Andrew Miller may have been superhuman in the 2016 playoffs, but his heroics wouldn’t be enough to save the bullpen this season.
To say that the Indians’ bullpen has struggled this season would be an understatement. They have gone from having the best bullpen last year to having the second worst in the league thus far. In order for the Indians to compete with the Red Sox, Yankees, and Astros to try to make it to their second World Series in three seasons, the Indians were forced to move one of their top prospects, Francisco Mejia, to the Padres for top reliever, Brad Hand and righty, submarine pitcher, Adam Cimber.
Losing Andrew Miller, one of the best relievers in the game, has undoubtedly hurt the Indians’ bullpen. However, as a whole, relievers can only have so much impact due to limited playing time. In order to determine how much impact losing Miller has had on the Indians for the first half of the season, I needed to project how many games he would have appeared in and how many innings he would have pitched had he not gotten hurt.
Since Miller joined the Indians in the middle of 2016 (not including playoffs), Miller has appeared in about 40% of the Indians’ games and has pitched a little over one inning per game played. Through the first half of the season, he would have appeared in about 20 more games had he not gotten hurt and taken 21 innings away from the other tribe relievers. The contribution of innings pitched for each reliever would look as follows:
Using Miller’s stats through his first 100 games with the Indians before his injury as a guide, he would have allowed approximately 4.4 earned runs in those 21 innings for a 1.87 earned run average. That is much better than the 5.28 ERA the Indians were sporting through the All-Star break (equivalent to 12.3 earned runs in 21 innings). This means that if Andrew Miller did not go on the DL this season, he would have saved the Indians about eight earned runs. Using the Pythagorean Win Expectation formula and the Indians runs scored (487) and runs allowed (405), the Indians would only have been expected to win one more game had those eight runs been saved (57 vs 56).
The following are the Indians’ ERA, WHIP, and K/9 without Andrew Miller and with the Andrew Miller projection. As you can see, the bullpen would be improved, but still towards the bottom of the league:
Just for a fun addition to my experiment, let’s say that Miller performed at playoff level usage for the portion of the season that he has missed. Miller famously became a workhorse in the 2016 playoffs, appearing in 10 of the Indians 15 games and pitching nearly two innings per appearance. The Indians reliever contributions and statistics would look like this:
In this scenario, the tribe bullpen ERA would go down 80 points, while the strikeouts per 9 innings would shoot up by 2.5 strikeouts. If Tito decided to use Andrew Miller in the regular season like he does in the playoffs, he would save the Indians a whopping 23 runs in nearly 60 additional innings of work. Wouldn’t that be nice if Miller had a robot arm and could pull that off? In the meantime, the additions of Brad Hand and Adam Cimber are sure to give a late-inning boost, and once Andrew Miller comes back, we can hope the Indians will once again boast one of the most feared pens in the game.