Arrowhead Pride: An SB Nation Community

Navigation: Jump to content areas:


Sports blogs for fans, by fans.
New Blog: RSL Soapbox for Real Salt Lake Fans!

Bringing valuation metrics to football


I've Wanted to write a post similar to this for awhile now. At first, it was because I wanted to prove the value of Matt Cassell, Then I wanted to Prove the monetary value of our key players, then I wanted to prove Why money is the most important factor to a football team. Frankly, I'm just going to present some stats and some very sabermetricy ideas.

I think part of the reason That this type of analysis of player valuation hasn't caught on in football, is that. there are alot of small sample sizes. In Baseball we look at 500 plate appearances and 162 games. In football it's different. We may look at 100 catches and 16 games. We all know from basic statistics that small sample sizes can greatly skew numbers like these.

We're going to jump right into it anyway.

I had a short debate earlier that centered around my thesis that Money drives the game far more than an altruistic desire to win championships. In baseball, There are a few metrics that have been used to put a monetary value on a playoff appearance and on a player's worth in reference to wins. I'm going to do my best to apply that to football. I'm taking general principles from the book "Baseball, between the numbers: why everything you know about the game is wrong".

I'm going to adapt the formula by basically multiplying the value of an individual game or win by 10 162/16 is about 10. I realize that the numbers won't fall in line perfectly. It assumes that the NFL and MLB have similar gross revenue. And it doesn't have an adjustment for park factors or local socio economic factors, it's just a rought draft. If you have a better way to do it, please share.

The book asses that in baseball an extra win, is approximately worth $1,196,000 so in Theory if a football player somehow can be worth an extra win by himself he's essentially worth $11,960,000. The hard part comes from the fact that we Have a stat Like wins above replacement for baseball, and really no stat that can value a win in terms of football. 

In all honesty we're basically going to have to make up a number based on DYAR on football outsiders. As a point of reference. DYAR means Defense adjusted Yards above replacement. In baseball the stat is already nice and tidy and estimates wins above replacement. We need to Figure out how many Yards equals a win.  Also, replacement level means any of the 22 guys off the street that Haley can pick up. As a side note. LJ has been -103 DYAR. That's pretty bad.

The best way I found to do this was to take the top QB, RB, WR, and  TE.  from each  team and add up their DVR and divide by the number of wins. I had to leave out the win less teams because a bunch of zero's would have screwed up the calculations. I know this isn't perfect, but it's just an estimation. If you've got a better idea of how to calculate this please let me know.

The numbers come out to be about 50 yards above replacement level equals 1 win, DYAR is heavily slanted toward QB, as it should be, in reality a QB is much more valuable than any other position, but in terms of DYAR it says Drew Brees is worth about 8 times what Adrian Peterson is worth. So it all should be taken with a grain of salt. Maybe I should be dividing the DYAR of a QB by three in order to get a more level playing field. Seeing how roughly a 300 yard passing game and 100 yard running game are equivalent. Maybe I'll try that later. Let's stick with the status quo.

Assuming all of these things I expected Matt Cassel to be more valuable than he has. Obviously it's easier for these metrics to work in baseball because one player is independent of another on most plays, in football, there are alot more variables.

There are a few things that really stick out. The value of a True Franchise QB, is basically off the charts and becomes a statistical anomaly All of these mythical wins are also based on offense alone, Meaning they might have to be halved in terms of total cash value considering you'd spend roughly half your cap on offense and defense...

Lots to think about here This is just a jumping off point... points of discussion...

Let's think about LJ for a second. He is -103 DYAR meaning a replacement level player would be about 2 wins more valuable than he is.... Negative numbers really throw the stats off, but in general you get the idea he's been pretty bad.

As for Cassell. He's been about -49 DYAR meaning he's roughly a game worse than a replacement level player. Certainly not playing up to that 60 mil contract (to be honest when I started this I didn't expect him to have a negative number.)

Bowe had a positive 71, so he's been worth about 1.5 wins Thus far. By far our most productive player. worth about 18 million to the front office in terms of gross revenue.

Like I said earlier, I know some of it is flawed due to the nature of football.

Here are the issues I can identify.

I need to adjust for roster size

The Salary Cap.

Find a way to quantify blocking skills.

And positional weights need to be adjusted.

I also divided by 4 (the number of player's DYAR I added together) If I multiply that back out. It takes about 200 yards to equal a win on your own. It balances out the skewed QB numbers but I think really devalues running backs. In that equation it values the best running backs at a bout .75 worth of a win. And it only put's LJ at about .5 of a loss.... Which might be accurate... Hmmm Not sure.

I think we can look back and say that if we based cassel's value at the end of they year on this rough valuation equation, he wouldn't get anywhere near 60 million. We can also look back and Say LJ is one of the worst 3 backs in the league. And we can also say that Bowe has been a lone bright spot.

The other thing that this shows, is that Management can put a number on a player's worth from a strictly business point of view. If we were to perfect the calculations we could say, this player was worth this amount of gross revenue to the team, and deserves this raise. We can also see who is under performing or outperforming their contract. I haven't figured out a way to turn gross revenue into player salary, but that's next.

This is a FanPost and does not necessarily reflect the views of Arrowhead Pride's writers or editors. It does reflect the views of this particular fan though, which is as important as the views of Arrowhead Pride writers or editors.

2 recs  |  Comment 23 comments  |  Add comment

Story-email Email Printer Print

Comments

Display:

On further review...

I don’t think I should have divided by 4 so let’s assume that 200 DYAR is 1 win That puts guys like drew brees around 4 wins all by themselves… and it puts guys like cassell at only about -.25 wins rather than -1….. which means he’s basically a replacement level player. In other words…. one of 22 guys off the street.

by callmesir on Oct 27, 2009 1:19 AM CDT reply actions   0 recs

You obviously put a lot of work into this. Thanks a lot for your contribution to AP!

I'm officially on the 'Draft Eric Berry' train.

by scottbwalters on Oct 27, 2009 1:28 AM CDT reply actions   0 recs

It probably was a little too long...

should have broken it up a little bit… it’s relatively constructive… I guess.

by callmesir on Oct 27, 2009 8:32 AM CDT up reply actions   0 recs

What is the point of posting something like this?

The guy obviously put a lot of effort and research into this. Post something constructive or don’t post at all.

by cvigz on Oct 27, 2009 9:08 AM CDT up reply actions   1 recs

This is actually very interesting

nice work.

"The spirit, the will to win, and the will to excel are the things that endure. These qualities are so much more important than the events that occur." - Vince Lombardi

by Shawn on Oct 27, 2009 1:39 AM CDT reply actions   0 recs

appreciate this

Nice work man. I wish football had the sample sizes you’re talking about in baseball. I love watching the metrics take shape there.

by Matt Conner on Oct 27, 2009 10:25 AM CDT reply actions   0 recs

Great work

Brilliant stuff. Keep doing this stuff — I love it.

by Nick Britt on Oct 27, 2009 12:33 PM CDT reply actions   0 recs

Good Stuff.

Well worth the read.
If I understand the stats Cassel, LJ, Bowe, and Ryan. would have to equal our 1 win or our adjusted 2.2 wins for a 16 games season?

We Will kick at least 4 Teams Asses in 09
And Succop will be the Key in two of them.
" Think and talk positive football off the field." Hank Stram

by Steve_Chiefs on Oct 27, 2009 12:44 PM CDT reply actions   0 recs

Part of the problem with all of this is that

In baseball, they base every thing off a mythical replacement level team that would win 63 ish games. This team would basically be 4A type players. Then they calculate the value of a player based on how many wins that player alone would add to a ball club if he replaced the mythical player at his position… For example. Albert pujols would add about 8 wins to that team by himself. And the calculations are then adjusted by position difficulty. If someone had the same offensive year at 2nd base as they did 1st base it’d be weighted more heavily.

We don’t have a mythical standard in football yet, so it’s kinda hard to establish exactly how many wins it would be worth. in theory, we could set that standard at about 2 wins. So we’d look a player’s stats and say, if we added Drew brees to this mythical team, he’d be worth 4 wins by himself. In the case of LJ right now, He’d take away from that teams win total by about .5 wins. meaning he’s worse than any random draftable college player.

I talked to the guys over at football outsiders and said the reason they haven’t been able to assign a number of yards to a win, to help extrapolate this out, is because they can really only apply it to offense skill positions, meaning there would be too many variables….

These calculations weren’t meant as a end all be all, but rather a some ideas of what could be done, in reality, There are entire teams (and by teams I mean QB RB WR AND TE ) That have negative DYAR and that really throws a monkey wrench in things. Because we all know you can’t have a negative win total.

by callmesir on Oct 27, 2009 2:46 PM CDT up reply actions   0 recs

But if you base it off average 8 wins

then those tandems that are negative would pull wins off the board

We Will kick at least 4 Teams Asses in 09
And Succop will be the Key in two of them.
" Think and talk positive football off the field." Hank Stram

by Steve_Chiefs on Oct 27, 2009 2:57 PM CDT up reply actions   0 recs

I think the problem with basing it off 8 wins

is that you’re assuming that the mythical team is full of league average players rather than “replacement level” meaning theoretically a team could go over 16 wins…. but thanks for the input/ideas man

by callmesir on Oct 27, 2009 3:25 PM CDT up reply actions   0 recs

I would think that Cassels numbers would be skewed from the OL not playing up to par. In other words, his DYAR is directly affected by his quality of talent surrounding him. In baseball its not quite the same. Maybe you could rate the offense as a whole and see if that number falls in line with other 1 win teams.

by pokerjoe on Oct 27, 2009 7:51 PM CDT reply actions   0 recs

You're absoloutely right.

That’s part of the problem with some of these metrics and the reason that they have a hard time gaining traction, but you keep tweaking and tweaking the formula the best you can. The more conversation you can get about it, the better.

by callmesir on Oct 27, 2009 8:59 PM CDT up reply actions   0 recs

Also, does LJ’s -103 DYAR mean he is causing the Chiefs to lose 24 million dollars by playing him?

by pokerjoe on Oct 27, 2009 7:55 PM CDT reply actions   0 recs

based on the first set of calculations.....

Considering we can put a number on about how much money a win is worth to an organization, that in theory, Larry is costing us -2 wins above a replacement level player…. so in theory…. he costs the the organization that much money with his poor play costing wins, Based on the second set of calculations, he’s worth about -.5 wins, would mean he’d be costing the chiefs somewhere closer to 6 million, on top of the salary we’re dumping into him.

by callmesir on Oct 27, 2009 8:58 PM CDT up reply actions   0 recs

Assigning dollar values doesn't have to be perfect.

Just apply the rule the same way across the league, and relative differences will give you something with traction. For instance, who’s to put a dollar value on the noise from jet-skis disturbing your beach experience? Well, ANYbody. And once that assignment is made, you can compare different strategies for which minimize the “noise cost.”

I think there might be more hay to be made by looking at salary profiles and see if, maybe, once you get past the franchise QB, that the more successful teams are spending more coin on, for instance, o-line…

Baseball’s much more of an individual sport. A great pitcher is going to shut down the opposition’s offense all by himself. You plug in a better hitting 2nd baseman and reap virtually instant results. I’d be concerned that any model such as this that you’re trying to build would see teams doing nothing but spend the whole shebang on a few skill players and neglect the trenches.

No question. Otis Taylor should be in the Hall of Fame.

by hmills110 on Oct 27, 2009 9:25 PM CDT reply actions   0 recs

you're right...

right now DYAR is only set up to judge offensive skill players. So obviously this has limited value. Any suggestions on how to judge blocking in terms of numbers? I talked to the guys over at football outsiders before I wrote this and they basically they said no attempt had been made to do those things because it’s almost impossible to parse out the value of the non skill players…..

At least the skill players are a start… I think my next post might be a little more simplified…. and I’ll take your suggestion about taking a player’s dollar worth and seeing what the difference is between their actual pay, similar to what they do in baseball, and see if there is a correlation between teams that have a high dollar amount invested in their QB RB an top wideout. and winning, or see if it’s a correlation is really with teams that invest in the trenches… thanks for the idea.

by callmesir on Oct 27, 2009 10:34 PM CDT up reply actions   0 recs

Maybe grouping

O-line
D-front seven
D secondary
Special teams
then
QB
WR
RB
TE
8 groups sorta like FF maybe

We Will kick at least 4 Teams Asses in 09
And Succop will be the Key in two of them.
" Think and talk positive football off the field." Hank Stram

by Steve_Chiefs on Oct 27, 2009 9:34 PM CDT reply actions   0 recs

Football outsiders has

DVOA and other metrics for O-line and D-line, but they aren’t standardized like the DYAR is for the skill positions that’s why I had a hard time including them…. but I bet I could figure SOMETHING out… wanna help???

by callmesir on Oct 27, 2009 10:35 PM CDT up reply actions   0 recs


User Tools

Welcome to SB Nation's Kansas City Chiefs blog!
Start posting about the Chiefs »

Join SB Nation and dive into communities focused on all your favorite teams.

FanPosts

Community blog posts and discussion.

Recommended FanPosts

Plainview_small
Talk Steelers with BTSC
Chiefs_pic_small
A little love for the Kool Aid drinkers (-!
Kool_aid_man_small
The Kool Aid Drinker’s Manifesto on the Kansas City Chiefs

Recent FanPosts

Small
Time To Go To The War Room
Turtle_article_small
AP Power Rankings Week 11; You Could Make It At Receiver For The Chiefs Edition
Us_on_thanksgiving_08_small
Beyond this season...
Fama150_small
TGIF Prediction Poll - Kansas City vs Pittsburgh Steelers
Small
Arrowhead Pride's name sealed in ..... brass?
Kool_aid_man_small
Log Jam at the Bottom
Itsgood2_small
Thursday Night Football Open Thread
Iron_cross_real_small
Nothing to Lose or Everything to Lose
104_4223_small
My new Mock 2.0
Tobias_small
Random Thoughts: Bowe, Bills, Charles & More

+ New FanPost All FanPosts >

Latest NFL Headlines from SB Nation

Gang Green Nation
How to Beat the Cheats-- Sunday Preview
Pride Of Detroit
Week 11 Injury Report
Mobile Live
Garrett Wolfe Out For the Year (Updated)

Managers

Arrowhead_pride_small Chris Thorman

Tg_small Joel Thorman

Editors

N505381175_257425_5488_small Matt Conner

Contributors

Ajax_small ChiefDJ

Small Jon Yoon