Randomly Add A Few More Wins Before Seasons’ End

In some leagues, certainly not all but some, when a fantasy manager runs out of starts, they can still collect stats from relievers. Some ESPN leagues are set up this way. They have a limit on starts, but no limit on innings pitched. The rule is put in place so that managers cannot stream a pitcher every single day of the season, effectively juicing their strikeout and, possibly, win counting stats. This strategy would almost certainly ruin ratios and should not be utilized. I would be curious to see a simulation of how a fantasy team’s ERA/WHIP would pan out if they streamed an average starter every day of the season, but that’s for another time.

Today, I’m writing this article to simply show 10 relievers who are not closers and have recorded the most wins this season. For those who have reached a starter max but can still record reliever innings, pick up these pitchers now! For those who just need a few more wins, start utilizing (if you haven’t already, why?!) the Roster Resource Closer Depth Chart and stream these pitchers in favorable matchups or after a day of rest. Here are the non-closer relievers who lead MLB in wins this season:

Predicting reliever wins is a difficult task. In fact, Steamer (ROS) is over it. If we look at all relievers (0 IP min) from 2021 and all relievers (0 IP min) from 2022, the correlation coefficient between their win totals is 0.42, good for a 0.17 R-squared. I’m also including pitchers who recorded zero wins in both years, which inflates the R-squared slightly, but still seems informative.

The fact that we see very few, if any, relievers in the top right corner of the graph shows just how unrepeatable high win totals can be for relievers across multiple years. However, in season, out of all relievers who collected at least one win by mid-July in 2021, 57% of them collected a second win in the remainder of the year. Change that threshold to at least two wins by mid-July, and 67% of those pitchers collected at least one win in the back half of the season. Here’s a breakdown:

What can we make of this? Relievers who perform well enough to be awarded a win in the first half of the season continue to get chances in the second half of the season. This is probably still very random. If a reliever is used in the sixth, seventh, or eighth inning often, they are increasing their opportunities to snag wins. But even the correlation coefficient between first half of the season wins and second half of the season wins in 2021 was still only 0.38 (0.14 R-squared). Reliever wins are mostly random, just like the official scorer’s awarding of the win can sometimes be.

Take for example Diego Castillo’s most recent win which occurred on July 16th. First off, that was a long time ago and if you’re just looking at the table of relievers above and adding them to your roster, you’re not doing enough. Regardless, Castillo threw in a tied (2-2) bottom of the ninth inning against the Rangers. He walked a batter, forced a double-play ball, and then struck out the third batter he faced. The Mariners then scored in the top of the 10th and Matthew Festa shut the door to get the save, giving Castillo the win. However, Paul Sewald got three quick outs in the bottom of a tied (2-2) eighth inning and was awarded nothing! He didn’t even walk a batter as Castillo did. Castillo got the win, simply because his team scored in the half-inning after he pitched. Random.

If you’re here at FanGraphs reading about baseball, chances are you have heard about the randomness of the win. Some of you may have even dropped wins from your fantasy leagues’ roto categories. However, it remains a common category and I would bet that the vast majority of fantasy leagues still count the win. If that’s the case, and you could use a few more before the season ends, you should bank on the relievers who up until this point in the season have been recording wins. Just hope that you’re rostering them on the random day that they record another.