Happ Falters
J.A. Happ is an interesting entity in terms of peripheral pitching statistics and traditional pitching stats. In 269.2 Major League innings, entering this afternoon’s start against the Milwaukee Brewers, Happ had posted a 3.10 ERA despite a 4.36 FIP, 4.57 xFIP, and 4.65 tERA. This season, since being traded to the Astros, Happ has been up to the same tricks, posting a 3.08 ERA despite a 3.76 FIP, 4.27 xFIP, and a 3.72 tERA. No doubt, Happ has pitched well this season, but there’s no way that he should be a 3.08 ERA pitcher – ace quality – given his stuff or his peripherals.
Happ strikes out a fair amount of batters – between six and seven per nine innings, typically – but walks far too many to sustain that kind of ERA, as he has conceded 3.6 walks per nine innings for his career and 3.9 this season for Houston. What’s sustaining him right now is a low BABIP – .244, particularly remarkable given the fact that the Astros have been terrible defensively this season, with a .681 defensive efficiency rating that ranks 25th in the league. The other helpful factors have been a 6.2% HR/FB and an 81% LOB rate, both far better than the MLB averages.
What happened to Happ today can’t be called regression to the mean, unless his mean is a pitcher with a 9.00+ ERA and far worse peripheral stats. Happ pitched 4.1 innings against the Brewers, allowing two home runs and two walks while striking out five. The Brewers also tacked on five more hits, including three doubles. Overall, Milwaukee scored five runs off Happ, including three out of the seven baserunners he allowed (the HRs don’t count, as they were never on base to be stranded).
The start pushes Happ’s HR/FB all the way up to 9.2%, his BABIP up to .257, his ERA up to 3.63, and his FIP up to roughly 4.10. He’s still well outperforming his peripheral numbers, but the J.A. Happ you see in statistical reports tomorrow is much closer to the J.A. Happ the Astros can expect to see in the future. He’s still a productive pitcher, somewhere near or slightly below the MLB average starter, but nobody should be under any sort of delusion that he’s “great,” can replace Roy Oswalt, or is a top half of the rotation starter.
Jack Moore's work can be seen at VICE Sports and anywhere else you're willing to pay him to write. Buy his e-book.
Happ seems to be the poster boy for outpitching FIP articles, but to me Happ is pretty much an identical pitcher to Jair Jurrjens, who seems to escape the same sort of discussions. The difference being that Happ is a lefty who throws a couple miles slower. But their peripherals are very similar and Jurrjens like Happ greatly outpitched his FIP and XFIP (by an even larger degree).
Jair isn’t escaping anything, his ERA is higher than his FIP/xFIP this season.
Jurrjen usually has a much higher ground ball rate that Happ and a much lower hr rate. Those are pretty significant differences in peripherals. Heavy ground ballers can get away more with low-mediumish k rates.
Happ had pitched in a HR friendly park. Jurrijens pitched in a non HR friendly park.
Happ shittens.
That was the best thing–post, stat, comment, chat answer, whatever–ever on Fangraphs. I sincerely mean that. That was seriously the best thing that ever happened in Fangraphs’ history.
The sample size for Happ isn’t substantial enough to really draw any definitive conclusions. Even less so when you factor in that in part of the season, he was pitching through injury, and part of it he was just coming back from injury. I think because of his age and stuff, people leap to the conclusion that Happ must not have much upside.
I’m not so sure this is true, though. 2009 was a fine season by a rookie pitcher’s standards. Lefthanders, especially big finesse lefties, are often late bloomers. His fastball is a good weapon for him, despite sitting around 90mph; it seems to have a lot of deception. That’s a good start. His secondary pitches are weaker, but have shown flashes.
My point? Don’t be surprised if he shows legitimate improvement to his skills and offsets his regression that way.
I should clarify that I don’t by any means think Happ is going to be an elite, sub-3 ERA starting pitcher. At the same time, I think if he stays healthy, his peripherals should improve at least a little, and his FIP/xFIP in turn. A starting pitcher whose true talent level winds up in the 3.50-3.99 range is still a very useful arm to have around.
last year if you sorted by pitchers with enough innings to qualify, happy was exactly the middle of the pack for FIP, which almost by definition is a #3.
oswalt is obviously better than that but as you point out an average starting pitcher is still pretty good, typically as people go through a litany of pitchers at the back end of their rotation
I always find it amusing that there are some players that analysts are so eager to run down that they race to post a negative article the moment something bad happens. “See?” the article seems to chime. “This proves what I’ve been saying all along!”
I don’t necessarily disagree with Moore’s analysis, but I always find it amusing when one bad outing leads to a fast post like this while the previous seven great ones are ignored.
In those seven games, Happ’s peripherals obviously don’t support the 2.16 ERA and 1.007 WHIP he registered. However, he also struck out over 7 per nine IP, barely walked over three per 9, and kept the ball in the park (0.59 HR/9).
We often accuse the mainstream media of falling into a narrative, and rightfully so. However, we must also be careful that we don’t do the same thing in the other direction. I don’t recall anyone saying that Happ was going to be remotely close to Roy Oswalt in terms of ability, and it’s not fair to imply that this was ever a valid comparison.
I’d be more interested in reading an article examining if Happ was doing anything different in those seven strong starts before today than reading that his one start today proved something that just looking at his FIP/xFIP would have told me anyway.
But this wasn’t even close to being his worst start of the year.
Tonight: -.262 WPA
8/4- -484 WPA
We all know Happ isnt that great but why is tonight’s start the sign of things to come?
I think this article is a prime example of why a lot of intelligent fans have a hard time bracing the sabermetric movement. J.A. Happ is exhibit A of a pitcher who has defied what some metrics have pegged him as. He falters in one start against a good offensive team at home, and I swear this article was up an hour after the game ended. Sorry guys, one bad start doesn’t make up for the fact that Happ has made a mockery of his FIP and xFIP. I know there is an overall vendetta against the Phillies’ organization in sabermetric circles, so Happ being a product of that system, I think it goes hand in hand.
Well said. I am a layman when it comes to sabr stats and it bothers me when a player is seemingly defying the odds because his peripheral stats are inflated. The community cries “small sample” or “luck” and I have to believe that someone like JA Happ is actually a worse pitcher that he seems to be. At the end of the day it’s all about how many runners cross home while he is pitching, not how many batters might have crossed home had he not been so lucky with his LOB and great defense in his small sampling of pitching. This is the same reason I can’t pick Cy Young based on xFIP. I guess I’ll never understand.
I can understand the frustration if “sabr people” abide purely by FIP, xFIP, tERA, etc. and say that these numbers are all that matters. But what bothers me more, is the outcry against the sabr communtiy because when a young pitcher, or any player for that matter, outperforms these types of numbers, “sabr people” say that he is probably not producing results at his true talent level and that he he will regress.
The fact is, sabermetrics exists precisely for these type of players! There will be players who do not regress to what the formulas predict, but the fact is that the vast, vast majority do regress. Happ has certainly produced good results, but the reason these articles are written is that these results are not necessarily definitive of a player’s true ability. We have seen tons of people come up and impress in their results at the beginning of their career, or players that have had seasons where their results suggest a significant change in talent. But the fact is that almost all of them regress towards what the statistics would suggest.
These articles are still necessary for the precise reason that there are people in these comments claiming that since Happ has had these results, they must be real! He has had two seasons of outperforming his peripherals, so he must be a great pitcher! The great fact of sabermetrics is that we do not necessarily buy these results. Yes, if you can pose legitimate, scouting types of reasons why they may not apply, feel free. But most people who complain just look at his ERA and say that those results speak for themselves.
Might Happ prove to be one of the very few pitchers who can consistently outperform his peripherals? Yes. But almost all players who have had results that significantly outperform their peripherals regress towards the mean of their peripherals. The backlash against this article is exactly what we, as rational, stat-minded analysts are trying to prevent.
So when sabr-oriented fans see these sorts of regressions happen, they feel that they need to point it out, because it is these types of games that bring players’ results back to their true talent and disprove that previous results are indicative of true talent.
Will every single player regress fully to the extent our statistics would predict? No. But they almost all do. The fact that Happ hasn’t yet in his short career is not good reason to claim that stats like FIP aren’t what we think they are. They are based on concrete data derived from decades of Major League Baseball. Unless you have truly observed skills that are just not able to be captured by the numbers, don’t claim that the numbers don’t apply to a certain player just because he has outperformed these numbers.
In yesterday’s start against the Brewers he did give up three runs in the first inning before regaining total command in 2nd/3rd/4th innings.
Happ’s success seems to revolve around avoiding the big hits, and the matchup with Milwaukee didn’t favour him, as the Brew Crew are 1st in the NL in home runs, and 3rd in SLG% and 2nd in ISO.
Prior to this start, Happ had pitched very well in MMP, and last night’s start shows that, while the park can favour some pitchers who know how to pitch (especially with Bourn in centre field), sometimes hitting teams come and expose the short left field porch. That is what happened last night, rather than Happ’s peripherals catching up with him.
his BB/9 is probably unsustainable, but if you compare his WHIP to Brett Myers, they are only .20 points apart.
You’re a math major and a senior and you think that regression to the mean is moving toward average? That’s sad.
I am neither a senior nor a math major, so please elaborate on your statement. I have heard the term ‘regression to the mean’ to often have the same definition as reverting to ‘long term average’ or even reverting to mediocrity in some business texts, but I’m not quick enough to understand your slight dig at the author.
Hal – depending on the data points, regression to the mean can be interpreted as regression to the average; it is technically an incorrect usage.
My problem is that your comment is meaningless and does nothing other than belittle someone. Astound us with your intelligence, but add something to the conversation? Please?
Waiting for one poor start from Happ and then publishing this article that says, in essence, “See! We sabermetrics folks always knew he wasn’t that good!” seems premature.
This article has been written many times since Happ has come to the majors, yet it never seems to quite work out that way.
What in the world are you trying to say? Where else has this article been written? Since Happ has been in the majors all of 1 full season, and then parts of a couple others I’m glad we have the single authority on how his career is going to turn out.
Are you actually implying that you expect him to have a career of under 3 ERA, which I guess is the opposite of this article? He’ll be a hall of famer if he can keep that up, please go buy some of his baseball cards now as an investment and let us all know in 10 years if you were correct.
I’m newish to the SABR stuff, so someone please correct me if I have this wrong, but aren’t softer-tossing lefties a category of pitcher than tend to be better than their peripheral stats would suggest? Take Barry Zito, for example. His career ERA is almost a full point lower than his xFIP, and has been so consistently each season of his career.
*Has been consistently lower, I should say, not a full run lower every year.
I will quite simply state you actually produce quite a few outstanding ideas and I will publish a variety of recommendations to add to shortly.
Heya i am for the first time here. I came across this board and I find It truly useful & it helped me out a lot. I hope to present one thing back and aid others like you helped me.
Hey there would you mind sharing which blog platform you’re working with? I’m looking to start my own blog in the near future but I’m having a hard time selecting between BlogEngine/Wordpress/B2evolution and Drupal. The reason I ask is because your design and style seems different then most blogs and I’m looking for something completely unique. P.S My apologies for being off-topic but I had to ask!