from Amazon
Wednesday, August 16, 2006
Using Our Baseball Guts
Can we use our guts to figure out how much a player can impact fielding? Sure…
The only number to remember is 70%. That’s the number of balls in play that is converted into an out. Using our guts, we construct this model:
Freq__Type of Play__Out Rate__Great Fielder__Bad Fielder
60%__Automatic____98.3%_____99.0%_________97.0%
10%__Some Effort__70.0%_____80.0%_________50.0%
10%__Alot of Effort__30.0%_____50.0%_________15.0%
20%__Highlight Reel__5.0%_____15.0%_________3.0%
100%__ALL________70.0%_____75.4%_________65.3%
What does this model tell us? The first thing we do is classify a ball in play into one of four categories: automatic out, requires some effort to get the out, requires alot of effort to get the out, and it’s almost impossible to get the out. That first column, “freq”, is simply my guts as to how often this type of play happens. I could be wrong, and so, you can feel free to update this column.
That third column is the average out rate, for each type of ball in play. Again, pure guts. However, when you multiply the out rate by the frequency for each of the four types of balls in play, you get a total of 70%. So, whatever numbers your guts are telling you, you have to stick to the 70% figure.
Now, how about a great fielder? I just picked some more numbers from my guts. I figured if an average player makes an out on 98.3% of routine plays, then a great fielder will do it on 99% of them. Anyway, I just threw a whole bunch of numbers there. The out rate for the great fielder is 75.4%, and for the bad fielder, it’s 65.3% In essence, plus/minus 5% per BIP.
The seven primary fielders (exlcudes pitcher and catcher) get about 600 BIP per 162 games each. Plus/Minus 5% of that is about 30 plays. That’s the range.
If you think the range should be higher, or lower, then you have to come up with a model to support that. Go ahead, you may be right.
As well, while I used only four categories of plays, you can easily extend this to as many as you like. Just come up with the appropriate model. But remember to stick to the 70% rule.
Here’s an even more granular model:
Freq___Out Rate___Great Fielder___Bad Fielder___Type of Play
40%___100.0%___100.0%___100.0%___Automatic
10%___97.0%___99.0%___94.0%___Automatic
10%___93.0%___98.0%___83.0%___Automatic
5%___80.0%___90.0%___60.0%___Some Effort
5%___60.0%___70.0%___40.0%___Some Effort
5%___40.0%___60.0%___20.0%___Alot of Effort
5%___20.0%___40.0%___10.0%___Alot of Effort
10%___10.0%___30.0%___5.0%___Highlight Reel
10%___0.0%___0.0%___0.0%___Highlight Reel
100%___70.0%___75.7%___64.7%___ALL
In this case, I have 9 categories of BIP. Results are similar.
I find I’d have to make very large concessions to get it to just plus/minus 7%, which means plus/minus 42 plays.
It’s important to note that the SS is involved in about 30% more plays, and therefore, when I say the average position would see players plus/minus 30 plays, at SS, it would be plus/minus 40 plays.
Similarly, 42 plays for the average position is 55 plays at SS. That, I think, is really the absolute limit.