open in college then...

I find this fun and informative. I'll revisit this stat another day mining for sleepers.

0 recs | Comment 10 comments

Story-email Email Printer Print

Comments

Predictiveness?

Probably the best way to assess the value of these statistics is to see how well they predicted future NFL performance for WRs drafted in previous years. Any chance we could get some of those numbers as well? Or maybe there's been previous stuff done saying that YPA is strongly associated with pro performance?

by cyberwulf on Mar 11, 2008 1:39 PM PDT reply actions 0 recs

Well, I only have two years of FEI...

to work with, and had to consult multiple sources to even find the complete play by play for the past two seasons. I could go back and find out how 2007's first round picks did, but that would only be one year of college stats and one year of pro stats to work with. So, no, I'm not sure how well this might predict future performance and am pretty sure I won't be able to tell for a few years, but it does give a better idea of how well these players actually performed in college.

by John Morgan on Mar 11, 2008 1:46 PM PDT up reply actions 0 recs

FWIW: Top Rookie Dwayne Bowe's 2006

Games: 5
%: 70
YPA: 9.39

by John Morgan on Mar 11, 2008 1:56 PM PDT up reply actions 0 recs

Calvin Johnson

Games: 4
%: 47%
YPA: 6.9

Reggie Ball's completion percentage was just 44.4, though, so he should get some leeway for that. The rest of the QBs were reasonably comparable. Also, Johnson's play by play data against Miami (6 receptions 79 yards) does not match his box score data (5/68). I went with the play by play, but that's weird.

by John Morgan on Mar 11, 2008 2:29 PM PDT up reply actions 0 recs

Interesting

YPA seems to have some potential as a useful predictive statistics. I wonder, though (particularly given your comment on Reggie Ball) how much it's influenced by the system the player's in? We know that's a problem when comparing QB passing numbers from option vs. pro-style offenses, but it might be less problematic for receivers.

by cyberwulf on Mar 11, 2008 3:56 PM PDT reply actions 0 recs

Early Doucet please. He seems totally off the radar after a nicked up year in college, could be a potential steal.

by Nate Dogg on Mar 11, 2008 5:05 PM PDT reply actions 0 recs

great analysis

might be useful to list a qb's completion percentage for the year and in the games vs top 25 opponents, just for comparison.

what do you think about expanding the sample size to top 35 or 40 defenses? obviously taking it too far defeats the purpose of limiting it at all, but it would give us a much bigger set of data to start with.

also, it might be interesting to compare the top 25 numbers against season completion % and YPA to determine if these receivers are just beating up on inferior opposition.

by minor threat on Mar 11, 2008 5:11 PM PDT reply actions 0 recs

Was Kelly just NA in games?

Or did OU seriously only play one top 25 team? Kelly did miss some games as I recall. (Not that staying healthy shouldn't factor into things.)

Oh, and Reggie Ball (G Tech), Brandon Cox (Auburn), and Blake Mitchell (S. Car.) have to be the three worst college football qb's I've seen in the last five seasons (on teams that are balanced run/pass offenses).

"Those who fear disorder more than injustice inevitably produce more of both." -- Rev. William Coffin

by dcrockett17 on Mar 12, 2008 8:21 AM PDT reply actions 0 recs

He missed West Virginia this year...

only recorded an incompletion against West Virgina last year and the rest of his schedule didn't qualify. Oklahoma's schedule was really easy. I still think he looks like a fine talent, but his stock might be inflated by inferior competition.

by John Morgan on Mar 12, 2008 9:20 AM PDT up reply actions 0 recs

I think it'd be more helpful if...

You scaled college defenses where 100 is average, then applied that as a percentage towards each receiver's production. This way you can take all games a player played into account.

As pointed out elsewhere, perhaps it's best to also apply the above logic to QB's for their stats minus that receiver's numbers.

Then apply this statistical analysis to current NFL players and see if anything provided by the combine shows a tendancy for a player to excel in the NFL versus another player.

I did this once for college basketball players trying to see how shot-blocking skills transferred to the NBA. I found that a player who blocks a lot of shots in college but has a shorter reach won't have that skill transfer was well to the NBA as those with a higher reach.

by Dobbs on Mar 12, 2008 12:30 PM PDT reply actions 0 recs