When it was revealed that Josh Rupe would be coming North with the Orioles to start the season, I speculated on Twitter that it was due to his Spring Training performance (0.00 ERA*), given that he has a career ERA/FIP/xFIP around 5 in the majors… as a reliever (plus an awful 1.2 K/BB ratio). A good point was made by Luke Jackson that the team was going with only 4 starters initially, so Rupe wasn’t likely to be up long. MASN’s Steve Melewski** replied that Buck Showalter and the team had used the “eye test” with Rupe, and liked what they saw – and that’s why he had made the team.
* Edit: Not anymore apparently, as Rupe have up a couple runs in the O’s last ST game.
** Steve and I argue occasionally, but he’s a good sport about it and I appreciate that.
Here’s the thing about that; it seems like baloney to me. Rupe has a sparkling ERA, but in 13.2 he’s struck out all of 4 batters. He hasn’t given up an earned run*, but he has given up a home run. And he’s only allowed 8 hits, which is the kind of luck that won’t continue while everyone’s putting the ball in play against him. So either they’re not using stats at all (ignoring his entire career and his poor ST strike-out numbers in favor of “he looks good out there”), or they’re using the wrong ones, incorrectly (ERA in a small number of innings against uneven competition).
* The earned run / unearned run distinction is kind of silly, but ERA is what FIP/xFIP are scaled to and it’s so common.
Not that it really matters. The difference between Rupe and, say, Mark Hendrickson is more or less negligible over the course of two weeks. I just don’t like the apparent process. The question shouldn’t be “does he look good?” or “how are his ST numbers?”; it’s “using how he looks, his ST numbers, and the much larger amount of other information we have available (career numbers, etc.), how is he likely to perform this season?” And that last group should be weighted significantly more.
Stats: ERA, FIP, xFIP, K/9 & BB/9