Simple Relief Pitcher Valuations

Patrick Gorski-Imagn Images

Every Sunday, in my waiver report I rank lowly rostered closers that might be worth a pickup. I add a note to each closer, in the following format:

Relievers: Saves-based ranks

Grant Taylor: Good reliever who could be the closer.

Andrew Kittredge: Good reliever who seems to be the closer.

Clayton Beeter: Average reliever who is the closer.

I don’t go into details on a pitcher’s talent, but use adjectives like “Good” or “Below-Average” to give readers a sense of how I value view that pitcher’s talent. I’ve winged those values using the season’s results from our Closer Depth Charts.

I was pulling the adjectives/valuations out of the air and other places and decided to both simplify them and put some data behind them. After considering several options, I decided to use our Depth Chart’s relief pitcher rest-of-season projections.

Begin of nerdy math stuff.

Initially, I thought of using several inputs and coming up with some fancy number to put a value on the talent and expecfed results of each reliever. In the end, ERA projections are heavily strikeout and walk-based, covering the most important inputs to pitcher talent, so I just went with ERA.

I started by finding the median and standard deviation of projected ERA for all pitchers from our Depth Charts projections. For this week, the values are a 4.28 ERA with a standard deviation of 0.48. Then I created benchmarks using those values and grouped the results into five categories. It’s not the most precise process, but it is simple, easy to understand, and serves our purpose.

I started by creating the “Average” group – any pitcher with a projected ERA Between the median and 1 SD below the median (3.80 ERA to 4.28 ERA). I know this is not technically “average.” But my ranking focuses on closers and closers are typically a subset of better-than-average relievers. So for this purpose, being better than the median is a good proxy for “average.” The next bucket – from one to two standard deviations below the median (3.31 to 3.80), I called the “Good” group. In all, 60 pitchers made this group, which averages to two per team. Finally, any pitcher whose projected ERA was more than two SD below median goes into the “Great” group. There are 21 pitchers in the Great group.

I also went the other direction. Pitchers between median and one SD above median (4.28 ERA to 4.76) are in the Below-Average group. Finally, any projected ERA over 4.76 is a poor pitcher. Here is a table of the divisions and a copyable spreadsheet with all the values.

End of mathy stuff.

Now, for reference, here are my educated guesses from Sunday and the new values, based on the categories defined above.

I wasn’t off by more than 1 ranking with any pitch. I was too low on three of the pitchers and too high on six of them.

Overall, the framework is in place, and the results seem fine. I plan on providing an updated sheet every week with the Waiver Wire and FAAB report.