Third Time's a Harm?

[Matthew Lenz]

“Will you teach Rocco how to manage a pitching staff”, said [name redacted]. Not being in the room at the time I knew what he meant. Rocco had pulled Dobnak before he could face the Tigers lineup the third time through. What he didn’t know was it was history shows it was probably the right call.[Name redacted] went on to say “... let him start the 6th and if he gives up a hit, then pull him ... that’s my strategy on MLB the show.” So if you ever find yourself playing MNSportsguy21 online in MLB The Show, you better believe he’s sending his guy out there for a third time through. So who’s right? MNSportsguy21 or Rocco Baldelli? Is there even a right answer? We’re about to find out.

 

Let's start by looking at the historical statistics of a starting pitcher facing a lineup for a third time in a game and then narrow it down to focusing on Rocco Baldelli use of the strategy. For this exercise, I wanted to limit my research from 1988 to current day as the National Baseball Hall of Fame (Cooperstown) refers to this era as “Today’s Game”. Unfortunately, Fangraphs only goes as far back as 2002 for this type of statistic so the last 18 seasons will be my timeframe. First, lets just look at a simple line graph that shows this growing trend across Major League Baseball.

Download attachment: Screen Shot 2020-09-04 at 9.40.00 PM.png

You can see that from 2002 to 2014 MLB managers stayed pretty consistently around 7,500 innings pitched but since then that trend has decreased to 5,408.1 innings pitched in 2019. A pretty drastic and clear shift in strategy amongst a majority of Major League Baseball teams ... not just Rocco. This strategy should make sense as most pitchers throw more pitches they start to wear down and as teams see them for a third time they know what to expect. That said, such an extreme shift doesn’t happen just based off of assumptions so let's look at another set of data.

 

Download attachment: Screen Shot 2020-09-04 at 9.55.44 PM.png

If you couldn’t see a theme by looking at the numbers I used a color scale to help. Our assumptions were true, teams have more success against starting pitchers the more they face them in a single game.

 

Now let's start to narrow the focus to Rocco Baldelli development of this strategy before taking a look at which, if any, Minnesota Twins starters should be given a longer leash. Coincidentally, Rocco’s first year coaching with Tampa Bay was 2015, which was the year that trend started, and was also the year they ranked 28th in starters innings pitched against the lineup the third time through. From 2015 to 2018, before he joined the Twins, the Rays starters were dead last in innings pitched in the same statistic. You may be thinking they just had a poor pitching staff, but in the same timeframe they were 12th in SP fWAR. If Rocco believes in this strategy, it’s clear that he adopted it from his time with the Tampa Bay Rays.

 

So how has this impacted the starting pitchers on the Twins? Interestingly, the Twins have actually allowed their starters to face the line up the third time through the 10th most in all of baseball. Contrary to what I suggested before, I do think this has more to do with our starters earning a longer leash through their play rather than Rocco ditching the strategy. So let's see how the 2020 Twins starters have fared the third time through a lineup throughout their careers. In the graphic below the two stats are wOBA / FIP in that order and the number in parenthesis is the innings pitched. For reference, since 2015 the median innings pitched in this scenario is 116.2

Download attachment: Screen Shot 2020-09-04 at 10.44.51 PM.png

Obviously, Dobnak’s terrific numbers have to be taken with a grain of salt when he’s pitched a total of 5 2/3 innings in his short career in this scenario. As with any data set there are exceptions to the rules here, as Berrios actually has better numbers the third time through than the first, but we can mostly summarize that turning the game over to the bullpen is likely the best option in most scenarios for this group of Twins starters. I was actually surprised that the Twins had the 10th most innings pitched in this scenario as it seems like Rocco pulls the starter more often than not.

 

MNSportsguy21, you might want to change your MLB The Show strategy.

 

MORE FROM TWINS DAILY

— Latest Twins coverage from our writers

— Recent Twins discussion in our forums

— Follow Twins Daily via Twitter, Facebook or email

 

Click here to view the article

[jorgenswest]

I think you missed pointing out how the data for the third time through is very skewed towards the top of the line up. A pitcher rarely sees the bottom of the line up a third time so the vast majority of the data from the third time through is skewed significantly toward the best hitters.

 

A better look would be to see how batters fare against a pitcher their third time they see him. Given the data you outline above you would expect a majority of hitters to have their best at bats the third time they see a pitcher. If the data is similar to looks two and three years back I am guessing that it won’t be near as striking. I think it was around 37-38% of batters fared better the third time through which is more than 1/3 but not much more.

 

 

 

[Jimv]

 

I think you missed pointing out how the data for the third time through is very skewed towards the top of the line up. A pitcher rarely sees the bottom of the line up a third time so the vast majority of the data from the third time through is skewed significantly toward the best hitters.

 

A better look would be to see how batters fare against a pitcher their third time they see him. Given the data you outline above you would expect a majority of hitters to have their best at bats the third time they see a pitcher. If the data is similar to looks two and three years back I am guessing that it won’t be near as striking. I think it was around 37-38% of batters fared better the third time through which is more than 1/3 but not much more.

I think your points are good and would love to see data broken down that well, but nonetheless, I thought the data in the article was illuminating and appreciated the analysis.

[Trov]

The third trip through is something that has been highlighted over the recent years.  To me, it really comes down to the pitcher, I have not looked up the numbers and there may not be much for them, but I recall Radke and Johan both being guys that if they made it through the first two inning without giving up runs they normally would do pretty well the rest of the game, my memory could be wrong though.  

 

I think starters that have a good mix of pitches and know how to set up hitters for future at bats can pitch deeper.  I am wondering how the pitch count has affected this third time numbers thing.  What I mean by it, with pitchers knowing 100 pitches is basically the cut off now a days, and 2 trips through order, do they not worry about that 3rd trip and not worry about setting up hitter for a third time, or saving some stuff for that 3rd time?  Also, as the first comment pointed out it is more likely the pitcher is facing the better hitters the 3rd time and not getting to end of line up, so the numbers may be a little skewed. 

 

What I mean by setting up hitters for a third time.  Some pitchers will go with same approach against hitter every at bat because it worked, but against a good hitter he will adjust to what pitcher is doing.  A good pitcher can think ahead and think about if the hitter will make that adjustment and bet one step ahead.  That is the chess game that is baseball.