ERA (earned run average) is computed as the average earned runs a pitcher gives up per 9 innings, and it is one of the most commonly used indicators of a pitcher’s ability. Being a rate statistic, versus an accumulating statistic like strikeouts or wins, it gives us an idea of how effective pitchers are at preventing earned runs. However, if we delve into the numbers further, we find that there are other important indicators that influence an ERA.
I have written copious amounts about BABIP, the batting average only on balls in the field of play. This is the average that opposing batters hit off the pitcher on all at bats that don’t end in a home run or a strike out. Pitcher’s have virtually no control over BABIP, and therefore high average on balls in play can show that a pitcher is getting unlucky, while low figures indicate some good fortune for the pitcher.
Another indicator is how many home runs a pitcher gives up per each fly ball. This stat also tends to level out for most pitchers, and once again, numbers above the league average (7.7%) indicate bad luck and numbers below good luck.
Then there’s Left On Base Percentage (LOB%). The percent of runners that reach base on a pitcher yet never score. Good strike out pitchers can expect figures between 75-80% while pitchers with low strike out rates and higher OBPs should see figures between 65 – 70%.
The final “luck indicator” is the percentage of total runs a pitcher allows that are earned runs. Some pitchers can actually get saved by unearned runs. Think about this: a pitcher loads up the bases with two outs. He then yields a ground ball to the shortstop, but the shortstop muffs it. An unearned run scores, but it does not hurt the pitcher’s ERA. Fair enough, but the next batter comes up and hits a grand slam. If I’m not mistaken, the pitcher’s ERA remains unchanged because the previous error should have been the end of the inning. I can understand if the three guys on base don’t count toward the ERA, but at least one run should count as earned because the pitcher gave up a homer. The run from this homer was completely stricken from his ERA record. Pitchers who allow more base runners tend to see these situations more often, where runs that should count toward their ERAs are not tallied due to quirks the definition of an earned run. Therefore, I tend to focus more on how many total runs a pitcher relinquishes and then use a common percentage – the major league average is about 92% of all runs are earned – to estimate an ERA.
During the course of a season, any one (or perhaps all) of these indicators can lean toward lucky or unlucky for a given pitcher. I have created a formula that puts all pitchers on a level playing field, almost as if they are playing in a vacuum. Though I understand that this is not physics nor economics class, and we don’t live in vacuums, this is still a better way to measure true ability than a simple ERA.
Here is a short list of some particularly fortunate and unfortunate pitchers this season based on my formula.
Pitcher ERA Projected ERA Difference
Jarrod Washburn 2.93 3.90 +0.97
Edwin Jackson 2.64 3.62 +0.98
Cole Hamels 4.68 3.75 -0.93
Randy Johnson 4.81 3.67 -1.14
These are just some examples of players whose ERAs have been especially influenced by the luck factors. While their current ERAs, or total runs per 9 allowed, are probably a better indication of how much they have actually helped their teams thus far, the projected ERAs are a much improved way of predicting how they might do the rest of the season. As a general manager or fantasy player, one should not be as concerned with the past as with the future. While Edwin Jackson and Jarrod Washburn are still above-average pitchers, even this season they are not phenomenally better (or better at all) than Randy Johnson or Cole Hamels, despite the ERA discrepancy.