Theo's Thoughts

The Starter vs. Reliever Dilemma

In Chapter 7 of Tom Tango’s The Book: Playing the Percentages in Baseball, there is an interesting study done on how pitchers of different roles (starter, swingman, reliever, etc.) succeed compared to how they would in a different role. Tango and his co-authors use 1999-2002 data, which cannot be directly applied to today’s constantly changing game. In this week’s blog, I perform a very similar study, but with 2017-2019 data (I did not include 2020, as it was a shortened season, nor did I include 2021, as it is not completed yet).

First, it’s incredibly important to know the baseball statistic Weighted On-Base Average, or wOBA. In simplest terms, wOBA is an all-encompassing offensive statistic that is set to the same scale as On-Base Percentage. An average wOBA is around .320, and any hitter with a wOBA above that is above average, and vice versa. So, the best pitchers have the lowest opponent wOBA.

Data was collected from FanGraphs, and split into two categories: opponent wOBA as a relief pitcher (wOBA relief) and opponent wOBA as a starting pitcher (wOBA start). To qualify, a pitcher needed to face 200 or more batters as a reliever or as a starter in the three-year span. This eliminated the statistics of position players pitching, which would surely skew the data if included.

Pitchers in relief collected a wOBA of .311, while pitchers in a starting role sported a wOBA of .329, a colossal difference considering the size of this dataset. So why was pitching in relief so much easier? Tom Tango says it perfectly:

“Starters have a lot going against them, as compared to relievers. They pitch a lot longer, forcing them to pace themselves. They face the same batters multiple times in a game, giving the batters an advantage. Relievers can come in and put all their efforts into a very short stint” (Tango, The Book).

So now that we know that relievers have an easier time than starters, let’s get into the nitty gritty by placing pitchers into five different categories:

  1. The Starter/Emergency Reliever: 95-100% of his innings are in starts
  2. The Starter/Spot Reliever: 75-94% of his innings are in starts
  3. Swingman: 50-74% of his innings are in starts
  4. Reliever/Spot Starter: 15-49% of his innings are in starts
  5. Reliever/Emergency Starter: 0-14% of his innings are in starts

There’s a lot going on in this table. Let’s break it down.

Role Percent, the first column, states the average amount of innings in a starting role: the Starter/Emergency Reliever pitched 99% of the time in games he started. Role Class, the second column, was discussed above. N, the third column, is the number of pitchers in the specific role class. TBF, the fourth column, is the total batters faced collectively by all pitchers in the role class.

Now, what does the table tell us?

Swingmen, Reliever/Spot Starters, and Reliever/Emergency Starters did significantly better in relief than in starts, as shown by the huge drop in wOBA in the rightmost column.

And although the Starter/Spot Relievers and Starter/Emergency Relievers did better in relief than in starts, they did worse than expected. See, as mentioned earlier, the difference between wOBA start and wOBA relief leaguewide is 0.18 (.329 – .311). But these two classes had differences less than 0.18, indicating that they are more effective as starters. Interestingly, the 1999-2002 Starter/Emergency Reliever class performed even worse than relievers in the 2017-2019 class: the data studied in The Book shows that the difference between wOBA start and wOBA relief leaguewide was -0.12, indicating that twenty years ago, pitchers who almost always started were much less accustomed to relieving, and that pitchers today are more flexible at shifting roles.

Finally, I wanted to look at three different pitchers who had the greatest individual differences between wOBA start and wOBA relief: Seth Lugo, Ryne Stanek, and Wilmer Font. These three pitchers faced 200+ batters in both relief and in starts, which guarantees that luck/chance was not involved.

I added one column to this table, Difference, which is wOBA start – wOBA relief. Seth Lugo, pitcher for the New York Mets, was absolutely elite in relief between 2017 and 2019. His wOBA relief was the third best in the league among qualifiers, behind only Brandon Morrow and Roberto Osuna. And it’s not like his performance was a fluke either: he faced 634 batters in relief. His wOBA start was a whole different story; not only was it not elite, but it was below league-average. And the Mets caught on—he hasn’t started one game this year.

Ryne Stanek and Wilmer Font were the opposite of Lugo. Their wOBA relief was way below league-average, especially Font, with a dismal .377. But as starters, these two were completely different pitchers. Now, there is a slight caveat to these numbers: both pitchers played for the Tampa Bay Rays at some point between 2017 and 2019, and the Rays use an “opener” strategy, where the opener will face the batters in a lineup once (by pitching one to three innings) before another pitcher comes in and pitches between three and five innings. So Stanek and Font weren’t really starters (they were openers), but their opener statistics fall into the wOBA start category because they did start the game. So the Astros, who have Stanek, should maybe try the opener strategy with Stanek starting. Font, on the other hand, has been used only as a starting pitcher in Korea (KBO) this year, and has found great success.

All in all, looking at opponent wOBA when starting vs. relieving can be useful for teams in making decisions on what part a player should play in the team’s pitching strategy. Some pitchers fare much better in the starting role, while others find greater success in relief.

I hope you found this post informative, and I thank you for reading. Please let me know in the comments of your opinions on this week’s topic, or any suggestions/requests for my next article!


3 thoughts on “The Starter vs. Reliever Dilemma

  1. If I were a GM I’d just stack my pitching staff with starters. Great article, thanks Theo Tobel!


    Posted by tribefan | August 11, 2021, 12:16 am
  2. A topic I knew nothing about, Theo. Excellent presentation! xo


    Posted by Saundra Tobman | August 11, 2021, 1:08 am
  3. Great job, as always, Theo!


    Posted by Brad | August 11, 2021, 4:20 pm

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

Blog Stats


Enter your email address to follow this blog and receive notifications of new posts by email.

%d bloggers like this: