Value.
In MLS there is arguably no greater importance than finding value in players. With a small salary cap (just $3.09-million in 2014) and only three exceptions (that are punished to the tune of nearly one-third of your cap if all three are used) you cannot afford to overpay for players.
And, if you are going to be a truly elite team you not only need to not overpay, but also underpay for the value you get. If you can find a player for $100,000 that plays like a $200,000 player you are ahead of the game.
How do you measure that though? Up until now there is no statistic you can point to that says you overpaid player X by Y amount and underpaid player Z by F amount. It's all subjective and the problem with subjective measures is that they aren't testable.
We've tried to address that. Over the last month we've been working to create a tool to place an expected salary value on players. With that we can then look at a player's actual salary and express his value in a +/- format.
Below the jump we present the Performance to Pay (P2P) figures for all starting Central Midfielders in MLS from 2013. Once we've received feedback we will possibly tweak the formula and expand the data to include all regulars in 2013.
Before we present the findings, a word about the format:
We first identified the six different outfield positions (goalkeepers will be looked at separately, using a different formula) that a player can be categorized in (central midfielders, attacking midfielders, wide midfielders, central defenders, fullbacks and forwards). Each position will be looked at separately, recognizing that different positions have different skill-sets.
Within each position a player ranking will be established using advanced statistics mined from the Website Whoscored.com. Whoscored is considered to be the best open data set of advanced stats available at this time. We relied upon Whoscored to classify what position a player played when there was confusion doing such.
That player ranking will be used to calculate expected salaries. By looking at the typical salary spread at each position from highest to lowest paid (capping highest at the DP cap hit of $368,750 to accommodate top end outliers) we were able to calculate how much of a pay raise each position in the ranking should expect.
For example, we found that each spot on the CM list should increase a player's salary by $10,204. Since the lowest paid CM was paid $46,500 we assumed that was the market value of the worst regular. The second worst would be worth $56,704, the third $66,908, etc.
Calculating the P2P number was a case of simply subtracting the actual salary from the expected.
More complicated is calculating the player's positional ranking. To do that we took two measures directly from Whoscored.com (Their player ranking (WSR), which is calculated using a cross section of fugues) and passing percentage (P%). Additionally, we created two measures. The measures created are Defensive Involvement (DI), which looks at the average amount of tackles, clearances, recoveries and interceptions per game that a player has, and Shots plus Key Passes (S+KP), which is simply the average of shots and key passes combined per game.
Each player is ranked in each category with the four ranking numbers then added together for a Overall Score Rank (OSR). The lower the OSR, the better the player is thought to have performed.
The results:
[TABLE=width: 500]
[TR]
[TD]Name[/TD]
[TD]Team[/TD]
[TD]WSR[/TD]
[TD]DI[/TD]
[TD]S+KP[/TD]
[TD]Pass %[/TD]
[/TR]
[TR]
[TD]Laba[/TD]
[TD]TFC[/TD]
[TD]6.99[/TD]
[TD]9[/TD]
[TD]1[/TD]
[TD]82.1[/TD]
[/TR]
[TR]
[TD]Osorio[/TD]
[TD]TFC[/TD]
[TD]6.68[/TD]
[TD]2.9[/TD]
[TD]1.8[/TD]
[TD]85.4[/TD]
[/TR]
[TR]
[TD]Beckerman[/TD]
[TD]RSL[/TD]
[TD]7.26[/TD]
[TD]6.6[/TD]
[TD]2.8[/TD]
[TD]81.8[/TD]
[/TR]
[TR]
[TD]Grabovoy[/TD]
[TD]RSL[/TD]
[TD]7.02[/TD]
[TD]5.5[/TD]
[TD]2.2[/TD]
[TD]86.5[/TD]
[/TR]
[TR]
[TD]Mejia[/TD]
[TD]Chivas[/TD]
[TD]6.57[/TD]
[TD]3.4[/TD]
[TD]1.7[/TD]
[TD]77.1[/TD]
[/TR]
[TR]
[TD]Minda[/TD]
[TD]Chivas[/TD]
[TD]6.48[/TD]
[TD]5.2[/TD]
[TD]0.7[/TD]
[TD]82.8[/TD]
[/TR]
[TR]
[TD]Larentowicz[/TD]
[TD]Chicago[/TD]
[TD]6.90[/TD]
[TD]6[/TD]
[TD]2.2[/TD]
[TD]76.4[/TD]
[/TR]
[TR]
[TD]Alex[/TD]
[TD]Chicago[/TD]
[TD]6.64[/TD]
[TD]3.4[/TD]
[TD]1.9[/TD]
[TD]78.2[/TD]
[/TR]
[TR]
[TD]Powers[/TD]
[TD]Colorado[/TD]
[TD]7.05[/TD]
[TD]3[/TD]
[TD]3.1[/TD]
[TD]81.4[/TD]
[/TR]
[TR]
[TD]Thomas[/TD]
[TD]Colorado[/TD]
[TD]6.96[/TD]
[TD]7.4[/TD]
[TD]1.6[/TD]
[TD]83.9[/TD]
[/TR]
[TR]
[TD]Trapp[/TD]
[TD]Columbus[/TD]
[TD]6.83[/TD]
[TD]6.1[/TD]
[TD]1.8[/TD]
[TD]84.0[/TD]
[/TR]
[TR]
[TD]Tchani[/TD]
[TD]Columbus[/TD]
[TD]6.76[/TD]
[TD]5.1[/TD]
[TD]1.1[/TD]
[TD]75.2[/TD]
[/TR]
[TR]
[TD]Jeffrey[/TD]
[TD]DCU[/TD]
[TD]6.77[/TD]
[TD]4.9[/TD]
[TD]2.9[/TD]
[TD]86.1[/TD]
[/TR]
[TR]
[TD]Kitchen[/TD]
[TD]DCU[/TD]
[TD]6.72[/TD]
[TD]6.4[/TD]
[TD]1[/TD]
[TD]82.5[/TD]
[/TR]
[TR]
[TD]Jacobson[/TD]
[TD]Dallas[/TD]
[TD]6.98[/TD]
[TD]7.2[/TD]
[TD]2[/TD]
[TD]79.8[/TD]
[/TR]
[TR]
[TD]Watson[/TD]
[TD]Dallas[/TD]
[TD]6.55[/TD]
[TD]3.9[/TD]
[TD]1.3[/TD]
[TD]78.0[/TD]
[/TR]
[TR]
[TD]Clark[/TD]
[TD]Houston[/TD]
[TD]7.04[/TD]
[TD]4.9[/TD]
[TD]2.4[/TD]
[TD]78.6[/TD]
[/TR]
[TR]
[TD]Creavalle[/TD]
[TD]Houston[/TD]
[TD]6.89[/TD]
[TD]6.4[/TD]
[TD]1.1[/TD]
[TD]80.2[/TD]
[/TR]
[TR]
[TD]Juninho[/TD]
[TD]LAG[/TD]
[TD]7.18[/TD]
[TD]6.5[/TD]
[TD]3.1[/TD]
[TD]85.6[/TD]
[/TR]
[TR]
[TD]Sarvas[/TD]
[TD]LAG[/TD]
[TD]7.18[/TD]
[TD]6.5[/TD]
[TD]2.5[/TD]
[TD]83.6[/TD]
[/TR]
[TR]
[TD]Bernier[/TD]
[TD]Montreal[/TD]
[TD]7.07[/TD]
[TD]6.3[/TD]
[TD]1.5[/TD]
[TD]82.4[/TD]
[/TR]
[TR]
[TD]Arnaud[/TD]
[TD]Montreal[/TD]
[TD]6.91[/TD]
[TD]4.8[/TD]
[TD]1.5[/TD]
[TD]81.3[/TD]
[/TR]
[TR]
[TD]Caldwell[/TD]
[TD]NER[/TD]
[TD]6.63[/TD]
[TD]5.9[/TD]
[TD]0.7[/TD]
[TD]82.9[/TD]
[/TR]
[TR]
[TD]McCarty[/TD]
[TD]NYRB[/TD]
[TD]7.01[/TD]
[TD]6.1[/TD]
[TD]1.3[/TD]
[TD]85.1[/TD]
[/TR]
[TR]
[TD]Alexander[/TD]
[TD]NYRB[/TD]
[TD]6.71[/TD]
[TD]2.7[/TD]
[TD]1.6[/TD]
[TD]82.7[/TD]
[/TR]
[TR]
[TD]Daniel[/TD]
[TD]Philly[/TD]
[TD]6.77[/TD]
[TD]3.6[/TD]
[TD]2.5[/TD]
[TD]84.8[/TD]
[/TR]
[TR]
[TD]Carroll[/TD]
[TD]Philly[/TD]
[TD]6.73[/TD]
[TD]6.1[/TD]
[TD]1.2[/TD]
[TD]80.6[/TD]
[/TR]
[TR]
[TD]Johnson[/TD]
[TD]PDX[/TD]
[TD]7.46[/TD]
[TD]6.3[/TD]
[TD]3.3[/TD]
[TD]83.7[/TD]
[/TR]
[TR]
[TD]Chara[/TD]
[TD]PDX[/TD]
[TD]7.23[/TD]
[TD]7.7[/TD]
[TD]1.2[/TD]
[TD]85.9[/TD]
[/TR]
[TR]
[TD]Cronin[/TD]
[TD]San Jose[/TD]
[TD]6.98[/TD]
[TD]7.1[/TD]
[TD]0.9[/TD]
[TD]79.7[/TD]
[/TR]
[TR]
[TD]Alonso[/TD]
[TD]Seattle[/TD]
[TD]7.35[/TD]
[TD]8.8[/TD]
[TD]1.8[/TD]
[TD]88.3[/TD]
[/TR]
[TR]
[TD]Evans[/TD]
[TD]Seattle[/TD]
[TD]6.92[/TD]
[TD]4[/TD]
[TD]1.8[/TD]
[TD]77.7[/TD]
[/TR]
[TR]
[TD]Zusi[/TD]
[TD]SKC[/TD]
[TD]7.48[/TD]
[TD]2.7[/TD]
[TD]5.1[/TD]
[TD]76.3[/TD]
[/TR]
[TR]
[TD]Rosell[/TD]
[TD]SKC[/TD]
[TD]7.11[/TD]
[TD]8.4[/TD]
[TD]1.5[/TD]
[TD]84.0[/TD]
[/TR]
[TR]
[TD]Koffie[/TD]
[TD]Vancouver[/TD]
[TD]7.0[/TD]
[TD]5.5[/TD]
[TD]1.8[/TD]
[TD]81.4[/TD]
[/TR]
[TR]
[TD]Reo-Coker[/TD]
[TD]Vancouver[/TD]
[TD]7.0[/TD]
[TD]5.1[/TD]
[TD]1.6[/TD]
[TD]81.9[/TD]
[/TR]
[/TABLE]
That's a lot of numbers. To help digest here is a list of the top five and bottom five in each category.
WSR
Top -
1. Zusi - 7.48
2. Johnson - 7.46
3. Alonso - 7.35
4. Beckerman - 7.26
5. Chara - 7.23
Bottom
37. Mirda - 6.48
36. Watson - 6.55
35. Mejia - 6.57
34. Caldwell - 6.63
33. Alex - 6.64
This is the stat that most meets the eyeball test. The players that are ranked high in thus measure are well known and generally considered to be stars in the league.
They also play for top teams, while the bottom five are four-for-five non-playoff performers.
It's possible that the WSR leans a bit too heavily on numbers that are team generated so we must look at the rest of the figures to find balance.
Central midfielders are expected to be the best of all worlds -- they must attack and defend. So, let's look at defensive involvement:
DI
Top -
1. Laba - 9
2. Alonso - 8.8
3. Rosell - 8.4
4. Chara - 7.7
5. Thomas - 7.4
Bottom -
37. Zusi - 2.7
37. Alexander - 2.7
35. Osorio - 2.9
34. Powers - 3
33. Alex - 3.4
33. Mejia - 3.4
A couple things are starting to jump out. Alonso and Chara are in both top categories; Alex and Menja in both bottom.
Also the odd, possibly outlier position of Zusi. Clearly one of the most well known stars in MLS it's possible he's just been mislabeled a CM when he's more of an attacking player. This will come up again on this list and it illustrates the difficulty of measuring CMs.
TFC and Vancouver fans are going to debate the Laba numbers. However, if all you want from him is defensive involvement then he provides it -- in fact, he did it better than any other CM last year on a statistical level.
Now we've looked at a defensive stat we need to look at the offensive contribution, with shots plus key passes.
The reason we've excluded goals is because they are so rare that they create outlier situations. We will include goals in our measure of forwards, but as midfielders go we're going to stick to S+KP. In the new stats world looking at shots as a measure of offensive contribution is fairly common and key passes is a highly desired attribute.
S+KP
Top -
1. Zusi - 5.1
2. Johnson - 3.3
3. Juninho - 3.1
3. Powers - 3.1
5. Jeffrey - 2.9
Bottom -
37. Caldwell - 0.7
37. Mejia - 0.7
37. Mirda - 0.7
34. Cronin - 0.9
33. Laba - 1.0
33. Kitchen 1.0
Yep, Laba is a pure DCM. A valuable and needed position, but is it worth a DP slot? That's a subjective argument that won't be won or lost here.
Lastly we look at pass completion percentage. The argument against using completion percentage as a measure is that not all passes are created equal. That's fair, but tracked over an entire season the numbers start to have meaning -- particularly if a high completion rare is combined with a high S+KP number.
Top -
1. Alonso - 88.3
2. Grabovoy - 86.5
3. Jeffrey - 86.1
4. Chara - 85.9
5. Juninho - 85.6
Bottom -
37. Zusi - 76.3
37. Tchani - 76.3
35. Larentowicz - 76.4
34. Mejia - 77.1
33. Evans - 77.7
Once again we see some familiar faces. Rather than try and predict what that means let's look at the overall rank.
Best overall CM ranking (combining the four categories)
1. Alonso
2. Juninho
3. Johnson
4. Sarvas
5. Chara
6. Beckerman
7. Rosell
8. Grabovoy
9. Thomas
10. Jeffrey
11. McCarty
12. Trapp
13. Bernier
14. Jacobson
15. Laba
16. Powers
17. Koffie
18. Daniel
19. Zusi
20. Clark
21. Reo-Coker
22. Larientowicz
23. Kitchen
24. Osorio
25. Cronin
26. Creavalle
27. Carroll
28. Evans
29. Arnaud
30. Caldwell
31. Alexander
32. Mirda
33. Alex
34. Tchani
35. Watson
36. Mejia
10 thoughts:
1) Alonso earned his pay raise
2) Chivas is really bad
3) The formula rewards all-around play. Zusi is the poster-boy to why.
4) Is Laba worth a DP slot when he's statistically average (overall), but exceptional at one thing?
5) But, maybe there is something to Zusi's low ranking. With a limited cap should teams be looking for overall play over specialist play. Was his truly exceptional S+KP number enough to make clear weaknesses elsewhere irrelevant? Dumb stats creating more questions than answers!
6) Harsh reality of MLS reflected in Montreal unloading Arnauld, eh?
7) How is Evans a US international? The eyeballs there (which tell you he shouldn't be) are absolutely backed up by poor numbers.
8) Clark might be next year's Arnauld. How much is "leadership" worth as an intangible?
9) With seven of the top 10 international players the US might need to reflect on how they develop CMs (also suggests that Will Johnson is the only Canadian national team member that would have a shot at a US cap right now)
10) Jered Jeffrey was one hell of a steal for DC United.
The purpose of this exercise was to determine whether player's were overpaid or of great value. As a reminder we placed a salary value in each spot in the rankings based on what the spread in salary was from No 1 to No 36 in 2013. We determined that the No 36 ranked player should be paid the league minimum of of $46,500 and that each spot up the ranking should be paid an additional $10,204 until we reached the maximum cap hit of $368,750.
Once we determined those salary numbers we simply subtracted the actual cap hit from 2013 (according to the salaries listed on the MLSPU page) with their actual salary.
The results (negative value represents a player being underpaid that amount)
1. Jeffrey -$241,400
2. Rosell -$213,492
3. Sarvas -$178,482
4. Powers -$161,376
5. Grabovoy -$160,342
6. Alonso -$158,750
7. Trapp -$154,192
8. Daniel -$147,281
9. Jacobson -$124,970
10. Johnson -$124,970
11. Osorio -$122,448
12. Thomas -$111,804
13. Chara -$108,866
14. Bernier -$108,655
15. Creavalle -$192,040
16. Juninho -$78,720
17. McCarty -$68,896
18. Koffie -$54,172
19. Laba -$50,580
20. Alexander -$26,653
21. Beckerman -$13,666
22. Kitchen +$11,298
23. Cronin $18,756
24. Carroll $38,164
25. Reo-Coker $38,202
26. Larentowicz $41,644
27. Alex $42,838
28. Evans $53,328
29. Caldwell $53,329
30. Mirda $56,434
31. Watson $60,171
32. Tchani $88,092
33. Mejia $93,500
34. Clark $97,986
35. Zusi $163,282
36. Arnauld $172,072
We've shared our thoughts. What do you think?