Optimal lineup
 

http://thesoberangels.blogspot.com/2....html?spref=tw

Best Simulator Lineup vs RHP
1. Shin-Soo Choo (L)
2. Joey Votto (L)
3. Brandon Phillips
4. Jay Bruce (L)
5. Zack Cozart
6. Todd Frazier
7. Ryan Ludwick
8. Pitchers Spot
9. Ryan Hanigan
Wins/162 Games Better Than Default Lineup: 1.14

Best Simulator Lineup vs LHP
1. Shin-Soo Choo (L)
2. Jay Bruce (L)
3. Brandon Phillips
4. Joey Votto (L)
5. Zack Cozart
6. Todd Frazier
7. Ryan Ludwick
8. Pitchers Spot
9. Ryan Hanigan
Wins/162 Games Better Than Default Lineup: 0.75

Main observations:
- Doesn't like seeing Votto bat with 2 outs and the bases empty.
- Likes Hanigan batting 9th to take greater advantage of his OBP and recognizing his limited potential to drive in runs.
- Cozart, Frazier and Ludwick are similar types of hitters, though I don't see why Cozart should be batting ahead of the other two

Re: Optimal lineup
 

I like it a lot. Of course I am the guy that has been advocating for putting Joey Votto in the 2-hole for a couple years now.

It is interesting that the only difference between the LHP and RHP lineups is that Votto and Bruce are flipped.

The simulator seems to like putting the pitcher in the 8th slot for all National League teams.

I like that the simulator recognizes that Phillips (and not Ludwick) is the ideal hitter to put between Votto and Bruce. Phillips crushes lefties, while Ludwick is better vs righties (despite his small sample size to the contrary in 2012). So Phillips is the guy to put in the 3rd slot to take advantage of the extra LOOGYs in the late innings.

It is traditional in baseball to show respect to your best hitter by batting him in the 3rd spot in the batting order. But pretty much every one of these simulators has shown over the years that the 3-hole is LESS important than the 1, 2 and 4 holes in terms of run production.

This sim used Bill James' projected statistics -- not the players' real 2012 stats, which likely accounts for why Cozart sits above Ludwick and Frazier. I would like to see the sim redone using last year's actual stats to see if it turns out different, especially when compared to Dusty's usual lineup with Stubbs and Cozart at the top.

James has Cozart projected for a .308 OBP, which is much better than the .288 OBP Cozart put up last year.
James also has Ludwick projected for a much inferior season in 2013 than he did in 2012 -- a .332 wOBA vs the .373 wOBA from 2012.

Re: Optimal lineup
 

How much of a difference are we talking in terms of runs here?

Re: Optimal lineup
 

You can assume that 1 win is roughly 10 runs.

Re: Optimal lineup
 

I agree with the simulator. To me, the basic rule of thumb in making a lineup would be to maximize what you've got. That means putting the best hitters (oversimplifying) in the immediate company of the other best hitters. Why, then, would you want your worst hitter--the pitcher--contiguous with the run of best hitters? The 9 guy should be an OBP type. And you really don't lose much by moving the pitcher up to 8. The guy in front of him should be the second-worst hitter.

Also agree with Votto in the 2 hole. That way, they can't be pitching around him as freely. And when they do, his walks are put to greater advantage.

Re: Optimal lineup
 

In this example the range is an extra 7-12 runs by using the optimal lineup instead of the lineup that is listed on MLBdepthcharts.com. This lineup is:

1. Shin-Soo Choo (L)
2. Brandon Phillips
3. Joey Votto (L)
4. Ryan Ludwick
5. Jay Bruce (L)
6. Todd Frazier
7. Zack Cozart
8. Ryan Hanigan
9. Pitchers Spot


Other studies have shown that using the best possible lineup instead of the worst possible lineup would net an extra 40 runs or so for an average MLB team. That would be worth an extra 4 wins on average.

The difference of an extra 7-12 runs (about one win) in this particular example is smaller than the 40 runs because the optimal lineup is being compared to the Reds' likely Opening Day batting order, which is much better than the worst possible lineup (which might have the pitcher batting 3rd for example).

If this study had been executed using the Reds' actual 2012 stats and their usual batting order with Stubbs 1st and Cozart 2nd etc it would have been interesting to see how many more runs the optimal lineup would have scored. I imagine last year's standard batting order was closer to the worst possible lineup than the optimal lineup.

Re: Optimal lineup
 

This looks a lot like the batting order that *BaseClogger* guy has been advocating around here!

Re: Optimal lineup
 

who?

:D

Re: Optimal lineup
 

Is the simulator using just 2012 numbers or career? If 2012 is it putting Votto 2nd due to the diminished power post injury?

Re: Optimal lineup
 

Not sure that I can get behind any lineup with Cozart hitting 5th.

Re: Optimal lineup
 

Because the Reds never had adequate lineup protection for Joey, I've advocated batting him 2nd for a long time. I hate seeing him pitched around. Opposing teams would have to pitch to him or put him on base and take their chances. Joey would probably bat close to .400 if he batted 2nd. LOL.

Hopefully acquiring Choo takes care of it though. If Choo and BP can get on base in front of Joey, no way they walk the bases full.

Re: Optimal lineup
 

The simulator is using projected 2013 numbers provided by Bill James.

Re: Optimal lineup
 

Does anyone know much about this simulator? Is this MATLAB code? I would be curious to see the algorithms he used to simulate PAs.

Re: Optimal lineup
 

Optimus Prime Lineup ;)

Joey Votto
Joey Votto
Joey Votto
Joey Votto
Joey Votto
Joey Votto
Joey Votto
Joey Votto
Joey Votto

Re: Optimal lineup
 

If cloning humans were legal Joey Votto's DNA would be worth more than his contract with the Reds.