Saturday, October 27, 2012

PART 2: Preliminary Evaluation, Crew

In my last post I took a good look at how players were doing in relationship to the Team as a whole. Looking at it that way sort of gives me an idea on which parts directed the 2012 Columbus Crew.

In this post I'm going to look at the similar topics but instead focus on the average number of points a player earns when starting. The way it works is similar to the way a pitcher earns a win in baseball, with a slight difference; I'm keeping the end match result with the starter even if different when he left the field - whereas a Pitcher only gets a result if it is the same as when he was taken out. Why, you ask?

The reason for this is three fold:

A. Much less scoring in soccer so the impact of the subbed player in soccer doesn't single handedly control the result of the match like a middle reliever or closer can, unless you have a guy named Messi on your bench but even then it would be a statistically rare event.
B. Play / pace / flow of soccer match is dictated by starters. The starters do most of the heavy lifting (good or bad). Much of the match determined by lineup.
C. I only have so much time, cut me some slack! As with any statistic it isn't the exact number that matters, the value in the number is to tell you whether the player contributing positively or negatively. Going out 2 or 13 decimal points is for mathematical engineers who have much more data to digest (and get paid).

I understand if the way I am measuring this is a turnoff, but I find the results valuable and reliable. I'm always trying to find different ways of measuring player and team results. Sometimes it works, sometimes not but always learning as I go.

Now, let's take a look at Average Points Earned per Game (PPG) when each individual player starts against team ranked in the Top Half of the League Table (2012). Why: I want to know who performed best against the best. The Crew finished smack dab in the middle of the table (10th).

2012 PPG v Top H : NAME : GMs STARTED v. Top H

2.20 : Cole Grossman : 5
1.75 : Carlos Mendes : 8
1.75 : Dilly Duka : 8
1.60 : Nemanja Vukovic : 5
1.25 : Emilio Renteria : 8
1.25 : Sebastián Miranda : 16
1.25 : Joshua Williams : 16
1.20 : Justin Meram : 10
1.20 : Chris Birchall : 10
1.18 : Eddie Gaven : 17
1.18 : Andy Gruenebaum : 17
1.14 : Tony Tchani : 7
1.14 : Danny O´Rourke : 7
0.91 : Chad Marshall : 11
0.90 : Jairo Obando Arrieta : 10
0.80 : Julius James : 5
0.44 : Milovan Mirosevic : 9

(Here are the guys with less than 5 starts)
1.50 : Shaun Francis : 2
1.00 : Bernardo Anor : 3
0.67 : Olman Vargas : 3
0.33 : Federico Higuaín : 3
0.33 : Eric Gehrig : 3

I talked about this in my last post but worth mentioning again because the Crew really struggled against top competition this year. Particularly notable at the bottom are Marshall, Arrieta, Mirosevic. James only had 5 starts, so like Grossman at the top, it should be only really looked at as likely trend.

On the positive side I like Mendes and Duka along with Miranda and Williams as the solid performers against the top. Remember, this is only against teams 1-9 on the table.

-----
LOOKING BACK TO MOVE FORWARD

To put some of the 2012 numbers in perspective I pulled the results for last year for each player when they faced a team in the top half. It worked out to be teams 1-8 on the table since the Crew were 9th.

2011 PPG v Top H : NAME : GMs STARTED v. Top H

2.00 : Rich Balchan : 8
1.86 : Kevin Burns : 7
1.73 : Eddie Gaven : 11
1.57 : Emmanuel Ekpo : 14
1.44 : Chad Marshall : 16
1.43 : Tom Heinemann : 7
1.43 : Julius James : 14
1.40 : Dejan Rusmir : 5
1.40 : Emilio Renteria : 5
1.36 : Joshua Gardner : 11
1.35 : Sebastián Miranda : 17
1.25 : Will Hesmer : 16
1.14 : Andrés Mendoza : 14
1.08 : Robbie Rogers : 13
0.63 : Dilly Duka : 8

(Here are the guys with less than 5 starts)
3.00 : Justin Meram : 2
3.00 : Andy Gruenebaum : 1
1.50 : Bernardo Anor : 4
0.25 : Danny O´Rourke : 4
0.25 : Shaun Francis : 4

Looking at it this way you can see that the 2011 team was better at beating the best in Major League Soccer. 2012 only saw four players with over 5 starts >1.50 PPG. Curious, because as a team the Crew earned a higher avg. PPG in 2012.

The stand outs in 2011 featured mostly players that didn't play this year... Balchan, Burns, Ekpo, Heinemann, Rusmir. Interesting that only Balchan and Heinemann are still with the team (didn't play due to injury).

I'm sure there are a few rolling their eyes at Kevin Burns. The guy, for whatever reason, fell out of favor with fans and management last year. Numbers don't lie - your eyes do (especially after you have been drinking, evaluating from field level or the Nordecke and tweeting the whole time). No question that Burns contributed against quality opposition. Both he and Rusmir performed significantly better than their 2012 replacements (Birchall and Mirosevic). If the move away from Burns and Rusmir had to do with injury or contract than that is one thing, but to replace due to performance and/or results is another. If it is the latter, it was a failure. The time and effort to find, sign and move Birchall and Mirosevic could have been put into something more productive.

Here are the two lists (2011/2012) merged. PPG represents combined results from both years.

2011+2012 PPG v Top H : NAME : GMs STARTED v. Top H

2.20 : Cole Grossman : 5
2.00 : Rich Balchan : 8
1.86 : Kevin Burns : 7
1.75 : Carlos Mendes : 8
1.60 : Nemanja Vukovic : 5
1.57 : Emmanuel Ekpo : 14
1.50 : Justin Meram : 12
1.43 : Tom Heinemann : 7
1.40 : Dejan Rusmir : 5
1.39 : *Eddie Gaven : 28
1.36 : Joshua Gardner : 11
1.31 : *Emilio Renteria : 13
1.30 : *Sebastián Miranda : 33
1.29 : *Bernardo Anor : 7
1.28 : Andy Gruenebaum : 18
1.26 : *Julius James : 19
1.25 : Will Hesmer : 16
1.25 : Joshua Williams : 16
1.22 : *Chad Marshall : 27
1.20 : Chris Birchall : 10
1.19 : *Dilly Duka : 16
1.14 : Andrés Mendoza : 14
1.08 : Robbie Rogers : 13
1.00 : Tony Tchani : 8
0.90 : Jairo Obando Arrieta : 10
0.82 : *Danny O´Rourke : 11
0.67 : *Shaun Francis : 6
0.67 : Olman Vargas : 3
0.44 : Milovan Mirosevic : 9
0.33 : Federico Higuaín : 3
0.20 : Eric Gehrig : 5

* Started more than 5 games both years. Pulling them out looks like this:

1.39 : *Eddie Gaven : 28
1.31 : *Emilio Renteria : 13
1.30 : *Sebastián Miranda : 33
1.29 : *Bernardo Anor : 7
1.26 : *Julius James : 19
1.22 : *Chad Marshall : 27
1.19 : *Dilly Duka : 16
0.82 : *Danny O´Rourke : 11
0.67 : *Shaun Francis : 6

Out of the players to see significant time both years only Dilly Duka and Danny O'Rourke improved (Duka, remarkably so). I believe it has to do with how they were used last year to this. As in 2011 both only saw spot starts while 2012 saw them starting more (O'Rourke still shaky).

-----
CLOSING COMMENT
The achilles heel for the Crew this year was when regular starters, in particular; new signings late in the year, faced quality teams. The Crew faced teams in the top half of the table an equal 17 times in both 2011 and 2012. 2011 saw two more wins in those 17.

Perhaps this has to do with unbalanced scheduling or perhaps it just has to do with new players adjusting to new lives in a different city (or country). The answer isn't really clear.

What is clear is that I hope the team makes informed decisions about players going into this offseason. Personal feelings, vendetta's, cultural differences, revenge, fan outrage/favor or bias should always take a backseat to results.

Create a reliable and valuable measurement system - explain the system to your team - reward those who perform above expectations / punish those below.

Is solely looking at record against good teams the only tool to use? Of course not. Is it something that should be valued? I think so.

No comments: