Batting Order

[satchel]

In a thread about Giambi possibly batting 5th, SmedIndy said:

Originally posted by SmedIndy
Of course, you have read the study by James that it doesn't matter a tinker's cuss where people hit in a lineup.

I am not familiar with this study so I'd like to ask a question. Did James show that (a) it doesn't matter where you hit in the lineup, your performance will be the same (as in, there's no such thing as protection)? or (b) that it doesn't matter what order you bat your nine guys in, you'll get the same results? or (c) something else?

I can understand and believe (a), but anything I'd like to hear more about. In any event, as Steve pointed out in the Giambi thread, hitting earlier in the order translates to more plate appearances.

[Xanadu Dragon]

http://www.alumni.caltech.edu/~raj/writing/BattingOrder.html

http://www.pankin.com/markov/btn1191.pdf

are two studies on the issue.

[SmedIndy]

Originally posted by satchel
In a thread about Giambi possibly batting 5th, SmedIndy said:



I am not familiar with this study so I'd like to ask a question. Did James show that (a) it doesn't matter where you hit in the lineup, your performance will be the same (as in, there's no such thing as protection)? or (b) that it doesn't matter what order you bat your nine guys in, you'll get the same results? or (c) something else?

I can understand and believe (a), but anything I'd like to hear more about. In any event, as Steve pointed out in the Giambi thread, hitting earlier in the order translates to more plate appearances.

From what I read in his Managers book, he did some studies where things like batting Babe Ruth last and the pitcher first or third only decreased runs slightly over several seasons.

So basically, protection is a bunch of hooey, and you could bat the pitcher second and it really wouldn't matter too awfully much. But, its silly to do it because of the heat it would generate.

[satchel]

Originally posted by SmedIndy


From what I read in his Managers book, he did some studies where things like batting Babe Ruth last and the pitcher first or third only decreased runs slightly over several seasons.

So basically, protection is a bunch of hooey, and you could bat the pitcher second and it really wouldn't matter too awfully much. But, its silly to do it because of the heat it would generate.

I'm sorry, perhaps I am being obtuse, but I still think there are two separate issues in the above and I am trying to figure out which James is (and you are) really talking about.

Protection seems to be about how well an individual batter does, depending upon where he is placed in the order; the idea being that with runners on base ahead of him and/or a strong batter following, the player will see more fastballs around the plate and will therefore hit better. I am persuaded by various studies I have read that this concept is "hooey" as you say.

The second idea is that whether you bat Babe Ruth 4th or 2nd in the order has no bearing on how your team as a whole does. I am more skeptical of this idea and therefore would like to see it expanded if it is in fact true. it seems to me that batting Babe Ruth ahead of three .400 OBP place setters has to be better for the team (if not for Ruth's individual stats other than RBI) than ahead of three .300 OBP slugs. Moreover, batting him near the top of the order gets more plate appearances for his prodigious bat over the course of the season, and is also thus better for the team. This seems to me a distinct issue from protection.

Is my question clearer now?

[satchel]

In addendum to the above, it appears that the studies linked by Steve in his post address the second issue. Thanks Steve!

[Duque]

I think if you have a bunch of great players, or even just one great player - it isn't going to matter much where you put him in the batting order. But if you have, say, a bunch of average players, each with varying skill sets, than constructing a good lineup is probably crucial.

[SmedIndy]

I didn't have a chance to look at the study again, but I think James simulated a lot of seasons and put players in different spots and the net change in runs was minimal.

[BuzzBuzzard]

I have not read any of the studies on this, but if there is no noticeable difference in where guys hit, how come no one is willing to do something unconventional?

[SmedIndy]

You remember the howls of derision when LaRussa hit the pitcher eighth?

I am probably dramatically paraphrasing the study, but it a difference of a few runs per year. Which may matter, or may not matter.

The best order may be in descending OBP order, but that flies in the face of "conventional wisdom".

[Xanadu Dragon]

Brass tacks, I don't think anything could convince me that getting your best hitters as many PA as possible and your worst hitters as few PA as possible is not the way to go. I'm not saying Giambi and Bonds should hit leadoff - - but, it would be foolish to bat them 8th and think it would be the same as having them bat 4th. Conversely, having Rey Ordonez hit leadoff would hurt you more than having him bat 8th.

[SmedIndy]

XD -

I edited my post while you posted. No way would any sensible man hit Rally Rey Ordonez first, but then Tanner led off Omar Moreno.

[sweaver]

Yeah, but Omar was really fast! And he hit for a (usually) good average, although he never walked.

Best I could cull from the studies Steve gave links to, if you did something really crazy like bat Ordonez leadoff and Piazza ninth, it might cost you two wins a year. That can make a difference, but not in typical year. Barring that, lineup selection doesn't make that much difference, although certainly the best hitters should get the most at-bats. That stands to reason.

[BuzzBuzzard]

I think that makes my point. My conclusion, again without a whole lot of investigation, is that there might be something wrong with the study if the outcome of Ordonez batting lead off and Piazza batting ninth is two less wins. That just doesn't seem to be believable.

[Fuzzy Bear]

Remembering my "Bill James Baseball Abstracts" from the 1980s, I remember him arguing the idea of "protection" as being bogus. He used the example of Dale Murphy, and how he hit with Bob Horner behind him, and without Bob Horner (a year in which Horner was injured). There was, essentially, no difference in Murphy's performance; in fact, he was a little bit better the year Horner was out.

In those same abstracts, James discussed the need for batters batting early in the inning to have the ability to get on base. In this mode, he praised Ralph Houk for having the good sense to put Wade Boggs at leadoff, and not Jerry Remy (who was on base a lot less. despite his speed).

The moral of this story, I believe, is that your #1 and #2 guy should have high OBP. Your #4 guy should be your power man. Your #5 guy should also be a high OBP guy, because the #5 hitter IS LIKELY TO LEAD OFF THE 2ND INNING!!!!! Batting early in an inning requires different skills than batting late in an inning, and while the batting order cannot fully plan for who will hit in what situation in the 7th inning, a coherent batting order can serve to prevent unnecessary substitutions, and help keep rallies alive.

Giambi is a cleanup hitter; the best in the business. Why anyone would move his position in the order is beyond me.

[KCBOOMER]

I like a lot of Bill James' stuff but I am not buying this one. I do think it matters that your best hitters get more at bats than your weaker hitters.

For every position in the batting order you drop you lose plate appearances. If a guy should be hitting third and you put him ninth you just gave up 108 PA's. On my club you don't give Rey Ordonez 108 PA's at the expense of Babe Ruth.

[sweaver]

The 1959 White Sox won the pennant with Luis Aparicio batting leadoff. The 1979 Pirates won the World Series with Omar Moreno batting leadoff. It really only makes a slim margin of difference.

[gyb13]

Originally posted by KCBOOMER
For every position in the batting order you drop you lose plate appearances. If a guy should be hitting third and you put him ninth you just gave up 108 PA's. On my club you don't give Rey Ordonez 108 PA's at the expense of Babe Ruth.
It seems like a lot, but it really isn't. Let's say in this case where you flip your 3 and 9 guys, there's a 100-point OBP difference between the two hitters. That means that, over the course of the season, you'll lose only about 11 baserunners.

[Xanadu Dragon]

:shockbig: a GYB sighting! Good to see you.

[gyb13]

....back from a lovely 4-day weekend hiking in Grand Canyon/Sedona. Back to office life, spreadsheets, coffee breaks, and of course...the (second ;) ) greatest sport on earth.

And to keep this thread going - for those of you who haven't, I suggest you go back and read through those 2 studies. They can probably prove these points better than we can...

[KCBOOMER]

I think it is more than eleven bases. I guarantee you that if you give Babe Ruth 108 plate appearances and you give Rey Ordonez 108 plate appearances you will see more difference than just eleven bases.

I grant you we are comparing the best to the least.

[gyb13]

Originally posted by KCBOOMER
I think it is more than eleven bases. I guarantee you that if you give Babe Ruth 108 plate appearances and you give Rey Ordonez 108 plate appearances you will see more difference than just eleven bases.

I grant you we are comparing the best to the least.
Baserunners, not bases. I didn't factor in SLG, just OBP.

[Fuzzy Bear]

Originally posted by sweaver
The 1959 White Sox won the pennant with Luis Aparicio batting leadoff. The 1979 Pirates won the World Series with Omar Moreno batting leadoff. It really only makes a slim margin of difference.

The '59 Go-Go Sox and the "We are Fam-i-lee" won in spite of these choices, in part because the other teams sought out "speedsters" who could run, but had low OBP due to "aggressiveness" (i. e., hacking away). The 1985 Cardinals had Tom Herr batting cleanup; is he your ideal choce for that role?

[Ytown Tribe fan]

My 3 cents worth is this: I'd put the guys in order of decreasing OBA, starting with the highest. Never mind SLG or BA or speed or anything else, just OBA.

For the Tribe, that'd mean batting Thome leadoff, with Lawton second, and so on. If it turned out that Thome was drawing fewer walks batting leadoff, we'd know we were on to something.

Outs are the name of the game. Not making them is the most important thing on offense. Since the leadoff batter gets more plate appearances than the number two guy, and so on down the line, over the course of a year, everyone would have more PA's if the batting order ran that way. More PAs = more baserunners = more runs.

[sweaver]

Originally posted by Fuzzy Bear
The '59 Go-Go Sox and the "We are Fam-i-lee" won in spite of these choices, in part because the other teams sought out "speedsters" who could run, but had low OBP due to "aggressiveness" (i. e., hacking away). The 1985 Cardinals had Tom Herr batting cleanup; is he your ideal choce for that role?

Exactly my point, Fuzz. The difference isn't that big, or those clubs would never have won.

Herr batted third for that team. On that particular team, he probably was the best man for the job.