A CITIZEN'S BLOG

The other day in my post on clichés I noted that one of the things I had always taken for granted was the fact that you have to steal bases and "manufacture" runs to be successful in baseball. Moneyball certainly shattered that (check out Chapter Twelve: chiefly Lewis' recount of the discussion between Ray Durham and the A's coaches about base-stealing and Beane's argument that the A's offense was better than the Twins in '02 because they sac bunted fewer times and attempted fewer steals, thus wasting fewer outs).

The June 27th Red Sox-Phillies game where Bowa sent Polanco to steal second with one out has reinforced my belief that Beane is probably right when he argues that base-stealing is too risky: Polanco was thrown out heading to second, so the Phils went from having a base-runner with one out to having no one on with two outs. Abreu's double, which could have scored Polanco (debatable, I grant you), went for naught as Thome struck out to end the inning. Had Polanco not tried to steal, the Phils probably would have scored a run to tie the game at 4-4, or had runners at second and third with one out. Thome's strikeout, instead of ending the inning, would have brought up Pat Burrell with an RBI opportunity.

So the A's argument that playing "stattion-to-station" baseball is a valid one because you run the risk that you are running yourself out of a big inning when you steal.

I wanted to test this assumption out, so over the weekend I looked up the top run-producing teams in the NL by looking at their runs created per 27 outs (RC27) stats on ESPN.com:

RC27: Top Five Teams... (as of July 6, 2004)
1. Phillies: 5.48
2. Colorado: 5.37
3. St. Louis: 4.27
4. Chicago: 5.14
5. San Francisco: 5.07

Here are the worst teams...

RC27: Five Worst Teams...
12. San Diego: 4.54
13. Milwaukee: 4.52
14. Florida: 4.50
15. Arizona: 4.47
16. Montreal: 3.75

What are the top teams attitudes towards base-stealing? Well, the Phils are 7th with 44 steals and have made fifty-eight attempts, which is middle-of-the-pack. The Cardinals have stolen 57 bases on 76 attempts (throwing cold water on my point a little), but then I noticed that the Giants are dead-last in the NL with just 21 steals (34 attempts), and the Cubs (24 on 41 attempts) and Rockies (25 on 45 attempts) are near the bottom in terms of steals too (13th and 12th respectively). It is a remarkable bit of information: three of the top five offenses in the NL are also three of the five most steal-adverse teams in the NL.

Conversely, offensively-challenged teams like the Brewers (81 steal attempts), Expos (70 attempts) and Marlins (67 attempts) try to steal a lot. (I note that the Fish's Jack McKeon has earned earned the ire of sabremetrics-types for dismissing the use of stats in the Sports Illustrated baseball preview issue). The sole exception to the rule seems to be the Cardinals, who hit with power and have speed to burn on the base-paths. Indeed, the Cardinals seem to be the exception to the rule. What got my attention was that the teams that fail to score runs not only run a lot, but that they are fairly successful at it: the Expos succeed 77% of the time, the Mets 79% (46 of 58), etc.

They steal, they are successful at it, and they still fail to "manufacture" runs.

(Walk, don't run!)

Pretty neat stuff. At a minimum it is another argument that the sooner Bowa ditches the "small ball", the better it is for everyone.

(A little off the subject of steals, but check out this piece from Hardball Times derriding ESPN's "Productive Outs" stat. Productive Outs are another no-no to sabremetrics believers: the most precious thing you have are your 27 outs, so why waste them playing for just one run? Larry Mahnken wrote an earlier piece on the subject too. Both are very good.)