For an AL lineup especially I think it makes tons of sense to bat your top guy #2. Without the "automatic" out in the #9 hole you can stick a "second" or even your standard leadoff hitter there. And it can make plenty of sense to put a Fielder type in the #1 spot. Every lineup, including every traditional one, is going to have those two bat back-to-back so speed really isn't a question. Put speed guys in the 8/9 holes and you've got your traditional lineup in every inning except for the first and you're much more likely to start off the 1st with one of your two guys getting on base.

In the NL it does make intuitive sense to keep your best hitters away from the automatic out so I can understand why the traditional lineup holds on in the NL even if it's sub-optimal -- it just seems nuts to have your no-hit SS and your pitcher hitting in front of Fielder and Cabrera.

Now, with so many projection systems available online, why would anybody use the Bill James ones?

Realistically (meaning Cabrera and Fielder will hit 3/4 in some combo), I think the best thing you can do is have Hunter hit 2 and make sure Miguel is 4th.

Walt, I agree there are better projection systems, but this was mostly just a fun article for Tigers fans, so I decided to use the one with the most optimistic projections. I didn't think it would change the line-ups much if I used a different system.

Your point about AL versus NL line-ups does make a lot of sense. However, the Tigers scored significantly fewer runs than they should have last year given their batting events and I think speed was part of it. So, I was a little more conscious of speed in constructing a line-up than I normally would be. Still, part of me would love to see what would happen with Fielder and Cabrera batting one-two.

Their league leading 156 GIDP didn't help. Crushing lack of production in the 5 and 6 spots is what did them in more than lack of speed or problems at the top of the lineup. Putting Victor in place of Delmon will make a big difference.

The DPs were absolutely a problem, one which will probably still be around this year. Another reason why they didn't score as many runs as they should have (wRC >> Runs) was too many solo homers. That was probably just random bad luck though and should take care of itself.

Oh, I don't really care what projection system you use.

too many solo homers

Gotta be the lack of speed! :-)

I think speed was part of it

Maybe. But my point is you can put the same speed in front of Fielder and Cabrera if you want by batting that speed 8th and 9th. They just wouldn't be in front of them in the 1st inning. And those two will always be back-to-back in the lineup so Cabrera will always have the slow Fielder batting in front of him (or vice versa) so it's not clear what you lose by putting them 1-2. I agree we'll never see it.

EDIT: by the way, Cabrera, Fielder and VMart (coming back from ACL!) is about 17 Molinas of slow. It's like Cash/Horton/Brown.

too many solo homers

Gotta be the lack of speed! :-)

I hadn't really thought of this before, but does a lack of team speed generally lead to fewer solo HR (as a percentage of overall HR)? Slower runners are less likely to score on all non-HR events, thus they are more likely to still be on the bases when home runs occur? Or does something else cancel this out? (For example, having slower runners leads to more forceouts and double plays, reducing baserunners.)

EDIT: by the way, Cabrera, Fielder and VMart (coming back from ACL!) is about 17 Molinas of slow. It's like Cash/Horton/Brown.

Cabrera takes a while to get going, but his peak speed might be within a standard deviation of "average." From my observation, he's nowhere near as slow as Fielder. On that basis, it makes sense to keep Cabrera in front of Fielder. Cabrera is more likely to score on a Fielder single/double than Fielder is to score on a Cabrera single/double.

...Table 2 shows that four of the five best line-ups have Prince Fielder leading off! In fact, eight of the top ten have Fielder at number one and all of the top thirty have either Fielder or Alex Avila. Remember though that this only looks at hitting and does not consider speed of which Fielder and Avila have none.

Does anyone still use lineup simulators that don't account for speed? Does the author know that his Apple IIe can handle the more advanced software people are writing these days?

I thought the consensus was you use your two best hitters w/power at #2 and #4?

I'd think Cabrera #2 and Fielder #4, with Jackson #1, and Hunter and VMart #3 and #5 (in whatever order) makes the most sense.

Guys, I understand that hitting is the primary factor in line-up optimization and that speed is secondary. I can't prove it, but I do believe that if you have some good base runners on the team who can also get on base, putting them in front of your best hitters can add a few runs to an optimal line-up over the course of the season. I could very well be wrong.

Jack, if there is an easily accessible line-up optimizer accounting for speed that everyone is using as you suggest, please give me a link. I have Diamond Baseball that I use for my own purposes on occasion. However, for a quick post it was easier just to plug the numbers into the Baseball Musings. It's also better for readers of my blog if there is a site where they can plug in their own line-ups if they choose.

I do believe that if you have some good base runners on the team who can also get on base, putting them in front of your best hitters can add a few runs to an optimal line-up over the course of the season.

I don't disagree but I think you're still looking past my point, so just as an example:

1. Jackson
2. Berry
3. Fielder
4. Cabrera ...

puts the speed of Jackson and Berry in front of Cabrera and Fielder.

1. Fielder
2. Cabrera ...

8. Jackson
9. Berry

puts the speed of Jackson and Berry in front of Cabrera and Fielder except in the first inning.

Also the supposed problem with

1. Fielder
2. Cabrera

is that the deathly slow Fielder is batting right in front of Cabrera. But that's also true of ...

3. Fielder
4. Cabrera

So you have to show that having speed in front of Fielder and Cabrera IN THE FIRST INNING ALONE is better than the improvement in the chance that your first two hitters of the game reach base and the extra PAs you'd get from Fielder and Cabrera throughout the season.

Now, in this SPECIFIC case, Jackson is one of the Tigers' best hitters, possibly their third best hitter. For this particular set of batters, Snapper has it exactly right -- it's an ideal group of players for The Book approach, Jackson/Cabrera/VMart/Fielder/Hunter. (flipping Hunter and VMart to your preference). But if you replace Jackson by, say, Cameron Maybin, then Maybin providing speed out of the #8 or #9 spot makes all sorts of sense to me.

One of the key reasons that lineup order doesn't matter, probably especially in the AL, is because the lineup order doesn't matter after the 1st or 2nd inning. After the first inning, your "leadoff" opportunities will arise pretty much by chance, so the #9 guy will lead off an inning as often as the #1 guy.

Just as Khan thought two-dimensionally*, far too often we think linearly when constructing the lineup. It's a batting cycle not a batting line. The desire to put speed in front of sluggers can influence how you order players in that cycle but it shouldn't solely determine where you choose to start the cycle.

*Seriously, how could you possibly captain a spaceship -- a faster than light spaceship -- while thinking only two-dimensionally?

Done with my eight-mile run and thinking more calmly and clearly...Good post Walt. Thanks for taking the time to lay it all out. I don't disagree with any of that except that I hope Berry isn't involved on a regular basis (I know you were just using him as an example to make your point.) Right now, it looks as if Dirks will get most of the playing time in left as he should. I also like Snapper's line-up. Of course, none of this matters as Leyland revealed his line-up today:

Jackson
Hunter
Cabrera
Fielder
Martinez
Dirks
Peralta
Avila
Infante

I'd forgotten about Infante, he's a more realistic guy to plug in than Berry.

Jackson (R)
Hunter (R)
Cabrera (R)
Fielder (L)
Martinez (S - very small platoon split over his career, but slightly better batting R)
Dirks (L)
Peralta (R)
Avila (L)
Infante (R)

That may not be the "ideal" lineup, but it looks pretty good to me!

Switching righty Hunter and lefty Dirks might make sense if Dirks hits like he did last year. That would give Cabrera a better "cushion" against relief matchups. (Or, more likely, would ensure that Dirks sees more right-handed pitchers.)