from Amazon
Thursday, March 22, 2007
Double Steals
The always resourceful Dan Fox comes through again, this time on double steals:
http://www.baseballprospectus.com/article.php?articleid=6003&mode=print&nocache=1174595744
This part struck me:
Breakeven points:
Outs BE%
0 .639
1 .558
2 .735
Actual results:
Outs Succ Att Pct
0 638 1127 0.566
1 1587 2258 0.703
2 769 776 0.991
What does this mean? With 0 outs, the breakeven point is 64%, meaning that you should only try the double steal if you are going to make it at least 64% of the time (i.e., you are 64% to 100% certain to make it). The overall average was only 57%.
On the other hand, with 1 out, the minimum level of success is only 56% (i.e., run when you are 56% to 100% sure to make it). And, they were successful 70% of the time. So, it seems that just about all the time, they ran above the breakeven level. And this is when they ran the most.
With 2 outs, the minimum level of success is 74% (i.e., 74% to 100% is the range to run on). And, they were successful 99% of the time! This is rather shocking. On the 776 attempts, they were just about all “sure things”. And, remember, our breakeven requirement level was the highest.
It’s almost as if the fielding team, knowing that you’d have to be crazy to double-steal, put their fielding alignment in such a way as to make it easier to let the other team steal, since they expect so few steals, and instead concentrate on putting their fielders in the position to get the batter out at first.
You could spend all day looking at this stuff. And of course, it would get even more interesting if you look at win expectancy rather than run expectancy.
I think the .99 success rate is misleading. If you attempt a double steal with 2 outs and one of the runners is thrown out, the inning is over and (I assume) the other runner won’t be credited with a SB. So that case won’t look like a double steal attempt in the data.
I’m not sure how to account for the 7 recorded two out failures—probably oddball plays. Maybe cases where both runners are initially successful but the lead runner is then thrown out trying to go all the way home? Or the runner on 2B steals third and home on the same play, but trailing runner gets nailed at 2B?