Friday, October 12, 2007

DCU use of Roster

DCenters highlighted (The Sour Smell of Success) the problems MLS teams have had translating regular season success into post-season success. The question "D" poses is:

"If we look at MLS Regular Season points per game (modified as above to account for the shoot-out), there's something shocking. Your top 5 historically have a grand total of 0 MLS Cups. Those teams are the 1998 LA Galaxy (2.19 ppg), the 2001 Miami Fusion (2.04 ppg), the 2005 Quakes (2.0 ppg), the 2001 Fire (1.96 ppg) and the 1996 Mutiny (1.91 ppg). Is there something about the regular season, and really tearing it up that makes playoff success difficult?"

With DCU looking at a 2.0 ppg if they win out should we be worried?

My gut reaction is that this post season problem is due more to team management than anything else. DCenters feels that it is random chance.
My thought is that less depth in your roster and a willingness to use it will cost you in the end (Look no further than United last season or the Revs every season.)

I looked at the top five teams and this season's DC United. I just looked at season minutes available versus minutes played by each player. This was a very cursory look, not adjusting for late-comers to teams or injuries, just Team's total minutes played versus total Team's minutes available. Here is the table:

If you look at this distribution a couple of things jump out at me. First and foremost that the Black and Red have only three players that have played eighty percent or more of the available minutes (Perkins 96%, Emilio 90%, Gomez 83%). The others that failed to bring home a Cup have 6-8 players playing 80% plus minutes. The other item is that United has 15 players that have played 30% or more of the available minutes, more than any other team on the list (Fire-14, Quakes-14, Mutiny-13,LAG-13, and the poor Fusion-12). I feel this distribution of playing to is going to ensure us a fifth star and Tom Soehn COY.

EDIT:The current Revs roster has 10 players playing 80% of the minutes and last season's United had 8.


The great thing about this game is that it is not baseball; statistics really say little about the game. I love the fact that three knowledgeable fans can go to a game, sit together and all see a different game! So enough with the stats, my gut tells me our management of the roster this year will pay big!

2 comments:

D said...

Guy -- Great work. excellent insight, which never occurred to me to make.

And you're right, the SABR models don't transfer to soccer (we analyzed the concept of Pythagorean expectation, and showed it doesn't transfer over about a year ago). Which is something to celebrate (of course, if they did transfer, I would celebrate that as well.)

But this is excellent work on showing why the "rule" may not be the "rule", or rather, what the true "rule" is or might be.

Amazed Daily said...

"But this is excellent work on showing why the "rule" may not be the "rule", or rather, what the true "rule" is or might be."

I'm guessin', you're working inside the beltline!