#2 Hitters
After I put up the graph on the relationship between fastballs seen by hitters and that hitters power rates, I started thinking about whether there might be some game theory issues that could be played around with. For instance, since we know that no power slap hitters see a lot more fastballs than guys with the ability to drive the ball, does this give us reason to think that traditional line-up roles are not optimized?
Most teams still use speed as a significant factor in choosing their leadoff hitter, and stolen base attempts from the #1 spot in the batting order dwarfs attempts from all the other line-up spots. There were 498 SB attempts from #1 hitters in the majors last year – no other batting order slot got more than 239.
So, since leadoff hitters are going to be stealing far more than any other line-up spot, we can infer that the #2 hitter will be at the plate most often when SB attempts occur. What’s the common wisdom on how pitchers defend against stolen bases? Throw fastballs. So which line-up spot should see the most fastballs? The #2 hitter.
Given that assumption, it would then follow that teams could setup a dilemma for pitchers by having a #2 hitter who pitchers do not want to throw fastballs too. If you had a high power #2 hitter, who pitchers only wanted to throw fastballs too 55% of the time, then you’d be forcing the pitcher to base his pitch selection on either the hitter or the runner. If you have a low power #2 hitter, then his desire to throw fastballs would align with both runner and hitter strategies, and there would be no conflict.
Despite this, #2 hitters had the second lowest ISO of any line-up spot in baseball last year, ahead of only #9 hitters. Indeed, the classic #2 hitter is a high contact hitter who is valued for his ability to give the manager confidence to call a hit and run or hit the ball to the right side if the leadoff hitter is able to steal second on his own. The current archtype #2 hitter is exactly the opposite of the kind of hitter that would force a pitcher to choose between pitching to the batter or the runner.
Now, I’m not suggesting that every team move their clean-up hitter to the #2 spot in the order. However, I do believe that players like Garret Anderson – moderate power free swingers who get a lot of breaking balls because they’ll chase them – could make a lot of sense in the #2 hole for a team with a prolific base-stealing leadoff guy. There’s value in making a pitcher choose between pitching to the hitter or the runner, and that value is being abandoned by teams who forgo any amount of power at the #2 spot in the order.
Print This Post
This idea makes a lot of sense. I’m surprised we haven’t seen a team that thinks outside the box try this yet.
Is there a benefit to placing a weaker hitter second so he gets more fastballs and has a better chance to suceed than he would elsewhere in the line-up like 7-8?
Does the advantage gained from moving a better power hitter to the 2-hole outweigh the production drop-off the other hitter would see from losing out on those straight pitches?
Similarly – why don’t these all-speed, no-hit guys bat in front of the pitchers in the NL.
Yes, if you are the Pittsburgh Pirates.
So does JJ Hardy fit this mold? He is set as the #2 guy for the Brewers behind a basestealing threat in Weeks.
i think d. wright fits this mold for the mets.
except…. Wright doesn’t hit second
who said that wright does hit second, josh? he’s saying that wright fits the mold…reading comprehension just isn’t that stressed in schools anymore i guess.
Well in Josh’s defense he did say “fits” this mold which would represent present tense, as opposed to “could fit”. I guess reading comprehensionand all that message board jazz…..
MPC-
Fits the mold- the article is about number 2 hitters. So for wright to fit the mold wouldn’t he need to be a two hitter? Or bat directly in front of a “prolific base-stealing leadoff guy”?
As for Matt H- yeah. Could the Mets bat Wright second? Sure. But that isn’t the point of the article. The point is to bat someone second whom would best take advantage of increased quantity of fastballs. One could say ANY Met would have better statistics based on seeing more fastballs. But the point here is-which Met would have the largest increase in the quality of their plate appearances?
I’m not sure Wright-arguably the most talented player on the team-would need that boost.
Anderson, indeed, saw time toward the end of the season in the #2 hole and, along with the arrival of Teixeira, seeing more fastballs was cited as a reason.
And, in fact, he hit .344 in 90 at-bats in the #2 hole.
Anderson isn’t a terrible breaking ball hitter, though he is prone to look more foolish than most on a lefty’s sweeping curve. He does hit them as well when he guesses right. In fact, I would contend that he’s more susceptible to the fastball up and in and a slider low and in than most. That’s how you get him out, especially when he guesses wrong. Quick adjustment has never been his strong point.
In Anderson’s case, it might be a little more complicated. It’s more about where the fastballs and sliders are located rather than particular pitch types. He feasted on guys that didn’t do their homework.
Now, I wonder how your idea will play out if Abreu sees time in the two hole as Scioscia has said he will do. Abreu sees A TON of fastballs already.
Could a move like that be a career extender?
Ideally you want your top hitters at the top of the lineup, based solely on the number of PA they will see over the course of the season. So I’m not sure giving Anderson the second-most PA opportunities is a benefit for the Braves. He’s really no more than a left-handed version of Jeff Francoeur minus the arm.
I like where this is coming from, but would you want to put a low-OBP guy like Anderson in the two-hole? You’re already creating problems by batting your Juan Pierre/Willy Taveras-types leadoff as it is, why compound it? Tooling around with this towards the bottom of the lineup might have some merit, though, and it works just fine if your speed guy and your power guy both have respectable OBPs.
If you bat in front of me you will hardly ever see a breaking ball. It does not matter if someone is on base.
Wrong.
the White Sox would greatly benefit by putting Alexei Ramirez in the 2 hole after speedster Jerry Owens who, if healthy, can steal 50+ bags. Alexei is the perfect aggressive, fastball hitter with plus power that would make this a great case-study for this theory.
Something tells me they would not greatly benefit from having two sub .330 obp guys atop the lineup.
What about Howie Kendrick?
Figgins-Kendrick would be a decent test case. Again, you’d like to see Kendrick get on base more, but he did it effectively in the minors, and doesn’t that usually translate unless you’ve got no power whatsoever?
Is it possible that the relationship between fastballs seen by hitters and power is influenced by players hitting second seeing more fastballs? Since #2 hitters generally see more fastballs because of speedy leadoff men getting on in front of them and prototypical #2 hitters are high contact low power hitters, wouldn’t this cause more slap hitters to see more fastballs? Not necessarily because they have no power but because more of them come to bat with fast guys on base. Or is this effect small relative to the overall relationship between power and fastballs seen?
I seem to remember Tony LaRussa advocating putting a bat with power at the number 2 hole. At the time I thought it was because he favored a hitter in the 9 hole instead of the pitcher, however maybe he has the same reasoning as Dave. And quite frankly it makes a lot of sense, as long as the player is not a complete obp dud.
This is definitely something La Russa likes to do. Chris Duncan, Ryan Ludwick, and Rick Ankiel have all hit there a good amount. Sometimes slap hitters like Aaron Miles or So Taguchi would fill that slot, but I know he prefers power hitters in that spot. Duncan was really good in that role in 2006 and early 2007.
Interesting article, Dave, and it sure makes a hell of a lot of sense.
Quick, unrelated question to anyone who can answer. Do the CHONE wOBA projections account for park factors and baserunning?
Stolen bases, yes. RBOE and park factors, no. Not yet, anyway.
Statcorner’s wOBA* has RBOE and adjusts for park.
Nice job getting Odalis Perez cut. Haha
Interesting concept, but I would like to see some empirical data to support #2 hitters getting more fastballs with the leadoff hitter on before making that assumption.
adam jones has a really good chance to “prove” that this is a good strategy this season.
Oliver’s wOBA does adjust for park, the projection is for a neutral park, which is why it may be lower than other projections for players from hitter’s parks.
I prefer a lefty in the 2-hole. Provides a decent ’screen’ to the catcher.
Actually, catchers do not throw out fewer runners when a left-hander is at the plate. That myth has been busted. See my THT piece on left-handed catchers for the data.
I still prefer a lefty for the provided ’screen’. It’s also never been proven catchers deserve the credit they are given. If the pitcher is quicker to the plate, the catcher will throw out more runners. Pretty simple stuff.
I had often thought that Hunter Pence should be No. 2 in the Astros’ batting order on similar reasoning (more FBs in the second slot). Pence is a tremendous FB hitter, but is prone to whiff on outside breaking pitches. And that seems to be consistent with how pitchers pitched him last year…(if I recall correctly) he had the lowest percent FB seen in the NL last year.
But his offensive stats out of the No. 2 hole are bad. Now, that may just be the fact that sample sizes aren’t big enough when you look at BA or OPS by batting order position (and that same reason, in general, will make it tough to test hypotheses like these). But it is also possible that we are wrong about FBs seen in the No. 2, particularly when you put a power hitter there.
Dan Uggla has also hit basically his career average out of the #2 spot. He’s a notorious fastball hitter and Ramirez is obviously a steal threat, so I find that sort of interesting.
Career average OPS, I mean.
I think Seattle may have tried this with Beltre when they were shifting him around to try and get his bat going. Seems like he hit pretty well in the two spot… or was it the eight spot.
Anyway, I’ve got an image of him whiffing on low and outside breaking balls burned into my retnas. And, it would be nice to see him hitting with Ichiro on base a bit more often.
Apparently Joe Torre said in his book that GM Brian Cashman suggested a line-up (I forgot which year) that had Giambi batting 2nd. I don’t think his reasoning was the same though. He had Abreu batting lead-off. I think he said he just wanted the highest OBP guys at the top. It would’ve been a coincidence, but hey, Cashman did want to try this.
The Yankees actually did do this in 2004. For like the last month of the season, Arod was hitting 2nd (or maybe it was 2005, but I think it was 04). It worked really well and I don’t know why they haven’t tried it since.
Anyway looking at this I think it might be a good idea to try again this year. Maybe:
1. Dammon (or Gardner)
2. Arod
3. Jeter
4. Teixeira
It won’t ever happen, but it does sort of make sense.
Wow, this is one of the rougher comment sections I’ve seen here. This is too bad, because the post is great. I think the misconception that hitters batting in front of the Pujols of the world get more fastballs has been addressed here fairly recently and pretty well busted. John, thanks for the THT article on left handed catchers – I had some trouble finding it, but it was worth the read.
http://www.hardballtimes.com/main/article/top-10-left-handed-catchers-for-2006/
(forgive the messy link)
Matt B., judging by your counterpoint, you didn’t read the John’s article. I think the study is pretty conclusive.
It looks like from this post that the ideal two hitter would be a guy who gets on base alot, hits fastballs disproportionately well, but doesn’t see alot of them. But, they should not have so much power that it is too your advantage to have more men on base for them. So, a team would still be better having someone like Huff or Uggla batting third or fourth. I’m looking at the list here of guys who don’t see alot of fastballs and fit this mold. Soto would fit this mold if he batted second. Maybe Soriano, would as well.
Bill, I didn’t read his article. I said it’s my preference. I don’t need stats and charts and arguments when it is something I just flat out prefer. Why don’t we ask a catcher/baseball player which they prefer as opposed to non athletes.
Why is it assumed that someone who analyzes data is automatically a “non-athlete”? I spent more time watching film and charting tendencies while I was an active player than I do now and I have to believe that there are plenty of former athletes in the ranks of sabernerdom.
You can “perfer” bad assumptions all you want, but to ignore a well researched opinion doesn’t help your credibility.
this sounds like a good idea for david ortiz
Interesting, but I’m still not sold.
Firstly, The Number 1 Hitter will be on base 1/3rd of the time. Which means that 2/3rds of the time, you’re going to have an inferior contact hitter up at bat. I’m assuming we’re taking for granted that this *new* number two hitter will post better numbers if he’s only getting fastballs, but worse numbers if he’s seeing a mixture of pitchers with no one on base.
Secondly, I’d like to see the difference in steal rates between a curveball and fastball. We’re dealing with a difference of half a second maximum in time to the plate.
Finally, the slap hitter that we’re replacing has a better opportunity to move the runner over / score him.
We’re dealing with our fastest player at the number one spot, so realistically if we can get him to second then a single will score him. Even if our *new* number two hitter hits it hard, our number one hitter will only get to third — and a single still scores him. We’d have to assume that our old number two hitter strikes out less, and moves the runner to second more often. I’d also assume that our old number two hitter would beat out more double plays and induce more fielders choices.
I think the best example would be…
Which Line – Up?
Granderson / Polanco / Ordonez / Cabrera / Guillen
or
Granderson / Guillen / Ordonez / Cabrera / Polanco
Maybe this isn’t the best example this year, but it sorta proves the point, as polanco is an extreme old-school number 2 guy.
Good, interesting article. It would be interesting to try and quantitatively estimate what kind of steals advantage would be seen for power versus contact #2 batter. My guess would be that the advantage over an entire season isn’t really that great.
I would also guess, and kris above alludes to this, that whatever advantage there might be to lower fastball rates – the more significant thing is the ability to hit and run and to bunt where the advantage goes to the contact hitter usually. I’m just guessing but it just seems like common sense *gasp* that this is why we don’t see the above scenario too often. Maybe someone with better sabr skills than me could do a real statistical analysis.