Devising the “Perfect” Sox Batting Order…Valentine has a better idea
By Earl Nash
For citizens of Red Sox Nation few items of pre-season speculation are more entertaining, than divining the perfect batting order for the Red Sox. No matter that Bobby Valentine will use at least 100 different combinations in 2012 and may use our perfect lineup in only 31 games, because they may be the games that are crucial to winning the AL bEAST flag.
Eggheads by the dozens in academia have been searching and re-searching their heuristics and statistics to discover exactly [plus or minus 3 or 4] how many wins the Sox could add in 2012, using the perfect lineup; so, drum roll please:
"“Effective optimization can increase a team’s win total by up to 3 wins per season.”"
[A Robust Heuristic for Batting Order Optimization Under Uncertainty, Joel S. Sokol, Journal of Heuristics, Vol. 9, No. 4, 353-370.]
OK, it’s written in an academic journal; there we have it: 3 more games won.
"“4 games per 162 game season.”"
[A Markov Chain Approach to Baseball, Bruce Bukiet, Elliotte Rusty Harold , José Luis Palacios, Operations Research, January/February 1997 vol. 45 no. 1 14-23]
OK, 3, maybe 4, more games won out of 162 [plus or minus 3 or 4]…
Can some expert decide; is it 3 or 4?
Oh, here we go, this guy used a sample size of 200,000 games…
“…simulation of over 200,000 baseball games, using a programmed embodiment of the main features of the Sports Illustrated baseball game, shows that…
"batting order exerts only a small influence on the outcomes of baseball games.”"
[An Analysis of Baseball Batting Order by Monte Carlo Simulation, R. Allan Freeze,
Operations Research
, Vol. 22, No. 4 (Jul. – Aug., 1974), pp. 728-735]
Uh, “small influence” is less specific that 3 or 4…
Maybe we need to drill down in whole chapters in books:
From a review of The Book: “Another way to look at things is to order the batting slots by the leveraged value of the out. In plain English (sort of), we want to know how costly making an out is by each lineup position, based on the base-out situations they most often find themselves in, and then weighted by how often each lineup spot comes to the plate.
Ah, now we’re getting there [maybe]…
“Here’s how the lineup spots rank in the importance of avoiding outs:
#1, #4, #2, #5, #3, #6, #7, #8, #9
So, you want your best three hitters to hit in the #1, #4, and #2 spots. Distribute them so OBP is higher in the order and SLG is lower. Then place your fourth and fifth best hitters, with the #5 spot usually seeing the better hitter, unless he’s a high-homerun guy. Then place your four remaining hitters in decreasing order of overall hitting ability, with base stealers ahead of singles hitters. Finally, stop talking like the lineup is a make-or-break decision.”
The old-school book says to put your best high-average hitter in the third slot. The lead-off hitter should already be in scoring position and a hit drives him in. Wham, bam, thank you ma’am.
But, The Book says “the #3 hitter comes to the plate with, on average, fewer runners on base than the #4 or #5 hitters. So why focus on putting a guy who can knock in runs in the #3 spot, when the two spots after him can benefit from it more? Surprisingly, because he comes to bat so often with two outs and no runners on base, the #3 hitter isn’t nearly as important as we think. This is a spot to fill after more important spots are taken care of.”
And the lower third of the order, 6-9?
The Book says: “Stolen bases are most valuable ahead of high-contact singles hitters, who are more likely to hit at the bottom of the lineup. So a base-stealing threat who doesn’t deserve a spot higher in the lineup is optimized in the #6 hole, followed by the singles hitters.”
[Book review of The Book, by Tom Tango, Mitch Lichtman, and Andy Dolphin. “Optimizing Your Lineup by The Book,” Sky Kalkman http://www.beyondtheboxscore.com/2009/3/17/795946/optimizing-your-lineup-by]
We tried to apply the theory in The Book to the Sox lineup. We took the top three hitters (Ortiz, Gonzalez, Ellsbury) and put them in the #1, #2, and #4 spots, with the player least reliant on homeruns first (Ellsbury), the player with the highest relative OBP second (Gonzalez) and the player with the best power fourth (Ortiz). That gives us Ellsbury — Gonzalez — #3? – Ortiz, so far.
Then, we slottted the next two best hitters (Pedroia and Youkalis) into the #5 and #3 spots.
Since the authors say “the outs for the #5 hitter are much more costly than those in the #3 slot,” #5 hitter (Pedroia) should be better than the hitter in the #3 slot (Youkalis).
The authors reason that “the #3 hitter gets a lot more plate appearances with two outs than the #5 batter. So, he has less chance to do damage, unless that damage is done with the homerun.”
This formula gives us:
- Gonzalez
- Ellsbury
- Pedroia
- Ortiz
- Youkalis
[“Optimizing Lineups: Twins, Mets, and Yankees,” Sky Kalkman on Mar 19,2009, http://www.beyondtheboxscore.com/2009/3/19/801318/optimizing-lineups-twins-m]
Another guy concluded that “most of the difference in expected runs between high and low scoring lineups using the same players occurs in the first inning.” Better get to your seat promptly for the first pitch.
And, that “Tests using the Markov model showed its makes virtually no difference bats fourth five and six slots in either order.”
[“FINDING BETTER BATTING ORDERS, Mark D. Pankin, http://www.pankin.com/markov/btn1191.htm]
Another dedicated researcher ran 810,000 simulated games [hopefully by computer] and organized lineups by On Base Average from best to worst, worst to best and also [just for laughs] created a lineup randomly. Random came in a close second [by just a tad over 1% and the best down to the worst came in third. Who’d-a guessed? It was interesting that they finished in a virtual tie; rounded off: 52-51-50, by my rudimentary calculations.
OBA best down to worst 49.89%
OBA worst down to best 52.00%
Random 50.879%
[“The Effect of Batting Order in Runs Scored,”Roger Moore, Cal. Tech. Inst., http://alumnus.caltech.edu/~raj/writing/BattingOrder.html]
In Speaking of On Base percentages and Carl Crawford, John Tomase, Boston Herald, just posted a chart showing [in order of success] where Carl did best at OBP:
Here are his best places in the order, arranged by OPS:
7th .874
2nd .806
3rd .784
6th .760
1st .734
9th .662
8th .624
4th .500
5th .000
Although it shows, using OPS and the driving factor, that Francona had the stats to back up his dropping Crawford to the 7 hole, Carl’s speed might over-rule to result and make the 2 hole his most productive home. On paper, using OPS, Carl was apparently correct when he said that he “sucked” at hitting leadoff.
On the grass, Crawford has not batted leadoff since the first week of the 2007 season. In all, he served as leadoff hitter in 367 games, and batted .288/.323/.421/.744. In his remaining years with the Rays to 2011, he primarily hit out of the 2-hole. Then Tito bounced him all over the lineup in 2011: 1,2,4,6,7, until Carl was dizzy with changes in teams, cities and lineup slots.
As always, Valentine has a different, baseball-rational take:
“It creates a bad mentality, to think you have one lineup and that lineup is the one that wins. It’s a Little League mentality that should not exist at the highest level of baseball. To say nothing of the fact that guys often need to play to be contributors and to feel part of a team.”
“I think it’s evolved,” Valentine said. “Just a combination of things. One, I started getting more information where I would realize some lineups probably worked better against some pitchers, some lineups probably worked better against different bullpens, some lineups cannot be together all year.”
Valentine adds to the case against quest for the “perfect lineup” and will apply solid baseball reasoning to devising lineups that are perfectly targeted for that day’s opponent’s strengths and weaknesses; this will likely result and at least 100 different permutations and maybe as many as 150 over the 162-game season.
If a lineup is used for less than 70% of games played, it can hardly be called the “perfect lineup” and Valentine makes the case that “one size cannot fit all.” The brainiacs in academia seem to agree that employing the perfect lineup [adjusted per Valentine’s theory] will probably result in winning 3-4 more games per season.
Or, some studies conclude that, if you put the players names in a hat and drew them out randomly to create your lineup for that day, your season advantage would also fall into that 3-4 more win range.
Now, we’re thinking, maybe:
- Ellsbury
- Gonzalez
- Pedroia
- Ortiz
- Youkalis
- Aviles
- Ross
- Saltalmacchia
- Crawford
No, wait! Maybe…
Hey here’s a comment on a blog from…
The All Knowing says:
1. Fastest player best average
2. Fast and contact hitter/bunter
3. Power hitter/average
4. Best hitter/Power
5. Power/protecter hitter normally not as good as 3 & 4 but can still hit bombs
6. Average hitter
7. Worst hitter
8. Not great but good because will not see a lot of pitches with the pitcher on deck
9. If no Dh a second lead off guy speed and average if pitcher him
Sounds good, but
Joe
says:
No, It’s Not. The Coach puts Players in This Order Generally.
Good-Good-Good-Bad-Bad-Good-Bad-Good-Good-Bad-Good-Bad
What i Mean to Say Is That order in Batting Lineups Dont Matter, Because Everyone Has the Same Chance Batting. My Above Example Shows That The Coach Needs Good Players Throughout, And Not Condensed into one Area.
Other experts say that Joe is all wrong with his scattering method and that clumping your best hitters together in one group and your worst hitters in another group will result in more runs being scored.
Maybe Valentine is right, again, forget the “perfect” lineup concept; run the numbers on the other team and set your batting order accordingly.
But, Ellsbury in the 3-hole with behind “Peedy” with Crawford as lead-off…and…
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
For all the latest news and analysis from BoSox Injection, follow us on Twitter, Facebook, or with our RSS feed.
OOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO
For all the latest news and analysis from BoSox Injection, follow us on Twitter, Facebook, or with our RSS feed.