Writers of Pro Football Prospectus 2008

Most Recent FO Features

MasonTre14.jpg

» Word of Muth: Rams' Big Win

Ben Muth's tape analysis shows that St. Louis rookie Greg Robinson is still a work in progress -- as is Seattle's defense.

31 Oct 2013

Varsity Numbers: Revisiting the RYPR

by Bill Connelly

Last year around this time, I took a stab at a receivers stat that took simple Yards Per Target (a lovely stat) to a new level. It was flawed, and I vowed to make improvements to the measure, but a year later, that hasn't happened yet. So to rekindle the conversation, let's take another look at what I called RYPR last year.

Below, you will find a measure that attempts to answer the following questions about a given pass-catcher:

  • 1) How much do you produce?
  • 2) How important are you to your team's passing game?
  • 3) How good is the passing game to which you are important?
  • 4) And how much is the forward pass featured in your team's offense?

The idea was to simply multiply the following four factors together: a player's Yards Per Target, his Target Rate, his team's Passing S&P+, and his team's pass rate. Target Rate x Yards Per Target x Passing S&P+ x Pass Rate = RYPR.

Simplified, it's basically this:

RYPR = (receiving yards / total team plays) * Passing S&P+

Again, it's not bad. It gives a single player too much credit for his team's Passing S&P+ overall, but the initial goal here isn't to produce the perfect receiver measure -- it's simply to answer the four questions listed above. As soon as I give myself the time, we'll take this further.

For now, though, here is the RYPR Top 100 for 2013 thus far. You can find all 899 FBS players with at least 10 targets rated here.

Team Player Targets Catches Yards Catch
Rate
Yds Per
Target
Target
Rate
Target
No.
Passing
S&P+
Pass
Rate
RYPR Rk
Baylor Antwan Goodley 51 39 904 76.5% 17.7 23.9% 1 209.9 46.9% 417.8 1
Baylor Tevin Reese 47 33 824 70.2% 17.5 22.1% 2 209.9 46.9% 380.8 2
Florida State Rashad Greene 56 39 690 69.6% 12.3 28.6% 1 172.8 53.2% 323.9 3
LSU Odell Beckham Jr. 70 48 1009 68.6% 14.4 30.8% 2 147.3 46.4% 303.5 4
Texas A&M Mike Evans 60 48 1101 80.0% 18.4 20.1% 2 149.9 52.8% 292.6 5
Florida State Kenny Shaw 39 31 574 79.5% 14.7 19.9% 2 172.8 53.2% 269.5 6
Oregon State Brandin Cooks 108 85 1256 78.7% 11.6 27.9% 1 119.0 69.8% 269.5 7
LSU Jarvis Landry 76 58 882 76.3% 11.6 33.5% 1 147.3 46.4% 265.3 8
Michigan Jeremy Gallon 63 45 831 71.4% 13.2 38.7% 1 124.0 39.4% 249.0 9
Arizona State Jaelen Strong 69 45 685 65.2% 9.9 24.6% 1 153.7 60.6% 227.9 10
Pittsburgh Devin Street 61 35 645 57.4% 10.6 35.5% 1 127.4 46.8% 223.4 11
Wisconsin Jared Abbrederis 68 43 752 63.2% 11.1 39.1% 1 108.1 47.6% 222.2 12
Colorado Paul Richardson 87 50 914 57.5% 10.5 39.7% 1 100.8 52.0% 218.7 13
Oregon Josh Huff 56 38 703 67.9% 12.6 23.6% 2 149.7 47.1% 209.3 14
Florida State Kelvin Benjamin 39 23 430 59.0% 11.0 19.9% 3 172.8 53.2% 201.9 15
Ball State Willie Snead 115 65 1057 56.5% 9.2 33.4% 1 109.6 58.6% 197.3 16
San Jose State Chandler Jones 57 40 805 70.2% 14.1 22.2% 1 114.3 54.1% 193.7 17
Indiana Cody Latimer 62 40 640 64.5% 10.3 23.0% 1 137.5 59.3% 193.1 18
Stanford Ty Montgomery 59 39 619 66.1% 10.5 34.5% 1 129.4 41.1% 192.7 19
Kansas State Tyler Lockett 54 39 586 72.2% 10.9 33.3% 1 121.8 43.7% 192.5 20
Houston Deontay Greenberry 88 58 843 65.9% 9.6 33.2% 1 108.1 55.6% 191.2 21
Duke Jamison Crowder 94 60 769 63.8% 8.2 38.5% 1 122.6 48.8% 188.7 22
Miami Allen Hurns 44 28 506 63.6% 11.5 24.4% 1 141.1 47.4% 187.8 23
Vanderbilt Jordan Matthews 96 66 890 68.8% 9.3 40.0% 1 97.1 51.0% 183.7 24
Central Florida Breshad Perriman 32 22 507 68.8% 15.8 15.5% 3 141.0 52.9% 183.6 25
Team Player Targets Catches Yards Catch
Rate
Yds Per
Target
Target
Rate
Target
No.
Passing
S&P+
Pass
Rate
RYPR Rk
Pittsburgh Tyler Boyd 50 34 526 68.0% 10.5 29.1% 2 127.4 46.8% 182.2 26
Oregon Bralon Addison 62 38 609 61.3% 9.8 26.2% 1 149.7 47.1% 181.3 27
Ohio Donte Foster 73 50 721 68.5% 9.9 30.2% 1 118.8 50.4% 178.4 28
Boston College Alex Amidon 57 46 585 80.7% 10.3 41.0% 1 110.2 38.5% 178.4 29
Wake Forest Michael Campanaro 99 65 792 65.7% 8.0 39.6% 1 99.2 55.8% 175.5 30
Clemson Sammy Watkins 80 58 813 72.5% 10.2 26.0% 1 127.5 51.9% 174.8 31
East Carolina Justin Hardy 84 67 765 79.8% 9.1 27.5% 1 112.6 62.0% 174.5 32
Notre Dame TJ Jones 67 44 631 65.7% 9.4 26.9% 1 129.4 52.3% 171.5 33
Utah Dres Anderson 64 33 642 51.6% 10.0 26.9% 1 121.7 50.1% 164.6 34
Fresno State Davante Adams 93 72 811 77.4% 8.7 27.0% 1 99.9 69.5% 163.6 35
Texas Tech Jace Amaro 97 64 861 66.0% 8.9 24.0% 1 117.1 65.1% 162.0 36
North Carolina Eric Ebron 50 35 599 70.0% 12.0 20.0% 1 127.9 52.7% 161.6 37
Missouri L'Damian Washington 53 33 635 62.3% 12.0 21.0% 2 123.9 51.1% 159.5 38
Central Florida Rannell Hall 41 28 437 68.3% 10.7 19.9% 1 141.0 52.9% 158.3 39
Penn State Allen Robinson 86 55 878 64.0% 10.2 34.8% 1 82.0 54.3% 158.2 40
Ball State Jordan Williams 79 56 792 70.9% 10.0 23.0% 2 109.6 58.6% 147.9 41
Michigan Devin Funchess 36 23 492 63.9% 13.7 22.1% 2 124.0 39.4% 147.4 42
Maryland Stefon Diggs 56 34 587 60.7% 10.5 21.8% 1 123.4 52.0% 146.7 43
Louisville Damian Copeland 50 33 486 66.0% 9.7 20.1% 1 136.7 54.9% 146.5 44
Florida State Nick O'Leary 22 17 307 77.3% 14.0 11.2% 4 172.8 53.2% 144.1 45
UL-Lafayette Jamal Robinson 56 33 540 58.9% 9.6 32.4% 1 110.2 41.6% 143.2 46
Old Dominion Larry Pinkard 32 24 271 75.0% 8.5 23.7% 1 100.1 71.2% 143.1 47
UNLV Devante Davis 84 52 802 61.9% 9.5 29.2% 1 100.3 51.0% 142.5 48
Ohio State Devin Smith 51 35 524 68.6% 10.3 23.8% 1 129.6 43.7% 138.6 49
Ohio Chase Cochran 33 25 557 75.8% 16.9 13.6% 3 118.8 50.4% 137.8 50
Team Player Targets Catches Yards Catch
Rate
Yds Per
Target
Target
Rate
Target
No.
Passing
S&P+
Pass
Rate
RYPR Rk
Louisville DeVante Parker 35 27 457 77.1% 13.1 14.1% 2 136.7 54.9% 137.7 51
Ohio State Corey Brown 49 37 520 75.5% 10.6 22.9% 2 129.6 43.7% 137.5 52
Indiana Kofi Hughes 53 28 454 52.8% 8.6 19.6% 2 137.5 59.3% 137.0 53
Georgia Chris Conley 53 30 418 56.6% 7.9 24.9% 1 144.0 47.3% 133.6 54
SMU Jeremy Johnson 99 64 752 64.6% 7.6 26.7% 1 96.0 67.6% 131.6 55
North Texas Brelan Chancellor 47 36 619 76.6% 13.2 19.0% 2 108.1 48.3% 130.9 56
Oregon State Richard Mullaney 59 38 606 64.4% 10.3 15.2% 2 119.0 69.8% 130.0 57
California Chris Harper 91 57 790 62.6% 8.7 22.7% 2 98.5 66.9% 129.8 58
Arizona State D.J. Foster 51 36 387 70.6% 7.6 18.2% 2 153.7 60.6% 128.8 59
Washington State Gabe Marks 85 59 668 69.4% 7.9 19.1% 1 106.4 79.9% 127.6 60
Louisville Eli Rogers 35 31 422 88.6% 12.1 14.1% 3 136.7 54.9% 127.2 61
Notre Dame DaVaris Daniels 64 30 463 46.9% 7.2 25.7% 2 129.4 52.3% 125.8 62
Auburn Sammie Coates 46 21 536 45.7% 11.7 25.8% 1 111.0 37.6% 125.7 63
Vanderbilt Jonathan Krause 45 32 608 71.1% 13.5 18.8% 2 97.1 51.0% 125.5 64
Central Michigan Titus Davis 63 34 627 54.0% 10.0 27.2% 1 89.6 51.5% 124.9 65
Tulane Ryan Grant 75 53 700 70.7% 9.3 32.2% 1 83.9 49.3% 124.4 66
Houston Daniel Spencer 47 31 543 66.0% 11.6 17.7% 2 108.1 55.6% 123.1 67
Maryland Deon Long 55 32 489 58.2% 8.9 21.4% 2 123.4 52.0% 122.2 68
Troy Eric Thomas 65 42 667 64.6% 10.3 22.0% 1 99.6 54.4% 122.1 69
San Jose State Tyler Winston 37 24 506 64.9% 13.7 14.4% 2 114.3 54.1% 121.8 70
Buffalo Alex Neutz 67 39 628 58.2% 9.4 30.0% 1 91.6 47.1% 121.6 71
USC Nelson Agholor 50 30 539 60.0% 10.8 24.0% 2 100.4 45.8% 119.2 72
Indiana Shane Wynn 31 22 394 71.0% 12.7 11.5% 4 137.5 59.3% 118.9 73
Texas A&M Malcome Kennedy 64 43 447 67.2% 7.0 21.5% 1 149.9 52.8% 118.8 74
Stanford Devon Cajuste 34 21 377 61.8% 11.1 19.9% 2 129.4 41.1% 117.4 75
Team Player Targets Catches Yards Catch
Rate
Yds Per
Target
Target
Rate
Target
No.
Passing
S&P+
Pass
Rate
RYPR Rk
Georgia State Albert Wilson 85 43 798 50.6% 9.4 32.6% 1 66.5 57.8% 117.3 76
South Carolina Bruce Ellington 44 31 468 70.5% 10.6 18.0% 1 127.2 48.1% 117.3 77
SMU Keenan Holman 65 42 665 64.6% 10.2 17.5% 3 96.0 67.6% 116.3 78
San Diego State Ezell Ruffin 61 40 668 65.6% 11.0 25.7% 1 84.1 48.8% 115.5 79
South Carolina Damiere Byrd 42 23 460 54.8% 11.0 17.2% 2 127.2 48.1% 115.3 80
Marshall Tommy Shuler 73 49 517 67.1% 7.1 29.9% 1 110.2 49.4% 115.3 81
Missouri Dorial Green-Beckham 50 31 457 62.0% 9.1 19.8% 3 123.9 51.1% 114.8 82
Western Michigan Corey Davis 100 55 761 55.0% 7.6 32.7% 1 82.6 55.8% 114.5 83
Florida Solomon Patton 33 28 426 84.8% 12.9 21.0% 2 106.2 39.5% 113.9 84
Nevada Brandon Wimberly 88 61 622 69.3% 7.1 32.2% 1 108.6 46.0% 113.7 85
Washington Jaydon Mickens 53 45 542 84.9% 10.2 20.9% 1 111.5 47.7% 113.5 86
Baylor Corey Coleman 24 13 244 54.2% 10.2 11.3% 4 209.9 46.9% 112.8 87
Central Florida J.J. Worton 34 20 311 58.8% 9.1 16.5% 2 141.0 52.9% 112.7 88
Missouri Marcus Lucas 60 39 447 65.0% 7.5 23.8% 1 123.9 51.1% 112.3 89
North Carolina Quinshad Davis 42 30 415 71.4% 9.9 16.8% 2 127.9 52.7% 112.0 90
Illinois Josh Ferguson 29 25 346 86.2% 11.9 13.7% 2 128.3 53.3% 111.6 91
Ball State Jamill Smith 67 44 592 65.7% 8.8 19.5% 3 109.6 58.6% 110.5 92
Texas A&M Derel Walker 41 30 414 73.2% 10.1 13.8% 3 149.9 52.8% 110.0 93
North Texas Darnell Smith 63 46 518 73.0% 8.2 25.5% 1 108.1 48.3% 109.6 94
Arkansas State Julian Jones 51 39 476 76.5% 9.3 24.4% 2 95.8 50.1% 109.3 95
South Alabama Shavarez Smith 39 26 462 66.7% 11.8 17.6% 3 101.4 51.3% 108.3 96
Baylor Levi Norwood 18 15 234 83.3% 13.0 8.5% 5 209.9 46.9% 108.1 97
Alabama Kevin Norwood 26 23 348 88.5% 13.4 11.6% 3 135.6 51.5% 107.9 98
Fresno State Isaiah Burse 64 52 533 81.3% 8.3 18.6% 3 99.9 69.5% 107.5 99
Boise State Matt Miller 62 47 504 75.8% 8.1 22.5% 2 114.5 51.0% 107.0 100

For the most part, this passes the eyeball test. Baylor's Passing S&P+ is 21 percent higher than anybody else's right now, so it makes sense that the Bears would have a couple of (awesome) players near the top. And a lot of other names you'd expect -- Rashad Greene, Mike Evans, Brandin Cooks, LSU's Beckham-Landry duo, Jeremy Gallon, Tyler Lockett, Sammy Watkins, L'Damian Washington -- are all in the top 40. There are some surprises, of course, but I like that, too. I had no idea Arizona State's Jaelen Strong was putting together such good numbers; UCF's Breshad Perriman, too.

Anyway, again, this is an attempt to restart the conversation. In theory, plan to take a look at some year-to-year correlations (and, in theory, some college-to-pro numbers) to see how each factor should be weighted, but for now, this will do.

And yes, Baylor's top two receivers are each averaging more than 17 yards per target. That is completely insane.

This Week at SB Nation

Monday
In college football, pain is our family tie

Tuesday
The Numerical, Week 9: Florida State gets it over with quickly

Wednesday
Win projections: How the four-team SEC East race will likely play out
South Carolina 27, Missouri 24: Beyond the box score

Thursday
Win projections: Big Ten's other division title goes through Michigan State
Florida State is a huge favorite over Miami for a reason
Week 10 F/+ Picks

Posted by: Bill Connelly on 31 Oct 2013

6 comments, Last at 01 Nov 2013, 10:21pm by Dan

Comments

1
by Anonymously (not verified) :: Thu, 10/31/2013 - 11:55pm

This is an intriguing stat, Bill. Thanks for this. Can you please add a simple twitter button to share this?

2
by CBPodge :: Fri, 11/01/2013 - 8:37am

This is definitely an interesting stat. I reckon it definitely needs some weightings assigned to the various ratings in it. But I think it needs a few more years' data (either going forward or back) before that happens. Because otherwise you'd probably end up weighting it so that you get the outcome that you want (best players at the top), rather than get the outcome that you want because of how it's weighted. If that makes sense.

And I don't care what its actually called: my head will always read RYPR as "Receiving Yards Per Reception".

3
by Chase (not verified) :: Fri, 11/01/2013 - 9:05am

Your formula no longer includes targets, which I would consider an improvement.

I might consider trying to do something like True Receiving Yards (linked to my name -- you would need to read part 1 first) for college football.

4
by Chase (not verified) :: Fri, 11/01/2013 - 10:47am

I also did something similar (link to my name again) for the NFL, putting two things on different axes: percentage of team targets and team ANY/A. I assume Passing S&P+ is similar enough to Team ANY/A, which means it's a measure of passing offense. Percentage of team targets was then my way of measuring how important a receiver was to the passing game.

One issue I didn't like was Michael Crabtree ranked so high, and that was because the 49ers didn't pass that often but had a great pass efficient rating, and when they did pass, it was to Crabtree. This effect is probably magnified in CFB (I'm thinking Georgia Tech some years), so it makes sense to instead adjust Team Passing prowess by a denominator that includes all plays, not just passing plays. I like the idea in theory, although I worry that it would shift things too much towards pass-happy teams (at least, for the NFL).

5
by Kal :: Fri, 11/01/2013 - 9:45pm

I think that's a good idea, but it also depends on what you're trying to demonstrate. Doing it against all plays shows how dominant a player can be relative to the overall team itself. Doing it against passing plays shows how important that player is to the success of the passing game. Both are useful. As an example - imagine taking out Tyler Eifert from the passing game from Stanford last year. Stanford had something like a 62/38 run/pass ratio, so in the first system Eifert wouldn't look like that big a deal - but if you just measure him by his passing, he's like half the value of the entire pass game for Stanford.

Both ways have value. You might use both on each player - showing how much production each person is worth as a percentage of the overall team's success AND showing how much production they're worth as part of the passing game.

6
by Dan :: Fri, 11/01/2013 - 10:21pm

I think that yards per team play, yards per team passing attempt, and yards per target all contain some independent information. Some styles of offense improve one of those stats at the expense of another - run-heavy hurts a receiver's yds/play but tends to help the other two, force-feeding one WR helps yds/play and yds/patt, but hurts yds/target, etc.

It's not very elegant, but one thing that I've done is to basically just average those 3 stats together. (First I took the square root of each of them, which prevents a player from being too great just on the basis of one of the three stats, and then I re-scaled each of them to be on the same scale.)