Pitching by the Numbers: Home run rates

I talked mostly backstage with Yahoo’s Scott Pianowski last week about homers and especially Tim Lincecum, a player I am very reluctant to back on a personal level but who my models, this year, are forcing me into recommending to some degree.

Scott says that homers are not random events and the pitcher has to take the hit for it. I agree to a point. I don’t like xFIP, for example, because it normalizes homer rate. I can’t go all the way in on that stat and stipulate that hitters totally control homers, even though I know that hitters mostly control most outcomes (let’s call it 60/40). So I am with Scott in saying that we cannot entirely absolve pitchers of their homer sins.

But this early, some homer rates are absurd and perhaps should be steeply discounted, attributed to bad luck/running into the wrong hitters. I am not saying that home run rates take so long to normalize (in other words, you need to face so many hitters) that any in-season rate is completely random/luck. For a number of pitchers, we can reasonably project a homer rate good or bad relative to league average.

Since 2012, this average is 0.99 homers/9. Let’s call “good” at 0.8 or under and “bad” at 1.2 or over. Is Lincecum going to be bad at allowing homers by this standard? I will bet, “yes.” But right now his rate is nearly twice that. So those six homers should be three. At the 2012-2014 average of 1.55 runs per homer, that’s about five extra earned runs allowed. Back those out (while STILL projecting that Lincecum is going to be bad at allowing homers going forward) and you get a 4.20 ERA. So if you can stomach an ERA in the 4.00-to-4.50 range, he’s worth rostering for the Ks.

Similar adjustments can be made to CC Sabathia. We’d knock two homers and about three runs off his ledger lowering his ERA to the 3.50-to-4.00 range. Do you want a 3.75-ERA Sabathia going forward if he continues to fan well over a batter per inning? I sure do, in all formats.

 There just aren’t many pitchers who are always bad at homers. Since 2012, here are the ERA qualifiers who have given up an average of 1.2 homers/9 every year (again league average is about 1.0):



That’s it. Just one guy. Here are the pitchers with two bad-homer years in the period, including this year, of course:



So you can reasonably think those guys are homer prone. But what’s the upper limit? What’s the most homers/9 we should expect from a pitcher? Understand there’s a selection bias here because if you’re giving up an incredible amount of homers, you probably aren’t going to qualify for the ERA title by pitching 162 innings. So let’s lower the search from 162 innings to 100 and see what pops.

Really bad is about 1.5 homers/9. Since 2011, minimum 100 innings, there have been 24 seasons out of 423 that beat that and the highest is Tommy Hunter’s 2.15 in 133.2 innings in 2012. Less than a quarter had HR/9 of 1.20 or worse/higher.

So the high homer guys with good peripherals are buy low possibilities, especially in the early season when the rates are more screwy due to smaller samples (presumably).

Here’s a list of the most homer-prone guys in the top 40 in (Strikeouts minus Walks)/IP. If their rate of homers allowed going forward declines to the 1.2 range (still bad/25th percentile), these ERAs should decline significantly.



Forget Greinke, who is the king of the solo homer. He's not even getting unlucky in ERA with all those dingers. Let’s recalculate Bailey, who has the worst HR/9 in baseball. His career HR/9 is 1.1. He should have allowed three homers by now given the career rates. Let’s call it four, meaning we subtract 4.7 runs that we’ll round to five. That would make his ERA now 4.50, which still doesn’t make any sense. The .400-plus BABIP explains the rest. Yeah, buy Bailey, even 80 cents on the preseason dollar will end up returning a tidy profit.

I don’t want to keep going back to the well with Salazar, whose lack of control/command can be troubling. But I do still believe. And Straily is Bailey-like (Bailey-light?) and definitely cheaper. His HR/9 for his career is 1.4, but last year it was 0.9. Knock three of his seven homers off and his ERA drops to 3.50-3.75. I’d expect that from Straily going forward IF (and this is a big if) his K/BB profile stays pretty much the same. I have no idea how Colon does it basically with one pitch. I can’t bet on him maintaining a playable, mixed-league K-rate going forward. But I do see his ERA declining dramatically. He can be the poor man’s Julio Teheran, at least the version of Teheran we’ve seen since April.

But what about the guys who we can reasonably bet to have very low (0.80 or lower) rates of HR/9.

Here are pitchers who are good in HR/9 every year since 2012 including this year. Andrew Cashner did not do it in relief in 2012, but I threw him on here just looking at his starting seasons (because I like him, I admit).



I have to say something about Burnett, a preseason favorite. He’s not a number's guy anymore. He’s in the medical department now. I have no idea how he will progress with his hernia/groin. He seems to be dealing with it but that may mean nothing in two weeks. All I know about Burnett is that if he is healthy, he’s very likely to be very good.