Mailbag: Pitching Questions

What is wrong with Harang this year? Image by Getty Images via Daylife

Had a few questioins arrive in the inbox. Questions may be paraphrased, and since I didn't ask permission to post them, they are anonymous (feel free to chime in if you wish).

Reader #1:

What is your take on Aaron Harang? Has he collapsed, or is he just unlucky?

Note: most of what follows is from Harang's player pages at FanGraphs.

Comparing the 2008-2009 Harang to the 2006-2007 Harang, there are two substantial differences I see. First is a decline in his strikeout rate from the mid-8 k/9's to the mid-7 k/9's. The second thing I see is a decline in ground ball rate from 39% or so to a fairly Milton-like 35%. All the while, though, his walk rate has stayed pretty constant--maybe a slight uptick, depending on how you measure it. But nevertheless, he still has a very fine k/bb rate, even if not as good as it used to be.

I recently read something about his slider's vertical movement having vanished. There is a pretty steep decline in slider vertical movement between 2007 and 2008/2009, which corresponds to a big drop in his slider's run value. Looking at this graph, it seems as though it happened about three starts before his DL stint. So there absolutely could be something to that, and it could explain why his strikeout rates have dropped a fair bit of late. But then again, there's an increase in horizontal movement in his change-up over the same time, without much change in its run value.

The various DIPSish metrics are mixed on him. His FIP is up from 3.7 to 4.4 or so, which is a steep climb. But his xFIP, which is a better predictor of future performance, is up only slightly, from 3.8 to 4.1 or so. So that's encouraging.

His tRA has gone from 3.75ish to 5.1 or so (yikes). But his tRA*, which includes appropriate regression and thus is a better predictor than tRA of future performance (FIP:xFIP::tRA:tRA*), has "only" increased from 4.1 to 4.9 or so. Again, not as bad as the unregressed data, though clearly not the level of performance we saw from him during 2006-2007.

He's essentially performing right at his ZiPS projection this year, which I would have said was pessimistic. Not surprisingly, his ZiPS the rest of the year is unchanged. That's the only projection system I have access to that gives in-season updates, but I think all the other projection systems (which had Harang in the 4.0-4.2 ERA range) will be closer to 4.35 ERA or so.

So I think the message here is that, at this point, Harang is still a quality starter, and can be expected to post something close to a league-average ERA (which is still a bit above-average for a starter). I'd put his expected WAR at around 2.5 to 3 by season's end, and probably about 2.5 next season. That's a quality pitcher. Not the 5 WAR beast he used to be, but still a good pitcher.

He's being paid $11M this year (expected 2.5 WAR) and $12.5M (~2.5-2.7 WAR, depending on the market) next year, which means he's pretty much being paid appropriately. No problem there either.

Reader response: Something you haven't mentioned is that Harang's DER is 2nd worst in the majors. I'd like to see some study on the affect that has on a pitcher's overall numbers. The reason is because if there is a pitcher that is getting less help from his defense, will his other numbers naturally decline simply because of the extra effort it takes to get through an inning. Put another way, a pitcher who basically has to get 3.5 outs an inning might be more likely to give up home runs or other hits than one that only has to get 3 outs an inning.

I didn't directly address the DER issue, though I did notice it. My feeling is that this year has to be luck-induced, as opposed to fielding induced. It's his worst DER of his career, but at least as far as Reds teams go, this has been the best defense he's ever had playing behind him. Granted, maybe there were some screw-ups while he was on the mound gven the small sample, but it makes no sense that his DER would get worse as his team's fielding undergoes a massive transformation unless it's largely a timing/"luck" issue.

Reading this again, I sort of missed the point as far as the psychology issue is concerned. My feeling is that you're right about it, but I don't know what the effect size is.

The next question comes from a different reader:

I am wondering if there is any way you could address this question/issue on your blog or maybe on BTB if you think it is interesting enough. You probably know about the Reds starters' struggles in the 1st inning of games this season. The stats, closely approximated about as near as I can tell are the following:

Bronson Arroyo: 9.46 ERA, 1.080 OPS
Aaron Harang: 7.36 ERA, .870 OPS (OPS worse after today's start against the Padres)
Johnny Cueto: 9.00 ERA (slightly worse), 1.150 OPS
Micah Owings: 7.50 ERA, .860 OPS
Edinson Volquez: 7.00 ERA, .913 OPS

Also, Harang, Cueto, and Arroyo are 3 of the top 5 worst pitchers in the league in BAA the first tie through the order, and the Reds have by far the worst 1st inning run differential in all of MLB. So my question is, do you have any idea or insight into why this might be? Any explanation for it at all? This phenomenon is very perplexing to me, as is the fact that apparently no adjustments are being made to correct it.

I honestly have no clue. Most teams do worse in the first inning than any other inning, simply because that's the only inning where you are guaranteed to see at least one (and often more) of the other team's best hitters. But why the Reds might be especially bad during the first inning isn't something I have an explanation for, unless they're not warming up enough or something (very unlikely, I'd think). Especially because the Reds rotation, while not awesome, isn't bad either.

It's not like the 8th-inning problem the Reds had a few years back, where the problem was that the Reds didn't have anyone competent enough to get the ball to Weathers in the 9th. There was a clear explanation for what we saw that year.

My honest guess is that it's noise. But I can't think of a clear way to test that further that would provide any sort of clear answer. Maybe some of the pitchf/x'ers can look at pitch motions and velocities in different innings and identify an issue. But that's not in my bag of tricks, at least not yet.

Sorry, that's the best I can do.