clock menu more-arrow no yes mobile

Filed under:

Bill James Stuff and I stir the pot of lineup argument again

Bill James is great. He's so much fun to read. I love picking up the Win Shares book or the most recent Historical Baseball Abstract and just reading something at random. It's almost always good, even if it's something I don't quite agree with.

What a lot of Bill James fans will ignore sometimes is the fact that much of his baseball writing isn't scary at all to Average Fan. It's very very accessible. It's often funny. And a good amount of the time, has nothing to do with numbers or stats at all. He talks about which players were real good looking or real ugly. What kinds of uniforms people wore. At one point in the Historical Baseball Abstract, he writes openly about how great the player Joe Morgan was and how much of an idiot the announcer Joe Morgan is. Makes me laugh.

And when he does get into his big number theories, he'll always preface it with something like, "You can either read the next 5 pages, or you can take my word for it, but there are numbers ahead and it's gonna be a bear." Very user friendly.

So on davidlizerblog, he points out this Bill James interview. We were debating the merits of different lineup strategies recently, and I was saying somebody should try 3 sets of 3. Basically have somebody who can get on base and move, followed by somebody who can make contact, followed by somebody who can slug. Then repeat two more times. The NL screws it up by having the pitcher in there, but aside from that, it sounds reasonable to me. Mostly I was saying that they should just try something different. Mix it up. Do something crazy. Here's Bill James' take on it:

Anyway, I am unable to prove that batting orders are highly relevant. I think that you can be ESSENTIALLY just as successful writing the names down at random. You'd lose a little bit.
And the reason why should be obvious. The batter at the front of the lineup should be one of your best batters as he'll basically get the most AB opportunities. But really, aside from that, how badly can you screw up? Even your best hitters will accomplish nothing two out of three times, so there's never a good way to guarantee a good run scoring opportunity for the one out of three times that he does decide to hit the ball. On the flip side, we've all seen the worst hitter on the team or sometimes the pitcher have three for four days. Since you can't predict that and since the only time you can guarantee a lineup will show in a particular order is in the first, then why should it matter? (Sorry about the horrendous run-on sentences.)

Consider the fact that in the National League, the double-switch is an important strategy for a manager. Once you pull a double switch though, you've essentially mixed up your original formula. The way I figure it, viewing the lineup as three sets of three gives you offensive opportunities that wouldn't normally appear. How many times have you watched a game and groaned that the bottom of the order was coming up? Why not have your second best hitter back there somewheres so you have a spark in every inning?

In any case, I don't have any good way of demonstrating that this lineup idea would work, but it gives me a little comfort that Bill James isn't able to demonstrate that it wouldn't work.

Oh and the rest of the interview's good too.