Luck against Morton

By Mike

If you had to describe Pirates starter Charlie Morton, you would probably include words like “embarrassing”, “horrendous”, or “abomination”. Unless you’re Buccos GM Neal Huntington, “unlucky” probably wasn’t one of them, as he related in a recent interview for the Pittsburgh Post-Gazette (http://www.post-gazette.com/pg/10142/1060038-63.stm).

I’ll take a look at Morton’s peripheral stats to figure out what Huntington is talking about and if there is any merit to this. Like last time, I’m pulling my info from www.fangraphs.com and feel free to leave a comment if you need me to go into detail about the philosophy behind them. Don’t be shy, because I’m sure there will be more than a few.

At face value, it doesn’t look like Morton could be pitching any worse. In nine starts this season, he is 1-8 with an ERA of 8.71. He only has 41.1 innings pitched, which means he’s averaging less than five innings each start. He doesn’t have enough innings to qualify for official statistics even though he hasn’t missed a start all season. That’s bad. Out of 126 pitchers who have over 40 IP this year, he’s dead last in ERA. Ryan-Rowland Smith is second worst at 7.29; almost a run and a half lower than Morton. So what is so unlucky about all this? There are three peripherals that are really jumping out that can actually explain most of this.

First is BABIP, which I spent a lot of time on in my last post. Since I did get a few questions on this, I thought I’d take a few sentences to explain this a little more. The formula is (hits-home runs)/(at bats-strikeouts-home runs+sacrifice flies). BABIP works as a pitcher’s stat just like a hitter’s stat, where it should hover around .300 (I did look into the history of BABIP a little more and found that no season in MLB history has had a BABIP fall outside the range of .293 and .303.) The only real difference is that a pitcher’s BABIP will more likely level out around .300, where a hitter’s natural speed or line drive hitting ability could raise their BABIP in the .320s or .330s.

Charlie Morton’s BABIP is an absurd .365. Even if he had a AAA team playing defense behind him, there’s no way that is sustainable and will almost assuredly fall relatively quickly. His high BABIP is correlated with his high left on base percentage (LOB%). Right now, he’s allowing over half of all men on base to score with a LOB% at 49.2 percent. To put that into perspective, the MLB average usually sits in the 71-72 percent range. In 2008, the worst LOB% was a tie between Brian Bannister and Greg Maddux at 64 percent. Last year, the worst LOB% was Carl Pavano at 66.1 percent. This stat is even unluckier than his BABIP and will have to drop sooner than later. 

For the third stat, let’s think back to the basis of BABIP. In the denominator, we’re subtracting out his home runs given up. So far, he’s tied with seven other pitchers at 10 home runs, and only Dan Haren has given up more at 12. So it would be reasonable to conclude that Morton’s low LOB% is due to all the home runs he’s given up, correct? Well, even his home runs have been relatively unlucky. His home run to fly ball ratio (HR/FB) is a lofty 22.2 percent. Statistical analysis of HR/FB concludes that pitchers have very little impact in this ratio (they can only really control GB/FB) and has more to do with variables such as ballpark’s outfield walls (Citizens Bank, Great American Ballpark, green monster at Fenway) and  weather/atmosphere (altitude at Coors Field, wind currents toward right field at Yankee Stadium). MLB average hovers around 10 percent.

In 2008, Braden Looper had the worst HR/FB in 2009 at 15.8 percent (which is high in itself, even at Miller park), with second worst being Rick Porcello at 14 percent. Morton’s 22.2 percent, again, is extremely high, and should regress back to the mean. Morton having a HR/FB is not sustainable, unless he actually tries to give up home runs (think Home Run Derby).

Well now that I’ve gone through these three peripheral stats indicating how unlucky Morton really has been so far this year, we need a way of not only quantifying this, but also a means of predicting what his ERA should be. We can achieve this by utilizing his expected fielding independent pitching (xFIP). A correlation study has been performed on all pitching statistics, standard and advanced, and xFIP has the strongest correlation to future performance. Meaning, xFIP is the most accurate statistic in predicting future pitcher’s performance. xFIP is a pretty complex formula that includes HBP and BB, as well as giving a value for the number of fly balls a pitcher gives up. Then, a constant is multiplied to normalize the discrepancy for a high or low HR/FB ratio. The result is the pitcher’s xFIP on an ERA scale.

(Note: If this normalization step is taken out, you get the basic FIP, which is used more for assessing current performance and not predicting future results).

Finally, we arrive at Charlie Morton’s xFIP of 4.25, which would put him at league average. Other pitchers with a xFIP around 4.25 are C.J. Wilson, Josh Beckett, Clayton Kershaw and Johan Santana. Morton’s xFIP is also better than that of Zach Duke (4.41) and Paul Maholm (4.68). 

Now, what will the Pirates do with Charlie if he still blows up? He won’t make it through June if he still gets demolished, but he needs to have more starts at the major league level to figure out if his statistics will regress toward the mean or if Charlie just doesn’t have what it takes. His stuff is filthy when he’s on, with sharp curveball and a fastball that averages 92.8 mph, good for 18^th best in the league, just a couple tenths below Tommy Hanson and David Price. 

When he’s on, Charlie Morton is a fun pitcher to watch. I’m hoping he can figure it out, because he’s the only pitcher on the current staff that could be a major part of the future. That is a BIG if though, since he hasn’t left himself for much more room for error.

…and if you’re able to follow any of what I just wrote, I’ll be extremely impressed.

The Dirty Heads - Lay Me Down