A Comparison
Let’s play the old comparison game. I’m going to show you four pitchers, and then you have to determine which one is the best.
Pitcher A: 8 GS, 48 IP, 57 H, 19 BB, 27 K, 11 HR, 6.43 FIP, 6.56 ERA
Pitcher B: 8 GS, 50 IP, 53 H, 20 BB, 25 K, 9 HR, 5.86 FIP, 4.86 ERA
Pitcher C: 8 GS, 55 IP, 51 H, 14 BB, 33 K, 5 HR, 4.10 FIP, 3.07 ERA
Pitcher D: 9 GS, 67 IP, 53 H, 12 BB, 42 K, 6 HR, 3.69 FIP, 1.76 ERA
I know, I know, it’s not very hard, right? Pitcher A is terrible. He probably doesn’t belong in baseball. Pitcher B is nearly as bad, and is a replacement level pitcher at best. Pitcher C is a solid middle of the rotation innings eater. Pitcher D is pretty darn good, though maybe not quite as good as his sparkling ERA would suggest.
What do they have in common? All four pitchers are named Bronson Arroyo.
That’s Arroyo’s 2009 season, broken up into quartiles. His first eight starts of the season, he was an absolute disaster – one of the very worst pitchers in baseball. He improved to just being lousy instead of totally miserable in his next eight starts, but at that point, half the season was gone and his FIP stood at 6.14. That’s a 98 inning sample of below replacement level performance.
His last 16 starts, though? He’s been the exact opposite. His FIP stands at 3.78 since July 10th, as he’s gone back to throwing strikes, missing bats, and keeping the ball in the park. His first half could not have gone any worse – his second half could not have gone any better.
The obvious question is why? I don’t have any idea. I’m not sure Arroyo does either. There’s always a post-hoc explanation for stuff like this, but it’s hardly ever true. The reality is that pitchers are remarkably inconsistent. Arroyo never was as bad as he pitched in the first half, while he’s also not as good as he’s pitched in the second half.
If there’s a lesson to be learned from this, it’s that pitchers are flaky, and even things that a pitcher can control can vary widely over fractions of a season. 100 innings may sound like a lot, but given the extreme variations that pitchers can face, it’s really not.
38
And in the end he was about the same pitcher he’s always been. Breaking this stuff up into quarter seasons or even half seasons is fun, but you’re right Dave, it barely means anything other than showing you how a pitcher got to around what he’s usually like over the course of a season.
This could also be written about other pitchers, and many hitters. Few baseball players are “consistent” over all small fractions of a season.
This question of consistency comes up all the time. I’d love to see two things in this regard:
1) A table which shows a wide range of statistics and the typical variability (Perhaps the 3x Standard Deviation #) at given sample sizes. I think over time we tend to internalize the amount of variance displayed by given stats,
1) An analysis which shows whether consistency is a skill — do some players consistently display greater variability in their performance than others beyond the typical variance we see in given components of performance? Assuming Arroyo is displaying more general variance (in FIP) is that a result of a person characteristic or simply the sum of variances of his general FIP skills?
It’s not exactly what you’re asking in #1 (which I think would be interesting — especially as a graph of each stat for the population of players as a whole, so we could superimpose a given player onto it to actually see how much of an outlier he truly is) but I always turn to “When Samples Become Reliable whenever I’m trying to decide how much weight to give a stat.
There might actually be an explanation in this case. Arroyo recently said that he got a second cortisone shot on July 10th (he had been struggling with carpal tunnel), and that’s about when he turned it around. Also, if you look at Arroyo’s career 1st half/2nd half splits it’s a little scary. He seems to be a much better pitcher in the 2nd half.
I dominated some fantasy leagues by picking him up at the all-star break in a few of them…
Those kind of splits don’t mean much to me, but taking chances on guys like Arroyo and LaRoche who have shown trends for years seems to pay off sometimes.
He stopped playing the guitar altogether, which by his own admission was known to be causing an aggravation of the carpal tunnel he has been suffering from since he came to Cincy (that must be great to hear if you are the Cincy GM). If that isn’t a coincidence it would be the first MLB guitar related injury that I know of.
I blame Sandfrog for derailing Scott Spiezio’s Hall of Fame career.
And how about Bernie Williams? His numbers fell off a cliff when he released his debut album in 2003.
Joel Zumaya had to sit out of 3 ALCS games for playing guitar hero.
Even if we set aside Zumaya’s pseudo-instrument injury, given all the ridiculous “washing my truck” / “carrying luggage on stairs” / “dreaming about spiders” / “ironing a shirt while wearing it” injury explanations (both real and fabricated), I’m almost certain a guitar has figured in one prior to this.
the Reds are not the only beneficiaries if Arroyo has given up his musical pursuits…
The Bronson Arroyo Band – not for you?
Joel Zumaya hurt his wrist playing Guitar Hero, kind of the same right?
And there is speculation after last nights game, including sketchy visual evidence that Arroyo had pine tar on the brim of his cap. He was dominant while Smoltz couldnt get a grip on the balls and blamed the Reds for not properly preparing them.
I think he sprained a finger, but yeah, it was Guitar Hero.
I believe Joel Zumaya suffered an injury playing guitar hero if that counts.
the thing I find most curious about Arroyo is that he does this EVERY YEAR
This year
5.38 ERA/1.48 WHIP in the 1st half
2.34 ERA/1.08 WHIP in the 2nd half
Last year
5.97 ERA/1.66 WHIP in the 1st half
3.47 ERA/1.19 WHIP in the 2nd half
2007
4.84 ERA in the 1st half
3.55 ERA in the 2nd half
for his career his ERA difference between the 1st half of the season and 2nd half is over 1 run.
4.81 1st half / 3.69 2nd half
Is this the sign of a guy who shuns hard work during the off season and gradually “plays himself into shape?”
Ironically, I had looked at his breakdowns this morning after seeing what his ERA was, and knowing he was a demolisher of fantasy squads for the first two months.
Over the last 4 seasons, consisting of about 460 innings, his pre allstar ERA is about 4.8. His post allstar appearances consist of about 410 innings, and a 3.2 ERA. Considering where he pitches, that is an elite ERA.
I have decided that my strategy for next year will be to draft Dan Haren, wait until the end of June and offer Haren for Arroyo and a bat to someone that thinks I have lost my mind.
I wonder if the pine tar on the bill of his hat has anything to do with it…
http://ballhype.com/story/duncan-says-pine-tar-gave-reds-the-upper-hand/
I think it’s Duncan that has pine tar stuck somewhere.
Duncan and LaRussa are pompous asses.
Arroyo should have said that he was borrowing the trick that Duncan taught Julian Tavarez when he was in Stl.
People have mentioned the guitar stuff. Arroyo also received a second cortizone injection in mid-July for the carpel tunnel, apparently, which coincides with his rebound. Whether it’s a causal connection, though, I dunno. I tend to think the guy’s just streaky, as are many pitchers.
-j
Didn’t someone hurt themselves playing guitar hero?
I was at the 4/25 Atl at Cin game in some very nice seats behind home. Arroyo kept throwing a side arm spinner. Every time the ball left his hand you knew it was going to get pounded. And it did. His stats for the day: 5.2 IP, 9 ER (though it should be noted Dickerson and Encarnacion did not help any).
Just looking briefly through his pitch f/x release location charts, it looks like he went from throwing his curve and slider sidearmed and his fastball and cutter overhand to throwing everything a little more from 3/4.
Arroyo has said he lives on his chemical compounds to perfrom. Perhaps his chemist got a better combination going?
Interestingly, almost the entire difference between first half Arroyo (5.45 career RA/9) and second half Arroyo (4.08 RA/9) can be attributed to BABIP (.309 vs. .276) and strand rate (67.2% vs. 74.6%) as his first and second half FIPs only differ by a quarter of a run.
Fuckin’ tremendous things here. I’m very glad to see your article. Thanks a lot and i’m looking forward to contact you. Will you please drop me a e-mail?
Rarely am I satisfied with the common with the online content Someone said today. If you utilize only the material I love to read simply because it makes me think.