The cost of poor lineup optimization

E

Enscheff

Well-known member
I recently found BRef's play index event finder after reading a FG chat with Jeff Sullivan. He used it to answer a question about how many PAs a particular spot in the lineup has had RISP.

I did a similar check for all team in 2016, and got results that seem reasonable (% values added by me):

4th: 5961 (12.93%)
3rd: 5579 (12.10%)
5th: 5487 (11.90%)
6th: 5011 (10.87%)
2nd: 4955 (10.74%)
7th: 4927 (10.68%)
8th: 4927 (10.68%)
9th: 4794 (10.40%)
1st: 4475 (9.70%)

Teams want their 3-4-5 hitters to hit with RISP, and the league as a whole accomplished that. The league as a whole was able to focus 37% of the PAs with RISP into the 3 slots in the lineup traditionally occupied by the highest impact hitters.

Looking at the Braves in 2016:

4th: 227 (13.67%)
8th: 198 (11.92%)
5th: 197 (11.86%)
9th: 187 (11.26%)
7th: 177 (10.66%)
3rd: 173 (10.42%)
1st: 171 (10.30%)
6th: 169 (10.17%)
2nd: 162 (9.75%)

Atlanta, we have a problem. Freeman should have had 201 PAs with RISP with league average lineup construction. That is 28 PAs with RISP that didn't go to Freeman, and instead went to the Braves 8th place hitter (mostly).

So what happened? Why did the Braves best hitter come to the plate in only 10.42% of the team's PAs with RISP when the rest of the league would have had him at the plate in 12.10% of those PAs?

The most obvious answer is to look at who batted in front of Freeman. So using the same tool, here are the PAs in the 2 hole from last season:

Adonis Garcia 217
Ender Inciarte 113
Chase d'Arnaud 111
Gordon Beckham 103
Erick Aybar 77
Daniel Castro 63
Nick Markakis 40
Dansby Swanson 4
Jeff Francoeur 2
Reid Brignac 2
Drew Stubbs 1
Julio Teheran 1
Rio Ruiz 1

There's the problem! The Braves insisted on treating the #2 hole in the lineup like it was the place to stick a garbage hitter just because he batted RHed. Almost 600 PAs went to hitters named Garcia, d'Arnaud, Beckham, Aybar and Castro. Ouch!

Remember when Garcia was "hot" for like 2 weeks and everyone said it was smart to stick him in the 2 hole? Remember when I said he would be left there way too long and hurt the team's offense?

How bad did it hurt the offense?

It robbed the best hitter on the team of 28 PAs with RISP.
 
Good stuff. We all know the Braves coaching staff isn't that bright

Looks like the league is starting to catch up with being smart. I love what the Cubs are doing - though they also have an embarrassment of riches to play with.
 
Remember when the Braves now have Dansby in the 2-hole, so it's a moot point?

It's a good thing to point out, and we definitely didn't have great roster construction last year, I don't think there's any doubt about that. But lineup construction ultimately doesn't have that much of an impact on wins/losses, and the Braves' primary issue has been that they don't have enough good hitters, period.

The main reason the league as a whole concentrates their RISP PA in the 'middle' of the order is because you stick your best hitters at the top, they get the most PAs, and once you get to 3-4-5 there have been enough good hitters to increase the chances of one of them getting into RISP.

It's not like the league has figured out optimal lineup construction. That's just the way it's going to work. Once the Braves have enough good hitters, you'll see similar numbers to the rest of the league because it won't matter as much who you stick in what spot.
 
smootness said:
Remember when the Braves now have Dansby in the 2-hole, so it's a moot point?

It's a good thing to point out, and we definitely didn't have great roster construction last year, I don't think there's any doubt about that. But lineup construction ultimately doesn't have that much of an impact on wins/losses, and the Braves' primary issue has been that they don't have enough good hitters, period.

The main reason the league as a whole concentrates their RISP PA in the 'middle' of the order is because you stick your best hitters at the top, they get the most PAs, and once you get to 3-4-5 there have been enough good hitters to increase the chances of one of them getting into RISP.

It's not like the league has figured out optimal lineup construction. That's just the way it's going to work. Once the Braves have enough good hitters, you'll see similar numbers to the rest of the league because it won't matter as much who you stick in what spot.
Click to expand...

I think this is one of his dumbest posts yet. Like he is struggling to find ways to dis the Braves. Has he not found a good white suxs board to chat on yet?
 
sturg33 said:
Good stuff. We all know the Braves coaching staff isn't that bright

Looks like the league is starting to catch up with being smart. I love what the Cubs are doing - though they also have an embarrassment of riches to play with.
Click to expand...

You bring up an interesting point with the Cubs. They typically bat Bryant 2 or 3, and here is their breakdown:

4th 253
5th 226
3rd 203
6th 202
7th 199
8th 197
9th 177
1st 166
2nd 165

which ultimately led to these PAs with RISP for each hitter:

Anthony Rizzo 213
Ben Zobrist 198
Addison Russell 194
Kris Bryant 164
Jason Heyward 153
Javier Baez 130
Dexter Fowler 119
Miguel Montero 92
Jorge Soler 80
Willson Contreras 79

Rizzo getting tons of PAs with RISP is good. Bryant getting less PAs with RISP than guys like Zobrist and Russell is not good.

As good as the Cubs are, it looks like they have some room to optimize a bit and get Bryant a few dozen more PAs with RISP in 2017.
 
bravesfanMatt said:
who was supposed to hit 2nd last year?

Adonis hit .285/.318 out of that spot.
Click to expand...

How about someone that could have gotten on base? You know, the guy that was getting on base in front of the 8th and 9th place hitters that allowed those 2 lineup spots to accumulate so many PAs with RISP.

I'll leave it to you to figure out the 2 guys that should have been hitting 2nd. You're smart, you can figure it out.
 
smootness said:
Remember when the Braves now have Dansby in the 2-hole, so it's a moot point?

It's a good thing to point out, and we definitely didn't have great roster construction last year, I don't think there's any doubt about that. But lineup construction ultimately doesn't have that much of an impact on wins/losses, and the Braves' primary issue has been that they don't have enough good hitters, period.

The main reason the league as a whole concentrates their RISP PA in the 'middle' of the order is because you stick your best hitters at the top, they get the most PAs, and once you get to 3-4-5 there have been enough good hitters to increase the chances of one of them getting into RISP.

It's not like the league has figured out optimal lineup construction. That's just the way it's going to work. Once the Braves have enough good hitters, you'll see similar numbers to the rest of the league because it won't matter as much who you stick in what spot.
Click to expand...

Then why did the 8 and 9 slots get so many PAs with RISP? Someone was getting on base, but they weren't hitting 2nd.

Hmmmm....a riddle!
 
Enscheff said:
You bring up an interesting point with the Cubs. They typically bat Bryant 2 or 3, and here is their breakdown:

4th 253
5th 226
3rd 203
6th 202
7th 199
8th 197
9th 177
1st 166
2nd 165

which ultimately led to these PAs with RISP for each hitter:

Anthony Rizzo 213
Ben Zobrist 198
Addison Russell 194
Kris Bryant 164
Jason Heyward 153
Javier Baez 130
Dexter Fowler 119
Miguel Montero 92
Jorge Soler 80
Willson Contreras 79

Rizzo getting tons of PAs with RISP is good. Bryant getting less PAs with RISP than guys like Zobrist and Russell is not good.

As good as the Cubs are, it looks like they have some room to optimize a bit and get Bryant a few dozen more PAs with RISP in 2017.
Click to expand...

True... but the question comes down to this:

Is it better for Zobrist to provide Bryant with less opportunities, but better results

Or is better for Bryant to provide Zobrist with more opportunities, but worse results

Bryant can't get on base for himself. And you flipped him, you'd have the better hitter getting the RISP chances, but there would theoretically be less of them, no?
 
sturg33 said:
True... but the question comes down to this:

Is it better for Zobrist to provide Bryant with less opportunities, but better results

Or is better for Bryant to provide Zobrist with more opportunities, but worse results

Bryant can't get on base for himself. And you flipped him, you'd have the better hitter getting the RISP chances, but there would theoretically be less of them, no?
Click to expand...

Right, a bit of a chicken and the egg conundrum going on here. Do you want Bryant batting more often in total, or batting more often with RISP?

Instinctively, I want to say the most important thing is to maximize the number of chances Rizzo/Bryant and Freeman/Kemp get to bat with RISP. That matters more than the total number of times those players come up to bat (which is influenced by lineup position, and is why Bryant bats 2nd).

Obviously, the simple act of Freeman hitting in front of Kemp and Bryant hitting in front of Rizzo give Kemp and Rizzo many opportunities with RISP. But how do you maximize RISP opportunities for Bryant and Freeman?

I think the bottom line is that if your 8th and 9th place hitters are near the top of this list, there is a severe flaw in your lineup construction.
 
Enscheff said:
How about someone that could have gotten on base? You know, the guy that was getting on base in front of the 8th and 9th place hitters that accumulated so many PAs with RISP.

I'll leave it to you to figure out the 2 guys that should have been hitting 2nd. You're smart, you can figure it out.
Click to expand...

So your 'optimal' lineup last year would have been Ender/Neck/FF..
 
Enscheff said:
Right, a bit of a chicken and the egg conundrum going on here. Do you want Bryant batting more often in total, or batting more often with RISP?

Instinctively, I want to say the most important thing is to maximize the number of chances Rizzo/Bryant and Freeman/Kemp get to bat with RISP. That matters more than the total number of times those players come up to bat (which is influenced by lineup position, and is why Bryant bats 2nd).

Obviously, the simple act of Freeman hitting in front of Kemp and Bryant hitting in front of Rizzo give Kemp and Rizzo many opportunities with RISP. But how do you maximize RISP opportunities for Bryant and Freeman?

I think the bottom line is that if your 8th and 9th place hitters are near the top of this list, there is a severe flaw in your lineup construction.
Click to expand...

Agreed with your last sentence.

This is interesting to me. I've always felt it's best to get your best hitters more ABs (which is why I like what the Cubs are doing)... but perhaps that's not the way to go.

I suppose we could solve it by looking at the number of opportunities Bryant creates for Zobrist, and then using Zobrist's average slash line to get an idea of run production, and vice versa to see which creates more runs
 
Enscheff said:
Then why did the 8 and 9 slots get so many PAs with RISP? Someone was getting on base, but they weren't hitting 2nd.

Hmmmm....a riddle!
Click to expand...

First, I'm not sure your numbers add up. The Braves, according to you, were well above MLB average in # of PAs with RISP. I find that difficult to believe, especially once you take the pitcher's spot into account.

But assuming your numbers are correct, I've already said the Braves didn't have good lineup construction last year. My point is simply that as you fill up the lineup with better hitters throughout, lineup construction doesn't really matter much.

And PAs with RISP is just one way to evaluate lineup construction. I do think it would help the Cubs if they had Bryant getting more chances with RISP...but he also had about as high an OBP as anybody on the team, and the 2nd spot will also get more PAs throughout the year than the 3-5 spots. So you have to take several things into account, not just look at one stat and determine optimal strategy based on that one stat. But with your intellect, I'm not sure nuance is something you're capable of considering, so carry on.
 
sturg33 said:
Agreed with your last sentence.

This is interesting to me. I've always felt it's best to get your best hitters more ABs (which is why I like what the Cubs are doing)... but perhaps that's not the way to go.

I suppose we could solve it by looking at the number of opportunities Bryant creates for Zobrist, and then using Zobrist's average slash line to get an idea of run production, and vice versa to see which creates more runs
Click to expand...

If it was my job to analyze this stuff, you can bet your ass I would have the most detailed lineup simulator imaginable as one of my top pet projects. It would be a piece of my larger "DecisionMaker-O-Matic" program managers had on a tablet in the dugout.
 
smootness said:
First, I'm not sure your numbers add up. The Braves, according to you, were well above MLB average in # of PAs with RISP. I find that difficult to believe, especially once you take the pitcher's spot into account.

But assuming your numbers are correct, I've already said the Braves didn't have good lineup construction last year. My point is simply that as you fill up the lineup with better hitters throughout, lineup construction doesn't really matter much.

And PAs with RISP is just one way to evaluate lineup construction. I do think it would help the Cubs if they had Bryant getting more chances with RISP...but he also had about as high an OBP as anybody on the team, and the 2nd spot will also get more PAs throughout the year than the 3-5 spots. So you have to take several things into account, not just look at one stat and determine optimal strategy based on that one stat. But with your intellect, I'm not sure nuance is something you're capable of considering, so carry on.
Click to expand...

Allow me to coax your intellect along a bit...

A team gets a RISP, and the next batter drives them in. The next PA is likely not one with RISP.

The Braves get s RISP, but the next batter is likely the 8th or 9th place hitter, who fails to drive the guy in. The next batter comes to the plate with the same RISP.

Do you see how failing to drive in runners can cause there to be more PAs with RISP?

That's why I used those fancy things called percentages, so the absolute numbers didn't matter. That's why percentages were invented, after all.
 
I'm also reminded that the cubs traded a pitcher for Rizzo, then traded a pitcher for Russell. Then drafted Baez (hitter), Bryant (hitter), and Schwarber (hitter) with their high first round picks.

Seems like a solid strategy
 
sturg33 said:
I'm also reminded that the cubs traded a pitcher for Rizzo, then traded a pitcher for Russell. Then drafted Baez (hitter), Bryant (hitter), and Schwarber (hitter) with their high first round picks.

Seems like a solid strategy
Click to expand...

The Cubs traded 2 solid starters for those guys. Who on the Braves staff could we have traded. We traded Miller for a good return. Wood was a big whiff.. Who else could they have traded.. JT??

The Cubs also took 5+ years.. we are entering our 3rd. Let things develop.
 
bravesfanMatt said:
The Cubs traded 2 solid starters for those guys. Who on the Braves staff could we have traded. We traded Miller for a good return. Wood was a big whiff.. Who else could they have traded.. JT??

The Cubs also took 5+ years.. we are entering our 3rd. Let things develop.
Click to expand...

We traded Heyward for pitching.

We traded Upton for pitching.

We traded Kimbrel for pitching.

We traded Simmons for pitching.

We traded Gattis for pitching.

Those were some really really valuable assets.

And in 2015, with the #14 and #28 picks, we took pitching

And in 2016, with the #3, #35, and #41 picks, we took pitching.

Not saying they weren't good pitchers... but the Cubs are loaded because they're lineup is loaded
 
sturg33 said:
We traded Heyward for pitching.

We traded Upton for pitching.

We traded Kimbrel for pitching.

We traded Simmons for pitching.

We traded Gattis for pitching.

Those were some really really valuable assets.
Click to expand...

fair point.. I was narrowly thinking about pitchers we traded.
 
You must log in or register to reply here.
Back
Top