The #3 hitter and other batting order riddles

[moonslav59]

This says it better than I could have. Five times out of nine, more than half the time, the #3 guy comes up with someone on in the first inning. That's why I want a hitter, an OPS guy there.

 

I want a high OBP guy there and my best or second best (with power) up 4th.

[cp176]

Exactly. Creating runs is not just about RBI's. It's about runs scored and increasing the odds of runs being scored by doing something.

 

There will always be exceptions to any rule, especially when we are talking about very marginal differentials between just about all batting slots near each other in the line-up.

 

It's also about how many more (or less) runs might have occurred with the alternate choice to JD in the 3rd or 4th slot and vice versa.

 

Maybe JD created 22 more runs in the 3rd slot than the 4th, but when he batted 4th, the guy up 3rd created 23 more.

 

 

My argument about having your best run producing bat in the number 3 spot is exactly that. A run producing bat! This argument has always been about producing runs via the bat. Everyone who knows anything about the game at all understands that there are other ways to produce runs. If one single second i thought that the argument has been about anything other than the bat, maybe i would feel differently. I don't. My best run producing bat, is in the 3 spot. It is becoming much like a political debate. Some don't like the obvious so they change the rules as they go in an effort to prove a pointless point.

Eliminate that first inning and since I have seen no data to dispute the claim - that guy hitting third takes on some serious level of importance.

[S5Dewey]

I want a high OBP guy there and my best or second best (with power) up 4th.

 

This may come down to a discussion about what is a "good" hitter and what they bring to the lineup.

 

OPB and OPS have something in common. The common element is that they both are dependent on how often a player reaches base. It's just that the player with the higher OPS has more power. Since even a successful hitter only reaches base about one time out of three I want my #3 hitter to have that one time out of three to have that power. I'll take that over my #3 hitter maybe getting a walk or getting a single and leaving the big scoring opportunity to the #4 hitter... who also is unsuccessful two times out of three.

Nobody is going to be successful every time, but optimize your chances to score.

 

Edit... The more I think about this the more amazed I am that the stat geeks have decided that the better hitter should be the #2 guy and that the #3 guy isn't as important as even the #4 or #5 guy.

 

Equally as amazing is that people bought into it!

Edited June 7, 2019 by S5Dewey

[S5Dewey]

This whole discussion seems to center around what happens in the first time through the lineup and yet that only happens once in the game. Shouldn't we be focusing on what happens when the #1 hitter comes up after the #9 hitter, which probably will happen two, three, four or more times a game?

[Bellhorn04]

This whole discussion seems to center around what happens in the first time through the lineup and yet that only happens once in the game. Shouldn't we be focusing on what happens when the #1 hitter comes up after the #9 hitter, which probably will happen two, three, four or more times a game?

 

It all matters.

[Bellhorn04]

Personally, I think while it's fun to talk in theoretical terms, it's even more interesting to look at what the real managers are doing.

 

I do tend to think Cora is a manager trying to blend new school and old school stuff.

 

And I find it interesting he puts JD in the 3 slot and the 4 slot about equally.

[notin]

This whole discussion seems to center around what happens in the first time through the lineup and yet that only happens once in the game. Shouldn't we be focusing on what happens when the #1 hitter comes up after the #9 hitter, which probably will happen two, three, four or more times a game?

 

 

It all matters. But the only time you can control it is the first inning.

 

And while the 7-8-9 hitters are supposed to be the ones on base in front of the leadoff hitter, the team is going to have a worst hitter. Where should he bat? To me, the ideal spot is last, where he gets the fewest plate appearances...

[Slasher9]

The more I think about this the more amazed I am that the stat geeks have decided that the better hitter should be the #2 guy and that the #3 guy isn't as important as even the #4 or #5 guy.

 

Equally as amazing is that people bought into it![/i]

 

agreed.

[Slasher9]

worst hitter. Where should he bat? To me, the ideal spot is last, where he gets the fewest plate appearances...

agreed

[moonslav59]

My argument about having your best run producing bat in the number 3 spot is exactly that. A run producing bat! This argument has always been about producing runs via the bat. Everyone who knows anything about the game at all understands that there are other ways to produce runs. If one single second i thought that the argument has been about anything other than the bat, maybe i would feel differently. I don't. My best run producing bat, is in the 3 spot. It is becoming much like a political debate. Some don't like the obvious so they change the rules as they go in an effort to prove a pointless point.

Eliminate that first inning and since I have seen no data to dispute the claim - that guy hitting third takes on some serious level of importance.

 

I think the differentials are so small between the 3rd, 4th and 5th slots that saying they are all very important run-producing slots is always going to be true.

 

I was a firm believer in the 3 slot for your best all-around hitter for decades, but I trust the studies. It's not something I will go to my grave defending, especially because the differentials are very small.

 

I think we've beaten this subject to death, but I will add two points:

 

1) Taking away the first inning seems to favor the idea that the 4th hitter is more important than the 3rd hitter, who has these 4 guys up before him: 8, 9, 1, 2, while the 4 hitter has 9, 1, 2, 3 and the 5 hitter always has 1, 2, 3, 4 before him, counting the first inning- all the best hitters on the team.

 

2) One question I'd ask about these studies that show the 3rd slot hitter should be your 5th best hitter is this: did the study differentiate between great hitting teams and poor ones? When you have a team with 5 guys with an OBP above .350 and an OPS above .800, maybe the numbers change. Maybe teams with crappy hitters skews the numbers.

 

I'll not complain when JD bats 3rd. I think Cora knows what he's doing, and right now Betts, Bogey, JD and Devers are pretty interchangeable, except for the speed component, so there shouldn't be much swing anyway we go. (Even Beni is heating up right now.)

[notin]

This may come down to a discussion about what is a "good" hitter and what they bring to the lineup.

 

OPB and OPS have something in common. The common element is that they both are dependent on how often a player reaches base. It's just that the player with the higher OPS has more power. Since even a successful hitter only reaches base about one time out of three I want my #3 hitter to have that one time out of three to have that power. I'll take that over my #3 hitter maybe getting a walk or getting a single and leaving the big scoring opportunity to the #4 hitter... who also is unsuccessful two times out of three.

Nobody is going to be successful every time, but optimize your chances to score.

 

Edit... The more I think about this the more amazed I am that the stat geeks have decided that the better hitter should be the #2 guy and that the #3 guy isn't as important as even the #4 or #5 guy.

 

Equally as amazing is that people bought into it!

 

 

According to moon’s numbers, the #3 hitter comes up 44% of the time with 2 outs and no one on in the first inning. But that follows through in the game.

 

If every hitter has an equal chance of leading off an inning, then the leadoff spot comes up 1st another 144 times, with the same percentages.

 

So in the first inning, that means 72 times per a 162 game season, the third hitter comes up with 2 out and none on, and in the following innings does so another 57 times. (I’m using moon’s .333 OBPs, but even if you increase them the numbers don’t vary by much).

 

So that means 129 times per year, the #3 hitter is up with 2 out and no one on. In a 650 PA season, that represents 20% of his plate appearances. If you up the OBP to .400 for both hitters, the number drops to 18.7%. But it’s in the neighborhood of 1/5 of his PA.

 

The argument rests on whether any other spot in the top 5 exceeds that number. And they don’t solely because the lineup is set up in the first inning.

 

In the 80%-ish PA where he does come up with men on, the 4th hitter should also bat unless there is a DP. Which is why the third hitter needs to be capable of avoiding them.

 

The 3 slot isn’t for bad hitters; he should still be one of the best. But because he is more likely than any other hitter in the top 5 to come up with two out and no one on, his run producing opportunities are limited.

 

And he should be able to stay out of double plays...

Edited June 7, 2019 by notin

[moonslav59]

According to moon’s numbers, the #3 hitter comes up 44% of the time with 2 outs and no one on in the first inning. But that follows through in the game.

 

If every hitter has an equal chance of leading off an inning, then the leadoff spot comes up 1st another 144 times, with the same percentages.

 

So in the first inning, that means 72 times per a 162 game season, the third hitter comes up with 2 out and none on, and in the following innings does so another 57 times. (I’m using moon’s .333 OBPs, but even if you increase them the numbers don’t vary by much).

 

So that means 129 times per year, the #3 hitter is up with 2 out and no one on. In a 650 PA season, that represents 20% of his plate appearances. If you up the OBP to .400 for both hitters, the number drops to 18.7%. But it’s in the neighborhood of 1/5 of his PA.

 

The argument rests on whether any other spot in the top 5 exceeds that number. And they don’t solely because the lineup is set up in the first inning.

 

In the 80%-ish PA where he does come up with men on, the 4th hitter should also bat unless there is a DP. Which is why the third hitter needs to be capable of avoiding them.

 

The 3 slot isn’t for bad hitters; he should still be one of the best. But because he is more likely than any other hitter in the top 5 to come up with two out and no one on, his run producing opportunities are limited.

 

And he should be able to stay out of double plays...

 

My numbers were based on the #1 and #2 hitters having a .333 OBP and no DPs, CS or picked offs before the #3 guy comes up.

 

So far this year, our top 2 slots are...

 

1. .328

2. .394

 

Last year,

 

1. .418

2. .340

 

The number is probably around 40% of the time, for us, that the 3 hitter comes up with 2 outs and none on in the first inning.

 

 

 

 

[notin]

My numbers were based on the #1 and #2 hitters having a .333 OBP and no DPs, CS or picked offs before the #3 guy comes up.

 

So far this year, our top 2 slots are...

 

1. .328

2. .394

 

Last year,

 

1. .418

2. .340

 

The number is probably around 40% of the time, for us, that the 3 hitter comes up with 2 outs and none on in the first inning.

 

 

 

 

 

And in 18-20% of all his plate appearances. That’s very likely the highest among the top 5 hitters. The 6th hitter is probably the second most likely, but he’s far enough down to not worry about. He should also be getting 54 fewer total plate appearances than the third spot...

[Kimmi]

I played around with Baseball Musings Line Up Analysis tool, just for funsies. FTR, I am aware of the flaws of this tool, but it was interesting to look at nonetheless.

 

Using last year's numbers, and the line up that I remember Cora going with regularly (Betts, Beni, Mitch, JD, Xander, Devers, Holt, Vazquez, JBJ), the team projects to score 5.615 runs per game, or 909.63 runs over the season.

 

When I moved JD to the #3 hole, Xander to #4, and dropped Mitch to #5, the team projects to score 5.599 runs per game, or 907.083 run over the season.

 

Two points of note here:

1. The line up does worse with JD in the #3 hole.

2. Tweaks to the line up like the one I made above make very little difference.

 

Mic drop!

[Kimmi]

Also interesting to note, not a single one of the top 30 'best line ups' has JD in the 3rd spot. Most have Holt leading off, Mookie batting 2nd, JD 4th, and various combinations of Mitch, Beni, and Xander in the 3rd, 5th, and 6th spots.

 

Again, this is based off of last year's numbers.

[Kimmi]

One final note from Jack Moore at Fangraphs:

 

"When it comes to sabermetric studies, no single item sees more energy expended with less gain than the analysis of batting orders. The Book basically opened and shut the door on the issue: the best three hitters should bat first, second, and fourth, but even the most egregious of lineup errors won’t cost a team more than a win. It’s also more important to split lefties to avoid LOOGYs than it is to get that perfectly chained lineup."

[Kimmi]

Whether one is a fan of analytics or not, I think we all have to agree that the argument that managers have always put their best hitter in the #3 spot and, therefore, it must be the right thing to do is a really weak argument.

 

For one, how would anyone know that there wasn't a better lineup? Someone explain to me how, for all those years, they knew that putting the best hitter in the #3 spot gave resulted in the most production.

 

Remember, these are the same managers that were at the same time putting one of their weakest hitters in the #2 hole. I think we all agree now, thanks to analytics, that that was a pretty poor decision, even though 'that's the way it's always been done'.

 

For two, run scoring environments change. Players change. What worked best in the 1950s might not necessarily be what worked best in the 1980s. Did anyone bother to check?

[Bellhorn04]

For two, run scoring environments change. Players change. What worked best in the 1950s might not necessarily be what worked best in the 1980s. Did anyone bother to check?

 

Check what and where, Kimmi?

[Kimmi]

Check what and where, Kimmi?

 

My point is that until the stat geeks started doing their thing, everyone assumed that the way the managers were managing was correct because that's the way that it's always been done. No one ever questioned it. No one ever actually 'checked' through any kind of statistical analysis. So how do they know what was right or wrong?

 

The analytics guys questioned these things and started looking for answers. They found many that contradict conventional wisdom.

 

Studies might show that over the next 10 years, the best hitter should hit in the #3 spot, due to Launch Ball or some other reason. Then again, they might come to the same conclusion as those found in The Book. But at least they will continue to search for answers.