Question ... Which would you rather have your team do, with RISP, on an average night: go 3/9 or go 4/16? In the first scenario, your hitters are coming through at a .333 clip whereas in the second one you are frustrated by failure 75% of the time. So in the first scenario it feels like your team is actually pretty darn good in those RISP situations, while in the second one it feels like they aren't and you have to endure 12 times/game that your team fails with RISP. Ouch.
It's a trick question, of course, because in the first scenario your team is getting 3 hits/game with RISP while in the second one your team is getting 4. You're going to score a lot more runs getting 4 hits/game with RISP than you are getting 3 hits/game.
We're used to gauging numbers as percentages, not absolutes, for example preferring -- for good reason -- a player with a .333 OBP to one with a .250 OBP. Yet Bob Geren said something really interesting on the manager's show the other day, explaining why he liked having Rajai Davis leadoff, and the more I think about it the more intelligent it sounds to me.
Geren said, essentially, that if Rajai Davis came up 5 times in a game, instead of 4, there were that many more chances for him to reach base and do his thing -- which even his most ardent detractors would have to admit he has done well this season, stealing a league high 22 bases while being caught only twice, and creating several "Rajai runs" in other ways. Let's look at this a bit closer...
Let's assign Rajai a .300 OBP. That's on the low side, as his career OBP is .330 and even this season, as bad as he has hit for most of it, he now has his OBP up over.300. But let's go with .300, as a reasonable "worst case scenario."
With a profoundly uninspiring .300 OBP, if Rajai bats 5 times, instead of 4 times, there is a 5% better chance (31%-26%) that he will reach base twice, a 5% better chance (41%-36%) that he will reach base once, and a 7% less chance (17%-24%) that he won't reach base at all. Each time he reaches, Rajai has a chance to seriously create, to disrupt, to skidaddle around the bases as only Rajai can do.
What Geren is saying is that while on a percentage basis, Rajai's odds of reaching don't justify batting him leadoff, on a "number of total times reaching base" basis, it can still be worthwhile. And this actually makes sense if you consider two things: How much good Rajai does when he reaches, and the fact that the PAs he takes away from others are mostly being taken away from less-than-great batters (e.g., Pennington, Ellis, Patterson, Gross) who don't even slug well enough to justify making sure they get another PA.
It's really the RISP example: Geren is arguing that even with Rajai's low success rate of reaching, if he gives Rajai more chances (more PAs), he'll come out ahead because Rajai will be on base more times. And let's look at an example where Pennington, or Ellis, or Gross, or whoever, can maintain a .350 OBP, but gets fewer PAs because he's batting lower in the order so that Rajai can leadoff. What about the fact that Rajai is taking PAs away from a hitter who gets on considerably more often than Rajai does?
If over the course of, say, 22 games or so, Rajai gets on 30 times out of 100 (.300 OBP), and the guy bumped to the lower part of the order reaches 32 times out of 91 (.350 OBP), it's better than that guy reaching 35 times of 100 (.350 OBP) while Rajai reaches 27 times out of 90 (.300 OBP). Because in both scenarios you wind up with 62 times reaching base -- but in the first one you have Rajai as the baserunner 3 more times.
It's counter-intuitive to put a .300ish OBP guy leadoff and a .350ish OBP guy in the bottom 1/3 of the order -- and that isn't even a risk, as the only batters capable of maintaining a .350 OBP are probably Barton, Sweeney, and Cust, all of whom can bat in the 2-6 spots where Rajai won't be anyway -- but Geren's point about worrying about Rajai's "times on base" instead of "rate of getting on base" is well taken, especially on a team that doesn't have a lot of ways to score runs.