Has Barry Zito been really bad or just really unlucky?

Here at Statistics 101 headquarters, we love statistics. That probably does not come as a surprise, but this week I want to introduce yet another statistic to the repertoire of my readers. Component ERA, usually abbreviated ERC, was developed by statistical guru Bill James as a better representation of a pitcher’s performance than the standard ERA. ERC is useful for evaluating a pitcher’s performance as the sum of his statistical indicators (hits, walks, home runs, innings) instead of just the runs he allowed. In other words, it is another measure of what a pitcher’s ERA “should” be, instead of what it is. Over time, a pitcher’s ERA and ERC should be very similar, so a pitcher with a better ERA than ERC will usually see his ERA decline as he accrues more innings.

The formula for ERC is quite complicated, but I will list the most commonly used variety here for the masochists in the audience (trust me, it’s much easier to find the numbers online than it is to calculate it by hand).

In spite of all of the complicated calculations involved, many pitchers have posted very similar ERA and ERC numbers. The San Diego Padres’ Chris Young has an ERC of 1.79 and an ERA of 1.82, and Jake Peavy has an ERA of 2.27 and ERC of 2.23. Other pitchers have also come close to matching their numbers: Dan Haren (2.48 ERC and 2.46 ERA), Kelvim Escobar (2.84 and 2.79), and C.C. Sabathia (3.55 and 3.56).

As always, there are pitchers who are exceeding their ERC (and thus due for a drop-off) and pitchers who are not living up to their ERC (and thus likely to improve). Of all pitchers, the Padres’ Kevin Cameron has the largest differential (in percentage terms): his ERA is an impressive 0.90 and his ERA is “only” a respectable 2.67. The Diamondbacks’ Doug Davis has been the luckiest among starting pitchers, sporting an ERA of 3.88 despite an ERC of 5.00. Chuck James (3.84 vs. 4.76), Zach Duke (5.79 vs. 6.94 – yes, Duke should have been even worse than he has been), and Livan Hernandez (4.83 vs. 5.82) have also been exceedingly lucky this year.

The unluckiest pitchers this year, those you should target if you still have the opportunity to do so, include massive disappointments Dave Bush (5.03 ERA and 4.63 ERC) and Barry Zito (5.08 ERA, 4.53 ERC). If their owners are ready to write them off, you might be able to take advantage. Of course, analysts have been saying that about these two all season, so you might also want to look at Roy Halladay (4.13 vs. 3.55), Greg Maddux (4.20 vs. 3.64) or Anthony Reyes (5.60 ERA, 3.98 ERC).

Of course, if you have been reading my columns for a while, you are probably already thinking of the caveat that I am about to issue regarding ERC. I would call ERC a “first derivative” statistic – it is a much more useful number than ERA because it is based on more than just runs, but the numbers used in ERC are still somewhat susceptible to factors beyond a pitcher’s control. Hits and home runs can be broken down further into “second derivative” statistics (note that I am using quotes because these are not derivatives in purely mathematical terms, but I want to indicate that the numbers are derived from other numbers).

As you probably remember, hits are really a result of a pitcher’s strikeout rate and his defense, but a pitcher’s BABIP (opponents’ batting average on balls in play) will be relatively consistent with the league averages over time. Similarly, a pitcher’s home run total is a result of his fly ball rate and his HR/FB% - the rate at which fly balls turn into home runs. These two considerations mean that a pitcher who is due for a correction in his hit rate or home run rate will still post a strong ERC, because the components of his ERC (hits and home runs) will be more positive than the indicators behind them.  Despite this, ERC is still a much better indicator of how well a pitcher is faring (and should be expected to pitch in the future), and should always be used in lieu of ERA whenever possible.

Questions and comments may be sent to John at johnfranco@sportsgrumblings.com