:format(webp)/cdn.vox-cdn.com/uploads/chorus_image/image/63655275/1134345285.jpg.0.jpg)
Some discussion as to whether Archie Bradley should be considered a high-leverage reliever. His ERA is fine at 2.00, but I noticed that this was not reflected in his Win Probability, which sits in negative territory at -17% this season. That perhaps is a better reflection of how fans view players: it’s all very well to put up zeroes in blowouts, but we remember more often what you do when the game is on the line, for better or worse. So I thought it might be interesting to plot everyone who has pitched in the Arizona bullpen this season (except for John Ryan Murphy!), and see how well (or not) their Win Probability matches up with their ERA. So, here’s a table, in descending order of WP.
D-backs relievers though Apr 21
| Name | IP | ERA | WPA | SD | MD |
|---|---|---|---|---|---|
| Name | IP | ERA | WPA | SD | MD |
| Greg Holland | 7.0 | 0.00 | 64% | 3 | 0 |
| Andrew Chafin | 7.2 | 3.52 | 49% | 2 | 1 |
| Matt Andriese | 11.2 | 4.63 | 28% | 2 | 3 |
| Yoan Lopez | 8.1 | 1.08 | 12% | 3 | 2 |
| Jon Duplantier | 4.0 | 0.00 | 5% | 0 | 0 |
| Taylor Clarke | 3.0 | 0.00 | 2% | 0 | 0 |
| Matt Koch | 13.1 | 11.48 | -1% | 0 | 0 |
| Archie Bradley | 9.0 | 2.00 | -17% | 4 | 3 |
| Yoshihisa Hirano | 8.0 | 5.63 | -78% | 2 | 3 |
The last two columns are “Meltdowns” and “Shutdowns.” These are measures of whether a relief performance was good or bad. If your game resulted in a WP of equal or worse than -6%, it’s a meltdown; equal or better than +6% is a shutdown. It’s a bit more context neutral than saves. There are some interesting differences here. Bradley appears much worse (by WP) than his ERA would suggest. Conversely, Matt Andriese has been much better. I note he has the same number of shutdowns AND meltdowns as Yoshihisa Hirano - but an overall much better WP. Not all shut/meltdowns are the same, it appears. Anyway, just something to keep an eye on, as the seasons progresses.