Sabermetric Stats for Managers

by M.C.Teufel on Apr 17, 2010 2:53 AM EDT

After having to deal with one bone-headed decision of Jerry Manual after another I was wondering if anyone has seen any sabermetric work done on the work of managers. I googled but came up short (the best I can find was Chris Jaffe's work on Manger tendencies: http://www.hardballtimes.com/main/article/evaluating-mike-scioscia-and-terry-francona/).

It seems that the common perception of managers is that they have "a feel of the game situation" that some robot with a print out can't see with just stats. If this is the case than it must show itself in actual results. If there are actual results than over the course of a 162-game season in a game like baseball there must be actual stats to back it up.

What I'm envisioning is something of the following:

1) We now have quite a bit of data to make this possible. We have live game win probability, leverage situation stats, sand seasonal projections like ZIPS. All these stats can be combined to can estimate what the probability of any manager change is.

2) We compare the optimal expected results would have been based on what the actual results were in the game.

For example removing Perez for Nieve in yesterdays game. Prior to the change we can project what the likely outcome would have been had Manual left Perez in or had used another pitcher. We can subtract the difference between what the system indicated the optimal choice would have been versus what the manager's choice led to (grand slam given up by a pitcher lent to us from the Mexican league). Over the course of a year we would have thousands of manager decisions that can be weighed and analyzed. It won't take that many months before we have very solid stats to base a manager's performance on. The stat could be similar to WAR and can be called "Wins Managed Over Replacement" or perhaps "Decision Wins", "Wins Over Jerry", or "Gangsta Wins."

I think the time is nigh for managers to be held accountable for their decisions in terms of actual wins or losses they bring to the table and with the level of stats we have now I don't think it we are that far off to be able to pull something like this off. I'm not a statistician so perhaps someone who is from SBNation can take this on.

This FanPost was contributed by a member of the community and was not subject to any vetting or approval process. It does not necessarily reflect the opinions, reasoning skills, or attention to grammar and usage rules held by the editors of this site.

Tweet Comment 2 comments | 0 recs |

Do you like this story?

Comments

That sounds very

Un-sabermetric.

After all, if we use the Perez game as the example, does any fault lie in the catcher that calls the pitch? Perhaps those in the scouting department who determined Lopez’ batting chart?
Why you’re attaching Manuel’s name to the result seems a little misplaced. Perez hadn’t pitched more than 6.1 IP in a game since September 14th, 2007. It seemed like a good spot to pull him.
Where the blame should probably lie squarely is Omar, after all, it’s the pitcher’s job to produce, Jerry doesn’t throw a pitch. It’s Omar’s job to equip the bullpen with capable arms. He obviously hasn’t done so.
Ironically, Felipe Lopez, the very reason we’re having this discussion, could’ve easily been signed by the Mets for cheaper than Cora had the role been in different hands.
It saddens me that people would remove a manager before Omar in this case. Haven’t people realized from going after Willie Randolph’s head?

by sagecoll on Apr 17, 2010 7:39 PM EDT reply actions

I believe btb did one after the 08 season

I don’t know if they did one after last season also. You could probably search their archives for it, I don’t have a link.

And to top off their greatest season yet the new jersey nets scored 86 points...in double overtime. yes a professional basketball team only mustered 86 points in 58 minutes of basketball.

by Gina on Apr 18, 2010 2:52 AM EDT reply actions