Let's ignore platoon splits and trade possibilities for a second. With the players they have in place right now on their roster, what is the most statistically effective lineup John Russell could put together?
The premise of this comes from numbers out of "The Book" by Tom Tango. In it, he goes over what the most likely run probabilities would be with players of various wOBA in different spots in the order.
Again, this is all statistical and based on an average team, but here's what his computers came up with. I will use a BA/OBP/SLG triple slash with wOBA to follow: I am arbitrarily excluding the ninth spot in the order.
Optimal Batting Order for League Average Team:
1) .273/.439/.439, .407 wOBA
2) .298/.390./.529, .405 wOBA
3) .275/.326/.466, .352 wOBA
4). 298/.362./,578, .407 wOBA
5) .260/364./.396, .351 wOBA
6. .250/.316/.402, .326 wOBA
7. .242/.307./.389, .316 wOBA
8. .234/.298./.377, .307 wOBA
Now, you're problem wondering how he got to these numbers. I was, too, and he explains it pretty well. For instance, home runs by hitters in the cleanup spot produce a 1.47 run expectancy, whereas the same home run by the eighth hitter only carries with it a 1.29 run expectancy. Things like that.
Singles are worth the most if they're hit by the first and second hitters, both carrying singles at a 0.515 run expectancy. Walks are worth the most out of the leadoff spot, and doubles are most valuable from cleanup..
An out carries with it the most negative value at leadoff at -0.328 runs expected, and an out hurts the least in the eight spot at -0.286. Strikeouts are worst in the cleanup spot at -0.332, the second worst spot for strikeouts being leadoff.
Tango and his colleagues came to these conclusions over thousands of logged games from decades worth of game collection. He plugged the numbers into a computer model, and he was able to come up with precise run values for game events as they happen to players in different spots in the batting order.
There's a full chart available in his book, which I really recommend to anyone. The chart shows run values for every single spot in the order in every base/out state for every event.
Okay, how do we apply this to the Pirates? Firstly, it all depends on if you believe in playing the percentages or not. If you want to be conventional and just put your most power at cleanup and the worst hitter batting eighth, that's fine, but let's say we want to put the best possible run expectancy on the field.
The book stresses that your leadoff hitter is only likely to lead off once in the entire game, and the order needs to be treated as a continuous loop.
Looking back to the optimal lineup that Tango came up with, we can make a few conclusions. I am going to take some results from the month of July, for the simple reason that so much of the Pirates' lineup is so raw and anything earlier is even more skewed by lack of experience.
- Your best hitters should be batting, in no particular order, in the first, second, and fourth spots. This comes from wOBA. The Pirates' best wOBA hitters this year have been McCutchen, Jones, and Milledge, . This is sort of difficult to apply to the Pirates since so many players are still developing (Tabata, Alvarez, and Walker, to be specific), but as of right now, the Pirates would be best served with McCutchen leading off, Milledge batting second, and Jones at cleanup. The catch is that Pedro Alvarez is very likely to be more efficient than Jones eventually, but for the right now, Jones has had more sustained success. Again, this shouldn't necessarily happen, just because it's not smart to make results-oriented lineup decisions involving rookie players.
- You worst hitter should bat eighth. Come on down, Ronny Cedeno.
- Your fourth best hitter should bat third. Especially with how he's been hitting lately, welcome to the three-hole, Pedro Alvarez.
- Your fifth best hitter should bat fifth, preferably with decent power. This is Doumit.
- Your sixth best hitter should bat sixth. I'd call this spot for Jose Tabata. After a little more than a month in the Majors, Tabata has turned it on and has a .345 wOBA for the last two weeks and deserves this spot. I like him
- That leaves Neil Walker for the seventh spot, where I can't help but think his bat would play very, very well.
Again, this doesn't factor in the future, just right now, as in for the next few weeks of this season. I'm not saying that this should be next year's opening day lineup. In fact, I'm almost certain it won't be.
Anyway, here's the Pirates' optimal lineup, strictly derived from run expectancy percentages. I'm using recent months for some players because they show them playing like additional playing experience suggests they could.
1. Andrew McCutchen, .358 wOBA
2. Lastings Milledge, .394 wOBA in June, .410 wOBA in July (!)
3. Pedro Alvarez, .399 wOBA in July (!.)
4. Garrett Jones, .337 wOBA
5. Ryan Doumit, .329 wOBA
6. Jose Tabata, .345 wOBA last two weeks
7. Neil Walker, .328 wOBA
8. Ronny Cedeno, .287 wOBA (.520 in July, but won't stay even close to that)
That lineup actually doesn't look too farfetched to me, and the percentages say that it's optimal for scoring runs. The glaring thing is that it takes Tabata's speed out of the top of the lineup, but once again, it needs to be considered that the order is a continuous loop.
Long-term, a few things are almost certain to change. Optimally, Alvarez will be an ideal cleanup hitter and Tabata will stay in the top two. Right now, though, this is what should theoretically score the most runs.This is all numbers, not subjective opinion. Subjectively, I like Doumit at seven and Tabata at two, but the numbers disagree.
What do you think?
P.S. Sorry for the longness. I appreciate it if you've actually read this, as always.