I made one of these last year over at talkingchop and I decided to do another one this year for the Royals. For those familiar with these kind of exercises you can skip the boring methodology part (which comes from The Book btw).

Before we continue, realize that the difference between an optimized lineup and what Yost runs out there is only 5-15 runs over the course of a season. Still, I enjoy this kind of stuff and maybe you will too.

Who's in the Lineups?

The opening day starters.

Where do the best hitters go?

1, 2, and 4. Ideally 1 has the highest OBP, 4 is the power hitter and 2 is a mix. The next two best slot in the 5 and 3 holes respectively. Yes the 5 spot is more important than the 3 spot (home runs have a higher run expectancy in the 3 spot though so a low OBP high slugger fits well here). After that sort the 6-9 batters in descending order.

What about Platoons?

It is imperative that the lefty are split up in the lineup to avoid the opposing manager leveraging a LOOGY late in the game. Lefties generally have larger platoon splits than righties. Because of these splits, we need to make different lineups for LHP and RHP.

What about GIDPs?

The Book says the most GIDP opportunities occur in 3 hole by far, followed by the 2 hole. So try to put the GIDP machines somewhere else.

Where does the speed guy go?

The Book says to put good baserunners in front of good contact guys and good basestealers in front of crappy or singles hitter who would have trouble getting them home from first.

Why did you switch the 8 and 9 guys?

The 9 hitter will hit in front of the top of the order (ie the best hitters on the team), so the penalty for less PAs is offset by this interaction. Note this really only works if the difference between the two worst hitters is severe (eg. a pitcher).

What stats did you use?

I used Zips projections from Fangraphs. The stat I used to determine batting ability is wOBA. Then I regressed each players platoon splits (1000 PAs for LHB and 2200 PAs for RHB). For more information on regression clicky here and here. Stolen bases Runs above average were obtained from BPro and are a 5/4/3 marcel type projection. In last year's post I used GIDPs, but the way I did it was terrible.  I was going to use the GIDP part of Chone WAR but Sean Smith got called up to the bigs. If you know of anyway to get GIDP runs above/below average let me know in the comments.

Lets look at the raw data.

Let's see what the lineup would look like against RHP:

I lead off with Billy because he has the best OBP and because it makes it easier to split up the lefties Kila and Gordon. I put Kila at 2nd because I think haveing 13 points in OBP at #2 here outweights having 5 points of ISO at #4. It could go either way. Between Aviles and Cabrera, Aviles is barely the better hitter overall, but his slight power advantage and Melky's minuscule base stealing prowess coerced me to put them in the 3 and 5 spots respectivly. Again they could be flipped if you think the 3 points of wOBA are more important. Melky and especially Chris Bletz's stealing abilities are leveraged well in front of the sub-.100 ISO sluggers and a neutered Jeff Francoeur. The whole switch the 8th and 9th hitters thing is why Escobar and Treanor are where they are. If you don't think 15 points of wOBA is a lot then switch them.

Now let's look at a lineup against LHP:

Against lefties it's basically Barry Bo- errr... Billy Butler and not much else. Aviles is league average(ish) so he slots in at the top. Jeff Francoeur sucks, but his bat is the third best vs. LHP on the team. I wish this weren't so. Ka'aihue has a lot more power than Melky so it makes a lot more sense to bat him third here. Besides Melky is better as a run producer. Tell 'em SabrNed! Everything else pretty much wrote itself.

I was going to do a lineup with Pena and Betemit, but I'm kind of tired. Maybe I'll put it up sometime tomorrow or Wednsday.

If anyone wants to see my spreadsheet where I calculated all this just let me know in the comments and I'll email it to you.

Lastly, I'd like to thank our own  Matt Klaassen aka Devil Fingers for his Lousy Lineup Series over at the now extinct Driveline Mechanics that inspired this post, and his fangraphs article that taught me how to regress platoon splits.