Twins Tidbit: Does Rocco Baldelli Pull Starting Pitchers Too Quickly?

[TwinsData]

A manager's decision to pull a starting pitcher or squeeze another inning out of them is often criticized. Rocco Baldelli, who tends to pull pitchers quickly, is no exception. But what goes into that decision and how should we assess the effectiveness of Baldelli's quick hook?

 

The above graph shows OPS (on-base percentage plus slugging percentage) against each starting pitcher separated by times through the order. So for example, it shows that hitters have a 0.786 OPS when hitting against Joe Ryan for the second time in a game. The size of the points indicate fraction of pitches. Chris Archer has only pitched to 10 hitters for a third time in a game (mercifully, it seems).

The pattern for most pitchers is clear: the scales tip toward the batter with each successive time through the lineup. So for Rocco Baldelli, the urge to call on the bullpen should strengthen each time opposing lineups turn over.

For comparison's sake, consider the group of middle relievers comprised of Tyler Duffey, Caleb Thielbar, Emilio Pagán, Jovani Moran, and Trevor Megill. That group has an OPS against of 0.636. So while turning to the bullpen often has not been pretty, it's still a much better option than having the likes of Chris Archer, Dylan Bundy, or Devin Smeltzer face a hitter for a 3rd time.

 

View full article

[Hosken Bombo Disco]

Thanks for posting! This is one of those enduring questions, with random sampling playing a role and variables that can be argued for either side. 

For example, when a starting pitcher is physically fit and is “making his pitches” that day, so to speak, and is already a high quality starting pitcher to begin with, I think the third and fourth time through the order can actually favor the pitcher. The pitcher has that data on what the batter is doing and swinging at, on that particular day.

what do you think?

[Squirrel]

LOL

[chpettit19]

But what does your gut say about it?

[Hosken Bombo Disco]

@chpettit19 @Squirrel

care to elaborate?

[Mark G]

If the computer spread dictated when a starter was pulled, wouldn't it be consistent each and every time with each starter?  Whenever each one entered the third porthole, they are done; middle of inning or end of it.  So why is it so inconsistent?  Because if it was consistent, we wouldn't be having this discussion multiple times a year when he makes us crazy again.    

As of last week our BP had pitched the 3rd most innings in the league.  And it shows, or we wouldn't be 5 games under .500 since May 24th.  So the strategy, and it is a strategy on Roc's part, to use 3-6 pitchers a game every game of the season, goes easy on the starters and burns out the pen.  And every year we debate if that is a good strategy.  And every year we rack up the injuries we are trying to avoid by limiting the starters pitches a game and innings in a year.  And round and round we go.    

[chpettit19]

11 minutes ago, Hosken Bombo Disco said:

@chpettit19 @Squirrel

care to elaborate?

It's the same discussion we go round and round on every game thread and recap. Some people like the strategy, some don't. Someone will inevitably complain that Rocco has no feel for the game and needs to use his gut. Others will point out stats that show the strategy is largely successful and others will complain that if they didn't baby guys so much they'd be better and able to last longer. Others will point out that there's far too many variables that go into these decisions to whittle it down to 1 variable and say it's always just a 3rd time through decision. Somebody is going to say their numbers are so high because they're facing the top of the order on the 3rd time through so it's skewed. I just wanted to get the gut comment out of the way before we dive into the rest of the back and forth.

[Squirrel]

47 minutes ago, Hosken Bombo Disco said:

@chpettit19 @Squirrel

care to elaborate?

I was laughing at the topic because this has never been discussed on here before

[Platoon]

It still doesn’t make sense to arbitrarily pull a SP at a set time. Maybe the stats say that jerking  SP in the fourth is working on an average? But surely individualizing this decision has to be part of the process. Nolan Ryan removed for Caleb Thielbar?   Or Joe Ryan for Joe Nathan?  Slight difference. 
 

The game requires x amount of innings by x amount of pitchers, the less innings you alot to your better ranked pitchers, pitchers able to get through a lineup at least twice, the more your lesser ranked pitchers, good for an inning, have to pitch. If you had a longer pen this might work. But in the current situation, you are throwing cannon fodder out there towards the end. 
 

You want to shorten the leash on a SP that’s one thing, but preplanning his inning limit seems more like scheduling than managing. 

[insagt1]

Look at history. For decades starters we're supposed to pitch deep into games. Today there are so many stats clogging up the works everything has been analyzed to death. So you have a staff of starters now conditioned to only pitch a handful of innings and a seriously overused bullpen. The game has gone cockeyed. Does everyone have to throw 100 mph to get hitters out today? I look at the 1971 Orioles pitching staff and remember when watching the good pitchers ply their craft was sheer fun.  Today you are a star if your ERA is 4.00 and you somehow struggle thru 5 innings.

[Doctor Gast]

We have to by pass our preferance and go with what we have and what works. I'd love to have 5 heaithy studs that can regular complete games but we don't have anyone that comes close, Or have 10 reliable BP arms to plug regularly into a 5 pitcher game, we have 1 (Duran) maybe Jax,

So trying to force our preferance into a formula that can't work is futile, We have 3 rookies which concensus is to limit their innings, our other SPs don't profile to pitch 6+ consecutive inning games. Trying to squeeze an extra inning or 2 has been disasterous. 

But we do have 4 establish SPs (Ryan, Grey, Bundy & Archer) and 4 spot starters/ long RPs (Ober, Winder, Smeltzer and Cotton). So the formula should be SP goes 5-6 innings / long relief goes 3 innings where they hand off to a closer or complete the game if need be.

So Is Baldelli pulling the SP too soon? I say no, he's leaving in SPs too long to help out the over relied short relief

[TwinsDr2021]

Besides Archer(17.63) the Twins are pretty consistent on how many batters they let their starters face on average.

Bundy - 21.19

Ryan - 21.86

Gray - 21.17

Smelzter - 21.33

Ober - 20.29

Winder - 21.14

From my research only six times all year has a Pitcher faced more than 25 batters (3 - Bundy, 1 each from Gray, Smelzter, and Ober) in a game and only twice more than 27 (Both Bundy).

What this tells me is while it seems wildly inconsistent on what Rocco is doing with the pitching staff from game to game, the plan seems to be to get a starting pitcher to end an inning right around the 18 batters faced mark even if that means pitching to a few guys a third time to accomplish that and in a rare few cases letting them pitch an extra inning if it is the bottom of the order. Also it tells me if you are a pitcher, you might as well go all out just about every pitch because it doesn't really matter how well you do or save yourself for late in the game, it isn't happening.

IMO the Twins have ZERO plans on changing this plan, because from what I can tell this is exactly what is happening in the minors as well, so hoping we get some stud like Shane Bieber for example isn't going to happen, thus also the reason I don't see them signing a high priced free agent starter, why would the Twins sign a guy like Musgrave/Ray/Rondon/etc.. to a high price salary when you are going to limit their innings, and if you are one of those guys and you want to leave a mark on being one of the best pitchers in your generation why would sign here? Also IMO that is why the Happ/Shoemaker/Archer/Bundy's of the world are willing to sign here because that takes advantage of their skills.

Just took a peak at Berrios and in 20 his average was around 22, and the start of 21 it was around 21-22 until mid may, but after than average around 26, speculation maybe that is when the two sides decided the Twins wasn't his future and the Twins realized they needed to pitch him more to showcase him? He only pitched less than 6 innings in 3 of his last 15 starts here.  and then again instead of being on to something maybe I am just on something.

 

[jorgenswest]

Did you know that batters 7 through 9 in the order have an OPS against Bundy of .582? The top two hitters in the order have an OPS against of .994. It probably isn’t that shocking as the top of the order is way better than the bottom of the order.

This has significant meaning for his split of 55 batters faced the third time through the order. Of those 55 a total of 5 are from the bottom three spots in the order. The top two spots have seen him a third time 23 times. The data for the third time through is made up almost entirely of the best hitters. Of course Bundy will do much worse in that context.

Do you wonder why we don’t here the same in context of hitters? You would expect a veteran like Correa to have a real advantage seeing a pitcher the third time through. The first time Correa is facing a pitcher this year his OPS is .962. The third time it is .724. Wouldn’t it be wise to pinch hit for Correa in this situation? Clearly the pitcher has figured something out the first two at bats. It is the same with Kepler. He really falls off the third time through to .564.

The reality is the data set is skewed for each. The data set for pitchers the third time through is mostly the top part of the order and the better hitters. It is the same for batters. The set of pitchers that make it to them a third time is going to be the better pitchers or a pitcher really on top of his game that day. It probably is more extreme for Kepler hitting further down in the line up. OPS also needs a pretty large sample size so that variance also plays in.

Is Bundy going to have trouble with the top of the order? Absolutely. He will have trouble with them the first, second and third time he sees them. I don’t think the data says that his third time against the same top of the order will be significantly more troublesome.

[ashbury]

Rocco needs to get his nose out of his spreadsheets and recognize the wisdom of the above charts.

 

 

/ there, did I manage to piss off everybody in one sentence?

[Platoon]

5 hours ago, Doctor Gast said:

We have to by pass our preferance and go with what we have and what works. I'd love to have 5 heaithy studs that can regular complete games but we don't have anyone that comes close, Or have 10 reliable BP arms to plug regularly into a 5 pitcher game, we have 1 (Duran) maybe Jax,

So trying to force our preferance into a formula that can't work is futile, We have 3 rookies which concensus is to limit their innings, our other SPs don't profile to pitch 6+ consecutive inning games. Trying to squeeze an extra inning or 2 has been disasterous. 

But we do have 4 establish SPs (Ryan, Grey, Bundy & Archer) and 4 spot starters/ long RPs (Ober, Winder, Smeltzer and Cotton). So the formula should be SP goes 5-6 innings / long relief goes 3 innings where they hand off to a closer or complete the game if need be.

So Is Baldelli pulling the SP too soon? I say no, he's leaving in SPs too long to help out the over relied short relief

Would not pulling the starters even earlier simply cause the bullpen more stress.  Years ago Bill Clinton said "it’s the arithmetic".  Whatever your SP doesn’t pitch your pen does. In another context if you had a group of starters consistently going deep into games, you would either drop a relief pitcher, or end up with rusty ones at the end of the pen bench! 

[Spth]

4 hours ago, jorgenswest said:

Did you know that batters 7 through 9 in the order have an OPS against Bundy of .582? The top two hitters in the order have an OPS against of .994. It probably isn’t that shocking as the top of the order is way better than the bottom of the order.

This has significant meaning for his split of 55 batters faced the third time through the order. Of those 55 a total of 5 are from the bottom three spots in the order. The top two spots have seen him a third time 23 times. The data for the third time through is made up almost entirely of the best hitters. Of course Bundy will do much worse in that context.

Do you wonder why we don’t here the same in context of hitters? You would expect a veteran like Correa to have a real advantage seeing a pitcher the third time through. The first time Correa is facing a pitcher this year his OPS is .962. The third time it is .724. Wouldn’t it be wise to pinch hit for Correa in this situation? Clearly the pitcher has figured something out the first two at bats. It is the same with Kepler. He really falls off the third time through to .564.

The reality is the data set is skewed for each. The data set for pitchers the third time through is mostly the top part of the order and the better hitters. It is the same for batters. The set of pitchers that make it to them a third time is going to be the better pitchers or a pitcher really on top of his game that day. It probably is more extreme for Kepler hitting further down in the line up. OPS also needs a pretty large sample size so that variance also plays in.

Is Bundy going to have trouble with the top of the order? Absolutely. He will have trouble with them the first, second and third time he sees them. I don’t think the data says that his third time against the same top of the order will be significantly more troublesome.

This makes a lot of sense. Thanks! I’d love to see the above chart broken down by each third of the batting order. 

[Doctor Gast]

1 hour ago, Platoon said:

Would not pulling the starters even earlier simply cause the bullpen more stress.  Years ago Bill Clinton said "it’s the arithmetic".  Whatever your SP doesn’t pitch your pen does. In another context if you had a group of starters consistently going deep into games, you would either drop a relief pitcher, or end up with rusty ones at the end of the pen bench! 

Yes it's arithmetic capable long relief covers for 3 incapable short relief. Incorporate long relief corp regularly,  that has been neglected, will take off the stress from both the rotation & short relief. You implement 3 extra innings from long relief that has been there before you don't need to squeeze extra innings from the rotation and not have to rely so heavily on short relief. 4 long relief pitchers goes a long ways.

Long relief are spot starters are just as much like the extention of the rotation as the BP. With long relief we can drop a few of the incapable short RPs. 4 long RPs can eat alot of innings

[Aerodeliria]

One can play these numbers games all day. What is the bullpen's OPS since June 15th? Not too good I imagine...

[Eris]

17 hours ago, insagt1 said:

Look at history. For decades starters we're supposed to pitch deep into games. Today there are so many stats clogging up the works everything has been analyzed to death. So you have a staff of starters now conditioned to only pitch a handful of innings and a seriously overused bullpen. The game has gone cockeyed. Does everyone have to throw 100 mph to get hitters out today? I look at the 1971 Orioles pitching staff and remember when watching the good pitchers ply their craft was sheer fun.  Today you are a star if your ERA is 4.00 and you somehow struggle thru 5 innings.

Inspired by your post I looked up Mike Cuellar’s stats. In 1969, the year he won 23 games and the Cy Young, his K/9 was 5.64. 
 

https://www.fangraphs.com/players/mike-cuellar/1002853/stats?position=P

On our current squad only Delvin Smeltzer and Joe Smith have a K/9 that is less.  All data is from Fangraphs. 
https://www.fangraphs.com/leaders.aspx?pos=all&stats=pit&lg=all&qual=10&type=8&season=2022&month=0&season1=2022&ind=0&team=8&rost=0&age=0&filter=&players=0&startdate=2022-01-01&enddate=2022-12-31&sort=8,d

One item that would add value to the chart is the number of pitches thrown per game. As an example, Joe Ryan averages 86 pitches per game and 5.35 innings pitched per game. Joe Ryan is likely being removed because he is reaching his pitch count limits rather than because he is ineffective.   
 

Three true outcomes has brought us here. The effect is a burned out bullpen—as many posters have indicated above.  Unless the game changes, a long term solution is expand rosters to add about 4 relief pitchers. The challenge with that is your 17th best pitcher isn’t as good as the 11th best pitcher. 

[TwinsDr2021]

12 hours ago, Doctor Gast said:

Yes it's arithmetic capable long relief covers for 3 incapable short relief. Incorporate long relief corp regularly,  that has been neglected, will take off the stress from both the rotation & short relief. You implement 3 extra innings from long relief that has been there before you don't need to squeeze extra innings from the rotation and not have to rely so heavily on short relief. 4 long relief pitchers goes a long ways.

Long relief are spot starters are just as much like the extention of the rotation as the BP. With long relief we can drop a few of the incapable short RPs. 4 long RPs can eat alot of innings

Maybe the solution to this problem is 8 starters that go 4 innings,

Smelter/Gray

Ryan/Bundy

Archer/Winder

Ober/Jax

That caps all their innings at 160 if perfectly healthy and go 4 innings each time. Then you have 5 short guys to rotate though. Duran, Thielbar, Pagan, Smith Duffey for example, and if need be the bus system to St. Paul.

Nobody would ever have to face a batter the super scary 3rd time, Seems to completely take injuries out of play by limiting pitches and innings, seems like the perfect solution plus you never have to spend big money on a starting pitcher because no starting pitcher worth anything would agree to that. So you have piles and piles of money to spend on offense/defensive. (I mean I hate everything about it, but....)

[Doctor Gast]

3 hours ago, TwinsDr2021 said:

Maybe the solution to this problem is 8 starters that go 4 innings,

Smelter/Gray

Ryan/Bundy

Archer/Winder

Ober/Jax

That caps all their innings at 160 if perfectly healthy and go 4 innings each time. Then you have 5 short guys to rotate though. Duran, Thielbar, Pagan, Smith Duffey for example, and if need be the bus system to St. Paul.

Nobody would ever have to face a batter the super scary 3rd time, Seems to completely take injuries out of play by limiting pitches and innings, seems like the perfect solution plus you never have to spend big money on a starting pitcher because no starting pitcher worth anything would agree to that. So you have piles and piles of money to spend on offense/defensive. (I mean I hate everything about it, but....)

Plus we could get Alcala and maybe Maeda back to help out the pen. That's not including any closer or frontline SP that we could pick up.

[Rosterman]

I'm also seeing waaaay too many short starts in the minors.

Yes, we are in the process or re-stretching out pitchers after a season lost and whatever last year was called.

If you are a starting pitcher, you used to be setup for endurnace, as well as the ability to use ALL of your pitches in a variety of ways so you can get thru the order multiple times and hit that 100-110 pitch mark which seems to be the ultiamte high bar these days, no matter how good you are pitching. That is the key to being a rotation arm. 

And, unless you have a truly elite bullpen, basically eight closers with a couple of those guys HAPPY to throw multiple innings, a game starts to be reduced to cannon fodder.