The myth of the balanced lineup

[Greenmonsters]

Not a new BB analysis, but a new analysis using BB data. I am hopeful that I can do something that benefits FP with the data.

The inning difference is huge. Huge. And I missed it, so thanks for that! I'm not interested in using TB data, though. The NCAA data is what I'd want to use eventually, which could then be applied to TB and even rec.

MLB is where the data is. It may not translate directly, but IMO it's better than "Well, that's just how we've always done it," as the rationale for lineup creation.

Another wrench in the works - there are run rule games in FP

 

[Greenmonsters]

So 1 bats 782.46, 3 bats 746.82, 4 bats 727.38 and 5 bats 711.18 in a season if you just carry the numbers out in 162 games. Runs in baseball are normally very hard to come by and when you very commonly have less than 10 hits in a game of a 9 inning game (also very little errors), you want your better hitters getting 35-70 more at bats in a season. For a .300 hitter, that's 10 (if he batted 3rd) to 21(1st) hits a year more than you would if he batted 5th.

This is why MLB does lineups like they do in my opinion.

One big difference that I don't know how you would account for is that in baseball the pitcher has to work from the stretch v. full wind up in most runners on base situations.

 

[BuckeyeGuy]

And do you agree with the book that the difference between a traditional and optimized line up is 1 win over a 162-game season?

I know you didn't ask me, but the claim is actually 10-15 runs over a season. That would translate to 1 or 2 wins.

Do you think the delta between the best and 9th hitter is the same in MLB and FP lineups? I don't. I also think that difference gets magnified in the FP game when facing an elite pitcher.

Same? No. Similar? Probably. Any way to determine that? Not yet, not until we have several to many years of NCAA data.

Another wrench in the works - there are run rule games in FP

In NCAA? I wasn't aware. That'd definitely change the results if there were a significant number of them every year (by "significant" I mean statistically significant).

I was doing some reading on Sabermetrics. Turns out that back in the 40's there was most of the data we see today, but it was kept by the teams. When writers asked for the data, the teams told them to go the heck away, they weren't going to give away "intelligence-type" information. I think that's where we are today with FP.

 

[Jquick823]

That's an unusually high pct of last PA's for 1 and 2. I hope most were walk-offs...

We batted 11 so they were at bat with 2 outs more often. #3 usually didn't bat in our last ups just based on how it fell. We did well as a young 12U team but many were not walk offs in our favor lol.

 

[BuckeyeGuy]

So 1 bats 782.46, 3 bats 746.82, 4 bats 727.38 and 5 bats 711.18 in a season if you just carry the numbers out in 162 games. Runs in baseball are normally very hard to come by and when you very commonly have less than 10 hits in a game of a 9 inning game (also very little errors), you want your better hitters getting 35-70 more at bats in a season. For a .300 hitter, that's 10 (if he batted 3rd) to 21(1st) hits a year more than you would if he batted 5th.

This is why MLB does lineups like they do in my opinion.

And also why they're doing it wrong. They've cherry picked a stat (PA/game or PA/season) and used it to justify going half way to being statistically sound. And they've done it in such a way as to put their best hitter in a spot where he bats the most often with 2 outs/none on and least often with no outs.

No, I don't think the MLB translates much at all to TB. Fastpitch is not baseball. They are 2 different sports and in my opinion should be treated as such when analyzing them. Strategies for each sport differ, thus you would set your batting order differently. Those 2 innings make a big difference. I think the number 3 hitter on a travel FP team sees a significantly higher amount of at bats due to time limits and few innings played.

I don't completely disagree. So, where is all the FP data to be analyzed? We are certainly getting there, but until we have all the data, all we can do is extrapolate from decades of MLB data. Is that a valid approach? Maybe. Most of the interesting new stats that apply to managing the game are derived from the run expectancy (RE) results (how may runs/inning for each base/out state). I would contend that if there is not a relatively large difference in RE between MLB and MCAA FP, we can at least use the MLB metrics as a guide, a starting point for the discussion. So, are the RE's close? You decide:



I would love to get my hands on NCAA data and mine it for results like we have for MLB. Based on the 2 inning/game difference alone, I think maybe there will be differences in how we'd want a lineup structured. It's a great question!

 

[JAD]

A couple things you need to add into the equation of creating a batting order in softball vs. baseball:

1) Some levels of softball allow pinch runners for pitchers and catchers.
2) Players are allowed to exit and enter the game more than once.
3) Some marquee teams carry 16+ players on the roster with some players sole purpose being to pinch run and steal bases.
4) The shorter base paths in softball put a lot more emphasis on speed.
5) A single + a stolen base = double

 

[GunnerShotgun]

And do you agree with the book that the difference between a traditional and optimized line up is 1 win over a 162-game season?

I agree that's what that book referenced in regards to baseball. Those that keep bringing up baseball and the differences between the two sports are making this discussion cloudy.

I coach fastptich and using fastpitch data we can see that there are ways to optimize your lineup. You aren't looking for wins per say but looking to increase run expectancy (which in theory leads to more wins). You need to be looking at projections based on experience when you make your lineup. I also stated *in my opinion* that the numbers aren't absolute in that certain players don't like to hit (don't perform well) in certain batting slots. You should factor this bias into your calculations.

For those of you that like to go traditional and go with your gut. Rock on. I love it when the opposing coaches punt when facing 4th down and less than two yards as well.

"Why do something wrong, when you can do it right?”

 

[BuckeyeGuy]

A couple things you need to add into the equation of creating a batting order in softball vs. baseball:

1) Some levels of softball allow pinch runners for pitchers and catchers.
2) Players are allowed to exit and enter the game more than once.
3) Some marquee teams carry 16+ players on the roster with some players sole purpose being to pinch run and steal bases.
4) The shorter base paths in softball put a lot more emphasis on speed.
5) A single + a stolen base = double

The lineup is predicated on the run expectancy of a given base/out/batting slot state. What should happen after the PA is governed by a completely different set of probabilities that apply to the base/out state and the success:failure ratio of your base stealer.

MLB example:

With a runner on 1st, no outs, a successful steal adds .24 runs per inning, caught stealing costs .68 runs per inning. Your break even success rate here is 73.6% - if you are successful stealing 73.6% of the time, go; if you are successful less than 73.6% of the time, don't go.

I get what you're saying: given the "interesting" substitution rules in FP, you can create a favorable base/out/batting slot state almost at will. That would make it even more important to know the probabilities. This is good stuff, really got to think about all this.

 

[GunnerShotgun]

5) A single + a stolen base = double

A single and a stolen base does not equal a double.

 

[MNDad]

Another wrench in the works - there are run rule games in FP

In NCAA? I wasn't aware. That'd definitely change the results if there were a significant number of them every year (by "significant" I mean statistically significant).

In NCAA the run rule is 8 after 5. I looked up a few random college teams and anywhere from 15%-30% of their games were stopped short due to the run rule. That would be pretty significant.