The Actual Effect of Lineup Optimization
I inputted the following lineup into the Baseball Musings Lineup Simulator. It's the lineup we used for most of last year and my best guess as to what they will start with this year.
Weeks
Hardy
Braun
Fielder
Hart
Cameron
Hall
Kendall
Pitcher
I used the BCB projections for these batters except Hall, who got .315/.440.
The above lineup produces 4.791 runs per game.
When I created my lineup in the Lineup Construction post on the front page, I didn't use the lineup optimizer, but it turns out that mine is the top result in the lineup optimizer (only flipping Hart and Hall). You'll notice the differences between my lineup and morineko's in the comments of the lineup fanpost, I used a .363 OBA for Weeks instead of .342, and I am not using splits. Almost all of the top lineups use Rickie in the top spot . Every one of the best lineups uses Kendall in the ninth spot in the order and Braun in the fourth spot.
Now for the interesting part.
Over 162 games, the practical lineup projects to score 4.791 runs per game.
Over 162 games, the optimized lineup projects to score 5.027 runs per game.
First, that amount of runs would assume that all of the players play every game, which is impossible. The suggestion by the optimizer is that changing from a traditional lineup to an optimized one could yield an increase of almost 30 runs, which I find to be impractical. But I can definitely see the benefits, and I'm starting to think that there is no excuse for not optimizing the lineup-- especially if the result could be as much as an extra win over the course of the year.
Check out the best and worst lineups for yourself.
20 comments
|
0 recs |
Do you like this story?
Comments
Two interesting things I see
1. I think it’s funny how every single line-up that produces the least amount of runs has Kendall batting clean-up.
2. It seems strange to me that Cameron is in the three hole for BOTH the best and worst line-up.
Eric Gagne DL time: May 23-June 29 Brewers record in that span: 20-9
Good observation, that seemed confusing to me at first (about Cameron)
But I’m generally saying that your fifth best hitter should hit third. So it would make some sense that if you switched everyone around and put the best hitter in the worst spot (9th), second best in the second worst spot (8th), and on down, the middle guy— the fifth best hitter— would stay in the three hole.
Kendall hitting cleanup is pretty funny. Last year, a lot of the worst lineups had Hart leading off, and on Ned’s last stand, he led off Corey to shake things up.
Scored three times and detonated an indisputable in four visits to the batting box.
Does anyone still have the MLB Gameday image of Hart’s at bat from last game of the season in the eighth inning with bases loaded, one out, that resulted in three swings on three sliders for the second out?
Preserved for posterity

I am very glad I thought to screenshot this. My whole photobucket account consists of stuff like this.
Scored three times and detonated an indisputable in four visits to the batting box.
by Jordan M on Feb 15, 2009 6:27 PM CST up reply actions 1 recs
All you need is a duplicate copy with fail across it.
Or, better yet, you could do it subtlely (sp?) by taking out “Corey Hart strikes out swinging” and replacing it with “Cory Hart fails epically. Two out”
Eric Gagne DL time: May 23-June 29 Brewers record in that span: 20-9
Or a picture of Craig Counsell wearing a cape
it would take some doing to replace the bottom. Maybe I’ll do that sometime when I’m really bored, or if someone has some mad photoshop skills, they could work on it.
That gameday doesn’t do the at-bat justice, those three sliders were in the same exact spot.
Scored three times and detonated an indisputable in four visits to the batting box.
I thought that I had seen another one posted somewhere with the sliders in the same spot.
I would offer to do the photoshopping, unfortunately I am as skilled at photoshop as Corey Hart is at hitting the down and away slider.
Eric Gagne DL time: May 23-June 29 Brewers record in that span: 20-9
The question is-
Do you keep trying to do it again and again regardless?
by friendo on Feb 15, 2009 9:25 PM CST up reply actions 1 recs
Or say that your photoshop skills might be better if the fans supported you more?
I have an unreasonable dislike of Bill Hall.
by Ted Simmons Speed Camp on Feb 15, 2009 10:19 PM CST up reply actions
He's just in a slump right now
He’s not going to sit there and fix the problem, he’s going to just keep doing what doesn’t work.
Scored three times and detonated an indisputable in four visits to the batting box.
I'll eventually find it
hopefully, this will still be relevant by then.
Eric Gagne DL time: May 23-June 29 Brewers record in that span: 20-9
I don't even know how to photoshop
My skills are MS paint and the Snipping Tool.
Scored three times and detonated an indisputable in four visits to the batting box.
don't forget some badass paint-penmanship
Also, cheese.
by Jeff Sackmann on Feb 15, 2009 10:42 PM CST up reply actions
That'll be handy
for your second career as a mohel.
"I will agree that the attitude [at BCB] is ridiculous and they have done so much to instigate animosity and then block us from responding. Real mature!"
Someone really needs to tape that in Hart’s locker so it’s the first thing he sees for the rest of his career when he sits at his locker.
I don’t want to uber-hyperbolize or anything, but I’m struggling to think of a worse at bat, given the context and all. Everyone knew what the pitches would be, everyone knew what was going to happen… and it still happened and was even more horrific than imagined.
What about this one:
Game 2, first inning:
R Durham walked.
R Braun doubled to deep left, R Durham to third.
P Fielder intentionally walked.
J Hardy walked, R Durham scored, R Braun to third, P Fielder to second.
C Hart grounded into double play, pitcher to catcher to first, R Braun out at home.
That doesn’t show the best part — first pitch swinging after a run was just walked in.
Ah, I suppose that sounds logical.
Looks the the simulator just verifies your point than. Nice!
Eric Gagne DL time: May 23-June 29 Brewers record in that span: 20-9
One thing about lineup construction
When it’s said that lineup doesn’t matter, do they mean that literally? If we led off with the pitcher and then catcher, and tried to create the worst lineup possible, would there really be only a 1-game difference between that and our best lineup?
"I will agree that the attitude [at BCB] is ridiculous and they have done so much to instigate animosity and then block us from responding. Real mature!"
No
The 10-run difference usually refers to the difference between the optimized lineup and the likely real-life lineup. If the lineup analysis tool is to be believed, the difference between best and worst is a staggering 117 runs. (That’s surely extreme, but a more realistic number is still huge.)
When people like me say that “lineup doesn’t matter,” it’s more from an awareness that lineup optimization isn’t going to happen like the algorithm wants it to. Swapping Prince and Braun in the 3/4 spots, or batting JJ 2nd instead of 5th, or “protecting” Prince with one guy instead of another … within the realm of stuff a MLB manager is likely to do, no lineup tweak is going to make more than a 1 or 2 run difference, if that.
Also, cheese.
by Jeff Sackmann on Feb 15, 2009 7:43 PM CST up reply actions

by 


























