Topic: Baseball game defense

So, the longer my White Sox' slump lasts, the more I seek diversion in "fictional" baseball -- namely, my ever-in-progress game.

Right now, I'm trying to work out a relatively simple defensive value that can be determined from easily-available historical stats, but is still somewhat normalized to the era. So far, I've gotten as far as setting team values by position. Let me explain:

First, you need to determine the number of "balls in play" (BIP) the team allowed. This is done by adding IPx3-SO and H-HR (i.e., the number of outs, minus strikeouts, plus the number of hits, minus home runs). Set this aside.

Next, determine the number of fielding chances for each position. In seven of nine cases, this is the traditional PO+A+E; however, there are two exceptions:

First Base = PO+A+E minus the assists total for the team's 2B, 3B, and SS
Catcher = PO+A+E minus the strikeout total for the team

Now, divide each position's chances by the BIP determined above.

As an example, let's take the 1906 Chicago Cubs (with Tinker, Evars, and Chance):

BIP = 4,469
Infield Assists = 1,212 (excluding 1B)

1B (Frank Chance)
PO = 1567
A = 91
E = 18

PO+A+E = 1,676 - 1,212 = 464

Divide this by the BIP and you get .104

2B (Johnny Evars)
PO = 351
A = 446
E = 44

PO+A+E = 841, divided by BIP = .188

SS (Joe Tinker)
PO = 308
A = 504
E = 48

PO+A+E = 860, divided by BIP = .192

Obviously, these numbers mean little by themselves; however, comparing them to the rest of the teams during the year, you find that the sum for Tinker-Evars-Chance (.484) is tops in either league, although not by much (Cleveland comes in at .480, while the St. Louis Browns are right behind at .479). Individually, Tinker (SS) is 8th, Evars (2B) 5th, and Chance (1B) 2nd. Clearly, the the sum was greater than the parts...

For another example, consider Derek Jeter. A lot has been made recently about him being the worst starting shortstop in the majors... by the numbers. Yet, he won a Gold Glove in 2005 (and 2004). What do these numbers say?

Well, in 2005, the Yanks had 4,638 BIP, and 776 chances at SS (assuming for the moment that the vast majority of them were from Jeter). This gives a defensive rating of .167-- good for 12th in the major leagues. So Jeter appears to be an average fielder. Not the worst, to be sure, but neither was he Gold-Glove caliber. (Who was? According to my numbers, the Minnesota Twins' defensive rating of .174 was tops in the AL -- but since several players split time there, it's hard to say whether any of them were the "best".)

I don't consider this a flaw, as it's generally accepted that the GGs are awarded on reputation rather than performance... but it's interesting, nonetheless. Any thoughts?

Daniel Kast
Majestic Twelve Games
cricket@mj12games.com

Re: Baseball game defense

Despite Kevin's "advice", I continue to plunk along... smile

As someone (Bill James?) has already developed the Defensive Efficiency Rating (DER), which is the ratio of outs recorded to balls in play, I thought a better way to express the number I'm looking for would be as a positional DER, the sum of which would equal the team value.

However, that means determining who is primarily responsible for recording each out made by a team. This estimated fielding chance (EFC) can be estimated as follows:

Catcher = (PO-Team SO)/2+A
First Base = Team Fielding Outs minus all other EFCs
Second Base = PO/3+A
Outfielders = PO+A
P, 2B, SS = PO/2+A

Using these numbers, we end up with the following splits since 1901:

P = 8%
C = 4%
1B = 10%
2B = 18%
3B = 11%
SS = 19%
LF = 10%
CF = 12%
RF = 9%

This "feels" right... as the left side of the defense is weighted slightly over the right side, and about 2/3 of outs originate in the infield.

Anyway, then continue as before; divide EFC by the team's BIP to get the positional DER. The sum of these numbers will then equal the team's overall DER.

For example, in 2005, the White Sox looked like this:

P (Various) = .053
C (Pierzynski) = .016
1B (Konerko) = .084
2B (Iguchi) = .123
3B (Crede) = .088
SS (Uribe) = .134
LF (Podsednik) = .074
CF (Rowand) = .091
RF (Dye) = .072

TOTAL = .734

With this modification, Jeter becomes the 13th-best SS last year, with our "Gold Glove" winners remaining in Minnesota and St. Louis.

Daniel Kast
Majestic Twelve Games
cricket@mj12games.com

Re: Baseball game defense

Dan:

Interesting (the gears in my head are trying to process all the stats).  I guess your stat translates into the "range" of the player to determine if he gets to a ball hit his way?

Eagerly looking for more!

Todd