Francisco Rodriguez has been in a Major League Baseball uniform for parts of thirteen seasons now, and parts of the last four with the Milwaukee Brewers.
When Ron Roenicke called Rodriguez into a tie ballgame in the top of the ninth last night against the Cincinnati Reds there was an audible groan from the Milwaukee faithful, after all we all KNOW that K-Rod is a bad pitcher in non-save situations.
But is he really? I looked into some numbers to see if K-Rod really is a bad pitcher in non-save situations. For this analysis I didn’t confine a save to the final inning of a game, I confined a save to any outing where K-Rod had to protect a lead of 3-runs or fewer late in the game.
So, tie games, large leads and games where the team trailed all count as non-saves. K-Rod has spent a lot of time as a setup man in his career, and in Milwaukee, and the talk of him being bad when he wasn’t protecting a save-like lead persisted then as well.
For his career, K-Rod has appeared in save situations 496 times and non-save situations 300 times. His career splits in the two areas are pretty even:
- Save Situations: 2.78 ERA, 11.3 K/9, .626 opponent OPS
- Non-Save Situations: 2.67 ERA, 10.2 K/9, .613 opponent OPS
So, for his career it isn’t the case that K-Rod is bad in non-save situations but most of Frankie’s career hasn’t been in Milwaukee and those first couple seasons where he was very good for the Angels he was behind Troy Percival on the depth charts at closer.
Rodriguez has been Milwaukee’s closer in 2014, and for most of 2013 before being traded to Baltimore. He got some chances in 2012 and 2011, but he wasn’t the team’s primary closer in either of those seasons.
In 2014, K-Rod has appeared in 46 save situations, and 20 non-save situations. 20 is a pretty small sample, but it is enough to look at some numbers:
- Save Situations: 2.60 ERA, 10.2 K/9, .650 opponent OPS
- Non-Save Situations: 4.50 ERA, 9.5 K/9, .708 opponent OPS
So at least we know that our eyes aren’t lying to us. K-Rod strikes out fewer hitters in non-save situations. His batting average against jumps from .188 to .244, and his on-base percentage against jumps from .244 to .298. His slugging against is virtually identical.
So yes, in 2014 K-Rod has been pretty lousy in situations where he doesn’t have a lead to protect, but what about in recent memory. Here is a chart of K-Rod’s ERA when protecting a save-like lead, versus not.K-Rod’s ERA in Sv vs Non-Sv Situations
2011 includes both his time with the Mets and with the Brewers, 2013 only includes his time with the Brewers not the Orioles.
I realize that ERA is pretty one-dimensional when looking at a player this way, but I think it serves its purpose here.
What this chart shows, more than anything, to me is the variability in relief pitcher performances from season to season. Relief pitchers are generally the hardest players to project from season to season, and while Frankie isn’t an elite closer anymore, his general performances have varied wildly over the last six seasons.
Why would Ron Roenicke put K-Rod into a tie game in the top of the ninth if he’s been so bad in non-save situations this season? Because that is how you manage “by the book.” It has been demonstrated over and over the last few seasons that Ron Roenicke has a sometimes slavish devotion to doing “what you’re supposed to do.”
The conventional baseball wisdom for ages had been to bring your closer into a tie game in the ninth inning at home. Why? Because he’s supposed to stop the other team from scoring and then you win it in the bottom of the ninth, which is exactly what happened last night.
Frankie has been bad this season when he didn’t have a lead to protect, but 20 games and 86 plate appearances is a pretty small sample to work from. You can see from the chart, that most of the time he’s actually better in situations when he doesn’t have a lead to protect.
It is probably a myth we’ve all concocted and perpetuated that K-Rod is bad in non-save situations, but, I still don’t want to watch him pitch them, at least not for the remainder of 2014.
I realize I’ve just scratched the surface here, we could go into all sort of fun areas including better stats like FIP and xFIP, Pitch F/X data and heat maps of K-Rods location, but I don’t think they are necessary today.