Earned run average

Origins

Henry Chadwick is credited with devising the statistic, which caught on as a measure of pitching effectiveness after relief pitching came into vogue in the 1900s. Prior to 1900 – and, in fact, for many years afterward – pitchers were routinely expected to pitch a complete game, and their win-loss record was considered sufficient in determining their effectiveness.

After pitchers like James Otis Crandall and Charley Hall made names for themselves as relief specialists, gauging a pitcher's effectiveness became more difficult using the traditional method of tabulating wins and losses. Some criterion was needed to capture the apportionment of earned-run responsibility for a pitcher in games that saw contributions from other pitchers for the same team. Since pitchers have primary responsibility to put opposing batters out, they must assume responsibility when a batter they do not retire at the plate moves to base, and eventually reaches home, scoring a run. A pitcher is assessed an earned run for each run scored by a batter (or that batter's pinch-runner) who reaches base while batting against that pitcher. The National League first tabulated official earned run average statistics in 1912 (the outcome was called "Heydler's statistic" for a while, after then-NL secretary John Heydler), and the American League later accepted this standard and began compiling ERA statistics.

Recently written baseball encyclopedias display ERAs for earlier years, but these were computed retroactively. Negro League pitchers are often rated by RA, or total runs allowed, since the statistics available for Negro League games did not always distinguish between earned and unearned runs.

ERA in different decades and baseball eras

As with batting average, the definition of a "good" ERA varies from year to year. During the dead-ball era of the 1900s and 1910s, an ERA below 2.00 (two earned runs allowed per nine innings) was considered good. In the late 1920s and through the 1930s, when conditions of the game changed in a way that strongly favored hitters, a good ERA was below 4.00; only the highest caliber pitchers, for example Dazzy Vance or Lefty Grove, would consistently post an ERA under 3.00 during these years. In the 1960s, sub-2.00 ERAs returned, as other influences such as ballparks with different dimensions were introduced. Today, an ERA under 4.00 is again considered good.

The all-time single-season record for the lowest ERA, according to www.mlb.com, the official website of Major League Baseball, is held by Dutch Leonard, who in 1914 had an earned run average of 0.96, pitching 224.2 innings with a win-loss record of 19-5. The all-time record for the lowest single season earned run average by a pitcher pitching 300 or more innings is 1.12, set by Bob Gibson in 1968. The record for the lowest career earned run average is 1.82, held by Ed Walsh, who played from 1904 through 1917.

Some researchers dissent from the official Major League Baseball record and claim that the pitcher with the all-time lowest earned run average is Tim Keefe, who had an earned run average of 0.86 in 1880 while appearing in 12 of his team's 83 games and pitching 105 innings (with a win-loss record of 6-6). But a purported record based on so few innings pitched is highly misleading. Over the years, more than a dozen part-time pitchers have pitched 105 or more innings and had an earned run average lower than 0.86. Major League Baseball recognizes many records from the 19th century — including Will White's 1879 record of 680 innings pitched, Charles Radbourne's 1884 record of 59 wins, and Pud Galvin's 1883 record for 75 games started, but does not recognize Keefe as the pitcher having the all-time lowest single season earned run average.

Infinite and undefined

Some sources may list players with infinite ERAs. This can happen if a pitcher allows one or more earned runs without retiring a batter (usually in a single appearance). Additionally, an undefined ERA occasionally occurs at the beginning of a baseball season. It is sometimes incorrectly displayed as zero or as the lowest ranking ERA, even though it is more akin to the highest.

Sabermetric treatment of ERA

In modern baseball, sabermetrics uses several defense independent pitching statistics including a Defense-Independent ERA in an attempt to measure a pitcher's ability regardless of factors outside his control. Further, because of the dependence of ERA on factors over which a pitcher has little control, forecasting future ERAs on the basis of the past ERAs of a given pitcher is not very reliable and can be improved if analysts rely on other performance indicators such as strike out rates and walk rates. For example, this is the premise of Nate Silver's forecasts of ERAs using his PECOTA system. Silver also developed a "quick" earned run average (QuikERA or QERA) to calculate an ERA from peripheral statistics including strikeouts, walks, and groundball percentage. Unlike peripheral ERA or PERA, it does not take into account park effects. Another statistic derived from ERA is Adjusted ERA, also called ERA+, which adjusts a pitcher's ERA to a scale where 100 is average for the league and takes into account the various dimensions and other factors of each ballpark.

Career leaders in the live-ball era (post-1920)

Because of rules changes post-1920, most notably the abolition of the spitball and frequent replacement of soiled or scuffed baseballs, the increased importance of the home run (largely due to Babe Ruth), and the American League's adoption of the designated hitter rule, ERAs have been noticeably higher than in the early decades of the sport.

This is a list of the lowest ERAs among pitchers that played their entire careers after 1920 (minimum 1,000 innings pitched). Note that three of the top six (Clayton Kershaw, Whitey Ford and Sandy Koufax) were primarily starting pitchers. Hoyt Wilhelm was a reliever for most of his career, while the other two are closers.