ERA and ERA+ Explained

By 4Baseball Staff · November 5, 2026 · 9 min read

How earned run average is calculated, why a 3.00 ERA means different things in different years and parks, and how ERA+ rescales a pitcher's ERA against the league so eras can be compared.

Earned run average is the most quoted pitching statistic that is not a won-lost record. It answers a simple question: how many runs does a pitcher give up per nine innings, not counting the runs that came from his defense's mistakes? A pitcher who allows 3.00 earned runs per nine innings has a 3.00 ERA, and the lower the number, the better.

The trouble starts when you try to compare two ERAs. A 3.00 in 1968 and a 3.00 in 2000 are not the same achievement, because the average National League pitcher in 1968 allowed 2.99 earned runs per nine and the average American League pitcher in 2000 allowed 4.91. A 3.00 ERA was ordinary in one season and excellent in the other. Park matters too: the same pitcher posts a different ERA in Denver than in San Francisco.

ERA+ is the fix. It is the pitcher's ERA expressed against the league, scaled so that 100 is average and higher is better, and usually adjusted for the park. This article explains how both numbers are calculated, works through examples with real figures from the historical record, and lays out what each statistic misses.

The formula and a worked example

The calculation is one line:

ERA = 9 × earned runs ÷ innings pitched

Take a starter who throws 180 and one-third innings and allows 65 earned runs. Nine times 65 is 585, and 585 divided by 180.33 is 3.24. His ERA is 3.24.

Innings are the one wrinkle. Baseball writes a partial inning as .1 for one out and .2 for two outs, so 180.1 means 180 and one-third innings, not 180.1 as a decimal. Before dividing, convert to thirds: 180.1 is 541 outs recorded, divided by three. The historical record stores the figure as outs for exactly this reason.

An ERA needs enough innings behind it to mean anything. To qualify for the ERA title, a pitcher must throw at least one inning per scheduled team game, which is 162 innings in a 162-game season. A reliever who throws 60 innings can post a 1.80 ERA, but a few bad outings would move it a long way, because a single three-run inning adds 0.45 to his ERA. Over 200 innings, the same inning adds less than 0.14.

What counts as an earned run

The word that matters is earned. A run is earned if it scores without the help of a fielding error or a passed ball. The official scorer replays each inning as though the errors had not happened, and any runs that would still have scored are charged to the pitcher as earned. Runs that would not have scored are unearned and do not count toward his ERA, though they do count against his team.

Two examples show how this works. A batter reaches on a throwing error and scores two batters later on a double: the run is unearned, because he would have been out if the shortstop had made the play. A batter walks, advances on a wild pitch and scores on a single: the run is earned, because the wild pitch is the pitcher's own mistake.

A pitcher is also charged with the runners he leaves on base. If a starter leaves with two runners on and the reliever gives up a home run, the two inherited runs are charged to the starter and the batter's own run to the reliever. This is why the earned runs charged to a reliever depend on how often he enters with runners on, and why the starter's ERA can get worse after he leaves the game.

A related statistic, runs allowed per nine innings (RA9), counts every run, earned or not. The difference between the two is largest for teams with poor defenses. In the historical team table, each club has both runs allowed (RA) and earned runs allowed (ER) for every season, so you can see how far apart they were.

Scoring context: the league sets the scale

An ERA only means something against the league around it. The table below uses the National League and American League ERAs computed from the team totals in the historical record.

A 3.50 ERA in 1930 would have been far better than the league's 4.97. The same 3.50 in 1968 would have been worse than the league's 2.99. Raw ERAs rank pitchers by how few runs they allowed. They do not tell you who beat the hitters around them by more.

The shift from 1968 to 1969 is the cleanest demonstration. Runs per team game across both leagues in 1968 were 3.42, the lowest figure since 1908 in the historical data. The next year the mound was lowered and the strike zone was made smaller, and the National League ERA jumped from 2.99 to 3.60, about 20 percent, in a single season. The pitchers did not get worse. The conditions changed. The 1969 reforms are covered in their own article.

Park effects: the same pitcher in two ballparks

The second distortion is the ballpark. The historical team table carries a park factor for every club-season on a scale where 100 is neutral and higher means more runs. In 2000 Coors Field was rated 125 and the San Francisco park was rated 93, a difference of about a third in expected run scoring. In 2025 Coors Field was 112 for batters and 115 for pitchers, while Oracle Park was 94 on both measures.

A rough conversion makes the point. A pitcher with a 4.50 ERA at Coors in 2000 was pitching in a park that inflated runs by about 25 percent. Dividing by 1.25 gives 3.60, the figure he would have been expected to post in a neutral park. The same 4.50 in San Francisco, divided by 0.93, is 4.84. The two pitchers have the same ERA and quite different performances.

Real adjustments are more careful than that. A pitcher works about half his games at home and half on the road, so the full park factor overstates the effect on him. Sources also differ in how they compute the factor and how many seasons they average. If you see two ERA+ figures for the same pitcher on different sites, a different park method is usually why.

The effect is visible in a single team's runs. In 2025 the Rockies allowed 1,021 runs in 162 games, which is 6.3 per game. That is a very bad pitching staff, and it was also a staff pitching half its games in one of the highest run environments in the sport. Both facts matter, and a statistic that cannot separate them is incomplete.

ERA+: the league-adjusted version

ERA+ puts the pitcher's ERA on a scale where the league average is 100. The idea is a ratio:

ERA+ = 100 × (league ERA ÷ pitcher's ERA), adjusted for park

If a pitcher has an ERA half the league's, his ERA+ is 200. If his ERA matches the league's, it is 100. If it is 25 percent higher, it is 80. It is a percentage of the league average with the direction flipped, so that higher is better. A pitcher with an ERA+ of 130 allowed about 23 percent fewer earned runs per nine innings than the average pitcher in his league and park.

The table below shows the ratio part of the calculation, before any park adjustment, for several famous seasons. League ERAs are from the historical team totals, so published ERA+ figures that adjust for park or exclude the pitcher's own runs will differ somewhat.

Two lessons come out of the table. First, raw ERAs mislead across eras. Gibson's 1.12 looks better than Martinez's 1.74, but Martinez pitched against an American League that averaged 4.91, and the ratio puts him ahead, before any adjustment for Fenway Park, which the historical record rates at 104, slightly friendly to hitters. Second, the last row shows why park matters. Freeland pitched at Coors Field, which in 2018 had a park factor of about 116 or 117. A fully adjusted number would raise his 141 into the 150s or 160s, depending on how the factor is applied. A 2.85 ERA there was a much better season than a 2.85 in a neutral park.

How to read an ERA+

As a rough scale, 100 is average. A starter at 110 to 120 is a good pitcher, 130 to 140 is excellent, and a season at 150 or higher is a star performance. Holding 120 or more across thousands of career innings is rare. These ranges are rules of thumb, not official categories.

Three habits keep the number honest. Look at the innings: an ERA+ of 200 over 70 innings is a hot streak, and over 270 it is a Cy Young season. Look at the role: relievers often post high ERA+ marks because they pitch in short bursts and are often given favorable matchups. And look at the league: ERA+ is relative to the pitchers around him, so a pitcher in a weak league and a pitcher in a strong league are not equivalent.

Seaver's career is a useful reference. He pitched 4,782 and two-thirds innings with a 2.86 ERA, and his best seasons, such as 1971, sit near 200 on the ratio above. The profile of Tom Seaver puts a full career against the league around him.

What ERA and ERA+ miss

The main limitations are the same for both statistics.

• Defense. A pitcher who allows a hard-hit ball that a good fielder turns into an out has a better ERA than one whose fielders miss it. ERA gives the pitcher the credit or the blame for both.
• Sequencing and luck. Three singles in a row score a run, and the same three singles scattered across three innings do not. Over a season, some of that evens out, but over a month it does not.
• Scorer decisions. The earned-run call is a judgment by the official scorer, and the way errors are scored has changed over time.
• Relievers. ERA treats a reliever's inherited runners one way and the starter's another, and a closer who pitches one inning at a time with a clean slate has a built-in advantage.
• Volume. A 2.50 ERA over 100 innings is not the same contribution as a 2.50 ERA over 220.

Because of these limits, analysts developed measures that try to isolate what the pitcher controls. FIP, xFIP and SIERA strip out defense and luck and look at strikeouts, walks and home runs. WHIP counts the runners a pitcher allows without caring when they score. And WAR combines a pitcher's innings and quality into one season number. None replaces ERA. All are useful when ERA and the eye test disagree.

Common questions

Is a lower ERA always better? Yes within a given league and park, but the size of the difference depends on the scoring environment. A one-run gap in ERA is worth more in a low-scoring year than in a high-scoring one.

What is the best single-season ERA? The historical data includes dead-ball and 19th-century seasons where ERAs under 1.00 appear in short schedules, such as Tim Keefe's 0.86 in 105 innings in 1880. Modern seasons with a full qualifying workload are a different matter, and the table above gives the idea.

Why do relievers have lower ERAs than starters? They face fewer batters per outing, often with the platoon advantage, and pitch with a rested arm. The best comparison for a reliever is another reliever.

Does ERA+ account for the pitcher's defense? No. It adjusts for league and park, not for the fielders behind him. FIP-style measures try to do that.

What is a good ERA today? Judge it against the league. The National League averaged 4.22 in 2025 and the American League 4.09, so a 3.50 ERA that year was well above average, while the same figure in 1968 would have been worse than the league's 2.99.

Where to go next

To see the formulas in use, read the FIP, xFIP and SIERA article for what a pitcher controls, WHIP explained for the baserunner side, and the walk and strikeout rates article for the components that drive both. You can compute your own numbers with the stat calculator or look up any pitcher on the leaders page.