Hector Rondon Is Breaking FIP
It requires but a single digit to understand how good Wade Davis has been over the last couple years. Scan the ERA column for this year or last, and the first thing with which you’re greeted is the number zero. You can stop right there. After the zero, the other digits don’t matter. For all intents and purposes, an ERA of 0.34 is the same thing as an ERA of 0.76 is the same thing as an ERA of 0.93. Whatever the number, as long as it starts with a zero, it signifies that the pitcher in question is allowing fewer than one run per nine innings, and it doesn’t get any better than that.
Sometimes, with metrics like Fielding Independent Pitching, it doesn’t even take a digit. It just takes a symbol, and you can stop reading. Like with Hector Rondon, all you have to see is the minus sign. Hector Rondon’s FIP this season is -0.15. That’s negative zero point one five. Rondon has broken the run estimator that is Fielding Independent Pitching.
Of course, even though it feels like we’ve been saying it for a while, it’s still early. And not only is it early, but Rondon is a reliever, and so the sample is even smaller. He’s thrown 13 innings. This means very little! But it is fun! And if the Chicago Cubs had one area that might’ve been viewed as a potential future weakness before the season began, it might’ve been the back end of the bullpen. Rondon, Pedro Strop and Justin Grimm have been great in recent years, but they didn’t have a Kenley Jansen or Craig Kimbrel-type name back there.
Well, nobody’s matched what Rondon’s done thus far. He’s faced 43 batters, struck out 21, walked none, and allowed zero home runs. Only four batters have reached against Rondon — all singles.
The disclaimers: he’s not throwing any harder, he hasn’t added any new pitches, he hasn’t generated any soft contact, and his BABIP is .182. Of course there’s some fluky stuff going on; he’s thrown 13 innings and we’re talking about a guy with a negative FIP.
What we’re seeing here is perhaps some improved command, working more on the edges than around the middle, and perhaps some deception — batters are swinging less than ever, even though he’s throwing more strikes than ever, and he’s currently got baseball’s second-highest called strike rate.
Going to a bigger sample, we can point out that, over the last calendar year, Rondon’s 1.17 ERA is second among all relievers, as is his FIP-based WAR, so it’s not like this is coming out of nowhere. Rondon has truly been one of baseball’s elite relievers as of late, and he probably doesn’t get enough credit for how good he’s been. For now, everything is still too muddy to tell whether Rondon’s made any significant changes since the end of last season, but it’s always fun to point out when a number that’s supposed to be positive is prefaced by a minus sign. So, congrats, Hector Rondon. You’ve broken FIP for the first month of baseball.
August used to cover the Indians for MLB and ohio.com, but now he's here and thinks writing these in the third person is weird. So you can reach me on Twitter @AugustFG_ or e-mail at august.fagerstrom@fangraphs.com.
Didn’t Andrew Miller have a negative FIP until a little while ago? Oh yeah, I forgot, it’s not worthy of a Fangraphs article until the Cubs do it. Even against their weak schedule so far.
Wow you’re salty. Isn’t it more likely that August just didn’t notice Miller’s FIP?
Yeah seriously, I wish more people talked about the Yankees! It’s like they don’t even exist!
JohnfortheGiants is breaking FAP
I wrote about Andrew Miller on the front page on May 3
There is an ‘l’ in Giants. I do appreciate that this commenter has now become parodied to the degree that he/she is, however.
Your troll attempts are getting weaker. Step up your game
OK, I agree that Hector Rondon is an otherwordly pitcher with jaw-dropping, possibly record-breaking peripherals. Well understood. That said, if the goal of Fielding Independent Pitching is to estimate how many runs per nine inning a pitcher would be giving up based on those peripherals, then if the number is coming out negative, doesn’t that really mean that FIP is literally broken and needs to be adjusted to account for performance at this level? After all, there’s no way a pitcher can perform so superlatively that he literally earns a run scored for his team every 60 innings he pitches.
Or is there?
FIP’s not broken. It’s a good fit to ERA over a wide range of typical pitching performances. The factors basically say that each strikeout saves you about 2/9 of a run, a walk costs you about 3/9 of a run, and a homer costs you about 13/9 of a run. But to fit the TTO components to ERA, you have to add a constant that’s based upon the MLB run environment, and for guys with tons of K’s and no HR’s allowed it’s possible for the value to shoot negative.
If you look at the game-level summaries on the player pages, you can see dominant relievers having single games with negative FIP all the time (say if they strike out the side). But over a full season, I can’t remember anyone breaking FIP. Pedro once had a FIP of about 1.39, which is just plain ridiculous.
I think the point being made is that there should be a cutoff. Say, if FIP comes out negative, it should just be adjusted to 0.
I don’t necessarily agree or disagree with this premise. On one hand, it makes the most sense to just round to 0 since you can’t allow a negative amount of runs. On the other hand, it’s just plain fun to see a pitcher so good that he breaks a formula.
I also don’t think a cutoff is good. Models always “break” when at the extremes of what they are meant to model. Having a cutoff of 0 is not useful either. A FIP of -0.15 is better than a FIP of 0. A cutoff would obscure this.
If you’ve hit the extreme edges of a model and it is behaving weirdly, I don’t see that as a big deal. Also even when someone is pitching phenomenally you get a FIP that is only just barely negative anyway.
FIP is a shortcut to the more elaborate and comprehensive DIPS. And even there, it likely breaks:
http://www.baseballthinkfactory.org/mccracken/dipsexpl.html
What you really want is the full BaseRuns version of DIPS. BaseRuns is what behaves properly, and by design, can’t go below 0 runs.
But, there’s a reason that FIP is what is cited and not BaseRuns. When you can get 99% of the way there, with 1% of the effort, that’s what will win the day.
I think the problem is if you cut it off at 0, you could have 2 pitchers with negative FIP, but they’re both just 0. This way, you can actually see just how ridiculous someone has been pitching, I like it.
Well, according to the Fangraphs FIP page:
“a more stable indicator of how a pitcher actually performed over a given period of time”
So, he’s actually gaining runs for his team with his performance, according to FIP. If it’s producing results like this, then as fun as it is to see, it’s not really doing its job.
When they say it’s more stable they’re referring to the fact that FIP has less year to year fluctuation than ErA. It accomplishes that goal just fine. If you see a pitcher’s ERA after 13 innings you take it with a grain of salt. Same idea with FIP.
I’m thinking maybe you started writing this a couple days ago and didn’t get all the stats updated. Rondon has pitched 13 innings now. I noticed because 11 1/3 innings = 33 outs + 4 singles = 37 (or fewer) batters faced, not the 43 cited. The 43 is correct, but in 13 innings = 39 outs + 4 singles = 43 batters faced.
He’s striking out almost 50% of batters with a 11% swinging strike rate? His outside swing percentage has dropped, his zone percentage is way up. For some reason, batters are just watching his stuff go by. This really doesn’t look like it will last.
At least on the broadcasts, Len and JD talk about him adding a change up this year. From your research, has he thrown more? Any?
Great article,
He takes runs away from the other team. Now that’s dominance.
Well, that didn’t take long.