Discussing/Analyzing individual player value and WAR for the 2014 season

[Sambo.]

I'm a bit of a stat nerd and have been fiddling around with some numbers lately. I've been trying to find a way where we can evaluate a football player's individual value to a specific team. Everyone probably knows that WAR (Wins Above Replacement Level) has taken over as baseball's premier and most accurate statistic when it comes to evaluating a player's performance. So why not try the same thing with football? The most logical place to start is with our team last year.

The first thing is you have to determine what the overall team's value is because everything will be based off this score. Using yardage differential, points, and first downs, I tested out the 2014 Auburn Tigers and got a raw score of 7.935 wins. I'd say that is pretty accurate since we finished the regular season 8-4. The next step is to determine how much each player contributed to that win total. Using total yards, points per game, average per play, and individual percentage of team's total output, you can determine every player's WIN SHARE. Obviously, these numbers fluctuate based on position. The quarterback is always going to have the highest win share because its the most important position on the field. Think about it- for every play a receiver or tight end makes, the quarterback is getting those numbers, too. So just because a QB has what appears to be unproportionally high score doesn't mean they're an All-American (or that the receivers suck). Some positions are just worth more than others.

So here's roughly how many wins each offensive skill player contributed to the team last season..
WIN SHARES:

1. Nick Marshall, QB 2.409 Wins
2. Cameron-Artis Payne, RB 1.350 Wins
3. D'haquille Williams, WR .544 Wins
4. Sammie Coates, WR .509 Wins
5. Quan Bray, WR .459 Wins
6. Ricardo Louis, WR .389 Wins
7. Corey Grant, RB .377 Wins
8. Jeremy Johnson, QB .365 Wins
9. Roc Thomas, RB .194 Wins
10. Melvin Ray, WR .126 Wins

To find the WAR, you have to compare these to what a REPLACEMENT player would've done in the same position for the same team. Replacement player level is roughly between a very low rated starter and active backup. For instance a replacement level wide receiver in 2014 SEC produces 31.44 YPG and 1.1 PPG over a 12 game season. You just have to subtract these from the player's total win share. This also fluctuates with position so here's a key..

QB WAR:
< 0.5 Bench Player
0.5-1.0 Low Tier Starter
1.0-1.5 Quality Starter
1.5-1.75 Honorable Mention All-Conference
1.75-2.00 2nd Team All-Conference
2.00-2.50 1st Team All-Conference
+2.50 All American

RB WAR:
< 0 Reserve
0-0.25 Low Tier Starter
0.25-.5 Quality Starter
.5-.75 Honorable Mention All-Conference/2nd Team All-Conference
.75-1.0 All-Conference
+1.0 All-American

WR WAR:
< 0 Reserve
0-.20 Low Tier Starter
0.20-0.30 Quality Starter
0.30-0.40 Honorable Mention All-Conference
0.40-0.50 2nd Team All-Conference
0.50-.75 All-Conference
+ 0.75 All American

2014 AUBURN OFFENSIVE WAR:

1. Nick Marshall +1.58
2. Cameron-Artis Payne +.791
3. Duke Williams +.337
4. Sammie Coates +.301
5. Quan Bray +.252
6. Ricardo Louis +.181
7. Melvin Ray -.08
8. Corey Grant -.182
9. Roc Thomas -.36

I know its not an exact science and its difficult to make sense of with such little data. But I thought it was interesting to see how many wins each player was actually worth to us last year. I'm going to try to do something similar with defense, although they will be very low because the overall unit's win share as a whole wasn't high in 2014.

 

[WarEagleG]

This is premium content. Please subscribe to view.

 

[Sambo.]

This is premium content. Please subscribe to view.

 

[Sambo.]

This is premium content. Please subscribe to view.