Was Kris Boyd really the 15th Most Productive Player in MLS last year?

Most everyone, to include the Front Office, is talking about recent transactions for Portland Timbers and I will too as time passes and the dust really settles. In the meantime something else sparked my interest a wee bit; was Kris Boyd, like the MLS Castrol Index indicates, really the 15th most productive player in all MLS last year? To answer that I first need to set the stage for your consideration.

1. I have three sources of data/information for this analysis. My first/primary source of data is the MLS Castrol Index; my second source of data is the standard MLS player data made available through the MLS Soccer team web sites; my third source of data are the results of the MLS Awards given to players this year after voting by MLS Management, MLS Media, and MLS Players.

2. My first step will be to test the waters of the MLS Castrol Index outcomes by comparing them to the outcomes of MLS Awards. In other words, do a simplified comparison on the outcomes of pure statistical data versus Professional Judgment of MLS Management, MLS Media, and MLS Players.

3. My second step will be to test the results of my own additional analysis using the standard data points 1) Goals Scored, 2) Shots Taken, and 3) Minutes Played. This second step will compare Kris Boyd to players who finished above and below him in the MLS Castrol Index ratings as well as compare Kris Boyd to players on the Timbers who scored at least 2 goals this year.

4. Finally, but foremost in order to begin, I will provide you information about what the MLS Castrol Index is supposed to provide using this link as my specific source for information: http://www.mlssoccer.com/castrol/faq

The MLS Castrol Index frequently asked questions along with their answers:

1. What is the Castrol Index? The Castrol Index is the Official Performance Index of Major League Soccer and uses the latest technology to objectively analyze player performance to help you see soccer from a completely different perspective.

2. How is the Castrol Index calculated? The Castrol Index tracks every move on the field and assesses whether it has a positive or negative impact on a team’s ability to score or concede a goal.

3. A key factor for all areas of performance in the Castrol Index is in which zone on the pitch the action takes place. Players receive points for each successful pass they complete, but the number of points awarded depends on which zones the ball is passed from and received in. Similarly, misplaced or intercepted passes are penalised depending on how much trouble the mistake is likely to land the team in.

4. The Castrol Index is also able to split up the rewards of a goal between penalising the goalkeeper for letting in a shot he should have saved and rewarding the attacker for scoring a goal. The number of Castrol Index points awarded for tackles, interceptions and blocked shots also depends on which zone they are made. Successfully taking the ball from a striker near the penalty spot will earn more points than a tackle out on the wing. Conceding free kicks and penalties will result in deductions.

5. Where does the data come from that powers the Castrol Index? Castrol uses the latest technology, working with the MLS and OPTA to log an average of 1,800 player movements per match. Castrol’s team of performance analysts then crunch all the data and award each player a Castrol Index score out of 1,000 – the higher the score the better the player’s performance.

6. Which players and matches does the MLS Castrol Index cover? Every match in the 2012 MLS season.

7. How often will the Castrol Index be updated? The Castrol Index will be update in the first week of each month and will be made available month via www.castrolsoccer.com and www.mlssoccer.com/castrolindex.</p>

8. Are there a minimum number of minutes a player has to play to be ranked in the Castrol Index? All players are ranked irrespective of how many minutes they have played in the season. Those who have not played a specified amount of minutes each month (equal to approx 60% of season game time) will have their Castrol Index score divided by that amount of minutes, therefore penalising players who are either injured, not selected or suspended for a considerable period of time.

9. Surely the player who plays the most games will top the Castrol Index? No, the Castrol Index is calculated on a per 90 minute score, measuring the quality of each player performance rather than the number of games played.

10. Are winning goals worth more than other goals? No, all goals scored are of equal value.

11. Monthly, Castrol’s team of performance analysts crunch all the data and award each player a Castrol Index score to determine the best performing players in MLS.

12. NOTE: The American spelling for the word ‘penalise’ is penalize; my assumption is that this product was originally created in Europe or Canada; not the United States… others may know better.

13. Although not a frequently asked question you can also access this website: http://www.castrolfootball.com/rankings/rankings/ to lookup Index scores for players in Primera Liga, Bundesliga, English Premier League, Ligue 1 and Serie A; so it is not just a tool used and recognized by MLS.

By the way; if you haven’t got a cup or pint of your favorite bevvy now would be a good time to do it…

Step 1: Comparing the MLS Castrol Index individual performance ratings of players against the MLS Awards; determined by combining votes from MLS Management, MLS Media, and MLS Players; in other words what I like to call “Professional Judgment”.

MLS Rookie of the Year was Austin Berry: The Castrol Index scored Austin with a 505 after logging 2520 minutes of play while netting three goals; a very good score but was it the highest among Rookies?

No… DC United’s, Nick DeLeon, scored a 601 while netting 6 goals and 4 assists; he also logged 2175 minutes of play.

No… Luis Silva (Toronto) scored a 559 with 5 goals, 5 assists in 2065 minutes played. Darren Mattocks (Vancouver) scored a 479 with 7 goals, 1 assist in 1299 minutes played while Connor Lade netted 3 assists in 2089 minutes played.

So when evaluating how well the Castrol Index was leveraged it appears that it may have been leveraged to determine an initial top 5 to vote from but it was not used as the definitive answer on who got Rookie of the Year.

Here’s the voting in case you missed it: http://www.mlssoccer.com/news/article/2012/11/19/fire-defender-berry-wins-mls-2012-rookie-year-award

Score: Castrol Index (Nil) – Professional Judgment (1)

Allstate MLS Goal Keeper of the Year was Jimmy Nielson. The Castrol Index scored Jimmy with a 646 after logging 3060 minutes of play; a very high score and the highest amount of minutes played by anyone in MLS this year; but was his Castrol Index score the highest among Goal Keepers this year?

No… Both Michael Gspurning (Seattle) and Troy Perkins (Montreal) scored higher; 695 for Gspurning and 668 for Perkins. Perkins did not even finish in the Top 5 vote getters while Gspurning finished 3rd. Dan Kennedy (575) received the second highest vote total while Nick Rimando (623) finished 4th and Andy Gruenebaum (636) finished 5th.

Here is the link in case you missed it: http://www.mlssoccer.com/news/article/2012/11/28/sporting-kcs-nielsen-voted-allstate-goalkeeper-year

So when evaluating how well the Castrol Index was leveraged it appears that it may have had relevance in picking out who might be worthy to begin with but the scores achieved did not mirror the voting totals.

Score: Castrol Index (Nil) – Professional Judgment (2)

MLS Comeback Player of the Year was Eddie Johnson. The Castrol Index scored Eddie with a 686 after logging 2120 minutes and scoring 14 goals with 3 assists. Other candidates were Chris Pontius (DC), Alan Gordon (SJ), David Ferreira (Dal), and Steve Zakuani (Sea). This was a close vote; the Clubs voted in favor of Chris Pontius while the Players and Media voted in favor of Eddie Johnson.

Here’s the link for results on this award: http://www.mlssoccer.com/news/article/2012/11/27/seattles-johnson-earns-mls-comeback-player-year

The Castrol Index scores, goals, assists, and minutes played are provided respectively for each player that finished in the top 5: {1} Johnson (686, 14, 3, 2120); {2}Pontius (661, 12, 4, 2336); {3}Gordon (712, 13, 7, 1299); {4} Ferreira (454, 2, 9, 1427); and {5} Zakuani (99, 1, 1, 319).

So when evaluating the Comeback Player of the Year the Castrol Index winner would have been Alan Gordon. Therefore the Castrol Index did not reflect an outcome relative to the winner.

Score: Castrol Index (nil) – Professional Judgment (3)

MLS Defender of the Year was Matt Besler. The Castrol Index scored Matt with a 640 after logging 2769 minutes with 2 assists. Other top 5 candidates receiving votes were {2} Bernardez (524, 2 goals, 1 assist and 2088 minutes); {3} Collin (610, 3, 1, 2488); {4} DeMerit (523, 1, 1, 2725); and {5} Valdes (488, 2, 0, 2921).

This was a very close vote; here is the link to view the vote totals: http://www.mlssoccer.com/news/article/2012/11/20/sporting-kcs-besler-named-2012-defender-year

What is unusual about this is that Nat Borchers (587, 2, 0, 2340) didn’t finish as a Top 5 vote getter, yet he finished 10 places higher than Bernardez, 11 places higher than DeMerit and 22 places higher than Valdez.

So when evaluating the winner versus the Castrol Index it appears that the Castrol Index was on-target; but when viewing the whole group it appears that the Castrol Index didn’t reflect the will of the voters.

Score: Castrol Index (nil) – Professional Judgment (4)

MLS Best XI (read more about it here): http://www.mlssoccer.com/news/article/2012/11/26/sporting-kc-land-four-players-2012-mls-best-xi

G – Jimmy Nielsen, Sporting Kansas City (already determined)
D – Victor Bernardez, San Jose Earthquakes
D – Matt Besler, Sporting Kansas City
D – Aurelien Collin, Sporting Kansas City
M – Osvaldo Alonso, Seattle Sounders FC
M – Landon Donovan, LA Galaxy
M – Chris Pontius, D.C. United
M – Graham Zusi, Sporting Kansas City
F – Thierry Henry, New York Red Bulls
F – Robbie Keane, LA Galaxy
F – Chris Wondolowski, San Jose Earthquakes

Formation (3-4-3): First off, MLS apparently decided to identify 3 Defenders, 4 Midfielders, and 3 Strikers (Forwards) for its BEST XI; while I don’t disagree with the idea of identifying more attacking players (per-say) I would offer that the initial filter of selecting 3 Defenders over 4 Defenders was most probably a “Professional Judgment” call on the part of MLS because most teams in MLS run a flat back 4.

For what it is worth I would recommend that the MLS select four defenders next year as opposed to three forwards since that formation better represents the formation of play used by the overwhelming majority of MLS Teams.

Defenders: The 3 Defenders selected were Victor Bernardez, Matt Besler, and Aurelien Collin. The Top 3 Defenders according to the Castrol Index are: Besler (640, 0, 2, 2769); Collin (610, 3, 1, 2488); and Borchers (587, 2, 0, 2340).

Victor Bernardez was actually 15th in the Castrol Index list for top defenders; Marshall, Hainault, Taylor, Hurtado, Hedges, Boswell, Cesar, Parke, Friedrich, Holgersson, and Myers all finished higher than Bernardez in the Castrol Index.

So in evaluating the overall information it would appear that the Castrol Index totals didn’t reflect the will of the voters.

Score: Castrol Index (nil) – Professional Judgment (5)

Midfielders: The 4 Midfielders selected were Alonso (562, 1 goal, 2 assists, 2687 minutes); Pontius (661, 12, 4, 2336); Donovan (660, 9, 14, 2256); and Zusi (586, 5, 15, 2789). The top 4 Midfielders in the Castrol Index this year were Pontius; Nysassi (661, 6, 3, 1720); Donovan, and Castrillon (647, 8, 4, 2413).

All those players, with the exception of Nyassi, bagged over 2000 minutes of playing time. The next player in the Castrol Index queue who bagged over 2000 minutes played, but didn’t get selected, was DeLeon (601, 6, 4, 2175); the player who finished second in Rookie of the Year Voting. It is interesting that DeLeon scored higher in the Castrol Index than Berry, Zusi and Alonso; yet he didn’t receive any honors from MLS.

So when evaluating whether or not the Castrol Index had a weighted impact on the final voting for the BEST XI Midfielders it appears it didn’t. NOTE: MLS pretty much indicated that Alonso was a ‘defensive midfield’ selection; oddly enough the MLS Castrol Index is supposed to take into account defending as well as attacking performances when rating players. If true then the selection of Alonso, who finished behind 12 other Midfielders in the MLS Castrol Index, should not even have had a look-in for the Award.

Score: Castrol Index (nil) – Professional Judgment (6)

Forwards: The 3 Forwards selected were Henry (796, 15, 12, 2107); Keane (854, 16, 9, 2520); and Wondolowski (836, 27, 7, 2813).

The top 3 Castrol Index players were Keane, Wondolowski, and Saborio (811, 17, 3, 2389); Thierry Henry finished 4th in the Castrol Index listing.

Not much to offer here other than Saborio makes about $5.2M less than Henry and he doesn’t play in New York; one of the largest spending MLS clubs this year.

So when evaluating whether or not the Castrol Index had a weighted impact on the final voting for the BEST XI Forwards it appears it did; but only with respect to Wondolowski and Keane.

Score: Castrol Index (nil) – Professional Judgment (7)

Conclusion on ‘testing the Castrol Index waters’: It appears there is compelling information to indicate that the MLS Castrol Index does not reflect outcomes when compared to the Professional Judgment of MLS Management, MLS Media, or MLS Players. It may have been used as a place to start but there is no compelling evidence to indicate that assumption when viewing the MLS Awards information.

Step 2: Using existing MLS data already available on all MLS Team websites: 1) Goals Scored, 2) Shots Taken, and 3) Minutes played. To set the conditions for this next step I am offering up two intuitive conclusions first:

1. Forward’s “productivity” is measure of the ratio between Goals Scored versus Shots Taken; nothing more nothing less; less is better than more. The intuitive part of this is that a striker who scores more goals with fewer shots taken is more productive than another striker who doesn’t.

2. Forward’s “efficiency”, by the team, is a measure of the ratio between Shots Taken and Minutes Played. The intuitive part of this is that a striker who has more shots taken in fewer minutes played better represents having a team that is more efficient in doing what they need to do; create goal scoring opportunities.

So here’s what my data collection showed and indicated: Name (Minutes Played, Castrol Index Score (Place), Shots Taken, Goals scored, Ratio of Goals Scored versus Shots Taken (Productivity), and Ratio of Minutes Played versus Shots Taken (Efficiency of the Team)

Robbie Keane: (2520, 854 (1st), 94, 16, 17%, 3.73%)

Chris Wondolowski: (2813, 854 (2nd), 127, 27, 21.3%, 4.51%)

Thierry Henry: (2107, 796 (4th), 88, 15, 17%, 4.18%)

Alvaro Saborio: (2389, 811(3rd), 87, 17, 19.5%, 3.64%)

Kenny Cooper: (2507, 720(9th), 84, 18, 21.4%, 3.35%)

Kris Boyd: (1891, 675 (15th), 63, 7, 11.1%, 3.33%)

CJ Sapong: (2208, 658(20th), 61, 9, 14.8%. 2.76%)

Will Bruin: (2501, 631(29th), 78, 12,, 15.4%, 3.12%)

Fabian Espindola: (2292, 576(39th), 69, 9, 13%, 3.01%)

Omar Cummings: (1995, 549(58th), 76, 6, 7.9%, 3.8%)

Chris Rolfe: (1759, 513(89th), 58, 8, 13.8%, 3.30%)

Oddly enough the top three strikers in this fundamental analysis are Chris Wondolowski, Thierry Henry, and Robbie Keane… Wondolowski had his team provide him 127 shots taken (opportunities) in 2813 minutes played for a ratio of 4.51%, Henry was second with a 4.18% and Keane was third with a 3.73%. Those same three Forwards were the top three Forwards voted by MLS Management, MLS Media and the MLS Players.

So although Saboria had a better Productivity ratio (19.5%) than Robbie Keane (17%) and Thierry Henry (17%) he was less efficient when it came to minutes played versus goal scoring opportunities created by his team.

So, in effect, Keane and Henry were better in using less service to create more goals. Therefore the efficiency rating created using more simplified data points provided a better representation of who won the top three Forward awards for MLS this past year.

Castrol Index (nil) My MLS Intuitive Data Analysis (1)

And when looking at all the sample data provided above, using the same intuitive approach, there is clear and compelling data to indicate that Chris Rolfe, identified as the 89th most productive player in the MLS Castrol Index was actually more productive, while also having less service, than Kris Boyd.

In addition, Will Bruin, CJ Sapong, and Fabian Espindola, who were also rated lower than Kris Boyd on the Castrol Index ended up being more productive while also having less service than Kris Boyd.

Castrol Index (nil) My MLS Intuitive Data Analysis (2)

Now how does the data stack up when using the same data points for players on Portland Timbers?

Darlington Nagbe (2777, 456(154th), 50, 6, 12.0%, 1.80%)

Bright Dike (896, 377(227th), 29, 5, 17.2%, 3.24%)

Jack Jewsbury (2765, 517(88th), 35, 3, 8.6%, 1.27%)

Danny Mwanga (740, 335(245th), 13, 3, 23.1%, 1.76%)

Kris Boyd (1891, 675(15th), 63, 7, 11.1%, 3.33%)

For Portland Timbers, the most productive striker this year was Danny Mwanga (23.1%), next was Dike, then Nagbe, then Boyd.

For Portland Timbers, the most efficient striker this year was Kris Boyd; so the idea that Kris Boyd was not provided good service from his team this year is not true. He was provided the best service of any player on the team; and only Jack Jewsbury was less productive Goal Scoring versus Shots Taken.

Castrol Index (1) My MLS Intuitive Data Analysis (3)

In conclusion:

1. Some managers succeed and some managers don’t; most look to succeed based on their individual choices of what players they want to work in the system of play they want. Critical to any system of play is service to strikers; Kris Boyd had service with Portland (compared to other Timbers players) and didn’t deliver. Regardless of philosophies about ‘systems of play’ John Spencer wanted Kris Boyd to help this team succeed; Boyd didn’t succeed this year and Spencer paid dearly for that decision.

2. Like Spencer, Caleb Porter is now making decisions on what players and system of play he wants to run in order to make the Timbers successful on the pitch; getting players to provide better service for his strikers will be the bottom line measurement stick on that success. Passes completed has value, but only with respect to creating a goal scoring opportunity. Others may have heard this phrase before “possession with purpose”; there is no other reason for possession with purpose when a game is nil-nil other than to create a goal scoring opportunity that might transform the game to 1-nil.

3. When 1-nil is achieved then “possession with purpose” also changes; it now includes the intent to possess the ball so that the other team cannot possess it to create that goal scoring opportunity that will change the game from 1-nil to 1-1. This same intuitive logic applies as the score-line continues to change.

4. The difference between managers is not just motivation it’s how that manager simply decides to take an outcome of nil-nil and transform it into three points; how the team creates goal scoring opportunities varies through long balls, short passes, free kicks, penalties, etc… variation in that approach to creating goal scoring opportunities creates havoc with defenses. The more variation a team and manager use in penetrating the attacking third the better; provided the manager selects the right talented players to do that.

5. Bottom line at the bottom: An Index for individual success of a team doing the small things will never provide value and benefit when considering the bottom line in this game. There are two individual statistics that matter in this game: 1) Strikers abilities to convert a goal scoring opportunity to a goal, and 2) A Goal Keepers abilities to deny a goal relative to shots on goal. There are two team statistics in this game: 1) The combined abilities of 10 players being able to create a goal scoring opportunity and, 2) The combined abilities of 10 players being able to deny goal scoring opportunities for the opponent.

6. I have developed a ‘team index’ to measure the success of a team to create goal scoring opportunities; I will use that ‘team index’ next year to see how it plays out. There is no need to create a new individual Index as MLS already charts and publicizes the critical data points in that Index; Shots Taken plus Goals scored for Strikers and Shots on Goal faced and Goals Scored Against for Goal Keepers.

7. I would submit there is compelling evidence for the MLS Castrol Index folks to revisit their algorithms in how they determine performance and productivity. This information, though pointed at Kris Boyd is not an editorial attack on Kris Boyd, (though some may view it that way); it is an assessment, using specific data publicized by the MLS Castrol Index, to assess the value of the Index compared to 1) Professional Judgment and 2) Other, more easily obtainable data, that can be used to ‘intuit’ an outcome that most likely has more relevance to management and fans as they make decisions about their team and its performance.

8. My own analytical approach to determine ‘productivity performance’ by a striker comes down to recognizing two things: a) The team provides the striker opportunities for shots taken and b) Can the striker convert those shots taken to goals? In simplified terms there are no other important statistics to consider.

9. If Kris Boyd stays with the Timbers next year you can bet that he knows he needs to score more goals relative to his shots that are taken; he had the best service on the team and didn’t execute, his goals scoring percentage must increase given an expected higher rate of increased opportunities.

10. Data remains, and always will remain, one tool that management can use in its decision making matrix; clearly the indicators in this analysis point to the need for Kris to increase his productivity in order to be more successful as a striker/soccer player.

11. When looking at the initial question and the outlay of information presented, Kris Boyd is NOT the 15th most productive player in MLS and the number crunchers working in the MLS Castrol Index arena need to revisit how they develop their data collection and analysis/algorithms to determine performance and productivity…

12. In viewing those same three statistical data points, and their ratios, the most productive (2nd) and most efficient team (2nd) in MLS last year was San Jose; that should be Capt. Obvious to most but what these ratios provide is actually relative to the overall outcomes of the teams.

13. So how about LA Galaxy, well LA finished second overall by being 3rd best in productivity and 4th best in efficiency. The top 8 teams in total productivity and total efficiency were San Jose, LA Galaxy, DC United, Chicago, Seattle, Houston, New York, and Kansas City. The four worst teams were Chivas, Philadelphia, Vancouver and Portland.

14. So apart from Vancouver the statistics pretty much hold true that those who finished higher in these two ratios finished higher in the standings. I would say ‘bully to Vancouver’ for making the playoffs as their overall productivity was 11th best while the overall efficiency was 19th… in other words they created the fewest goal scoring opportunities with a high minutes played; but when they did create a chance they delivered!

15. Overall, Portland did just the opposite; they created many more chances to score goals (12th overall) but failed to finish (17th in productivity).

16. Is it reasonable to expect that Portland Timbers might sign a new Striker this year? Will Valencia step in? Will Boyd stay and turn things around? How about Mwanga, he’s already got a Productivity rating of 23%. I’m still waiting for the dust to settle but it seems to me Caleb Porter is already doing the right thing; bolstering the Fullbacks to cut down goal scoring opportunities (efficiency) by the opponent. Perhaps we see a new Center-Back or perhaps another Midfielder (to add another dimension to creating goal scoring opportunities)?