Why FIP and xFIP Beat ERA for Predicting a Pitcher's Future

ERA measures past pitching results, but FIP and xFIP offer better insights into a pitcher's true talent and future performance by removing defensive and luck influences. Understanding the differences between these metrics is crucial for making informed fantasy baseball decisions.

Marco Moretti

October 27, 2026 · 5 min read

If you want to know what a pitcher did, look at his ERA. If you want to know what he'll do, look at his FIP and xFIP. That's the whole game. ERA describes past results, but those results are polluted by defense, luck, and the sequencing of batted balls. FIP and xFIP strip most of that noise out, which is why fantasy owners who use them tend to find the breakouts and the bouncebacks before the rest of their league notices.

Beyond ERA: Introducing FIP and xFIP for Pitcher Evaluation

ERA is earned runs allowed per nine innings, where an earned run is any run that scored without the aid of an error or passed ball. It's the most commonly accepted tool for evaluating pitchers, and as MLB.com's glossary puts it, it "should be an ideal evaluation of pitchers" — the goal of pitching is preventing runs, and ERA tells you how well a pitcher did that. The formula is simple: nine times earned runs, divided by innings pitched.

But the same glossary is blunt about the flaws. Defensive mistakes are accounted for, but great defensive plays are not, so a pitcher with an average defense is at a disadvantage to one with a great defense. Ballparks matter too, because certain stadiums are more conducive to run scoring. ERA tells you what happened; it does a poor job telling you why, or whether it will happen again.

Enter DIPS theory. Voros McCracken's research — Defense Independent Pitching Statistics — is, per FanGraphs' library, "the building block of much of today's pitching analysis." The core insight: pitchers have limited control over what happens to balls put in play. What they do control are strikeouts, walks, hit batsmen, and home runs. FIP measures exactly those outcomes, stripping out the role of defense, luck, and sequencing. xFIP takes it one step further by normalizing home runs against a league-average home run-to-fly-ball rate.

ERA vs. FIP vs. xFIP: A Quick Reference

Here is the side-by-side breakdown. Read each row and ask: which real-world events does this metric hold the pitcher responsible for? That question is the entire difference between the three stats.

What each metric measures, excludes, and is for
Metric	Core Inputs	What It Excludes	Primary Use Case
ERA	Earned runs allowed per nine innings pitched	Runs scoring on errors or passed balls — but still influenced by defensive quality, luck on batted balls, sequencing, park, and league	Describing past results; the standard run-prevention measure
FIP	Strikeouts, walks, hit by pitches, and actual home runs allowed	Defense, luck, and sequencing of batted balls	Assessing true talent level and portending future performance
xFIP	Strikeouts, walks, hit batsmen, and expected home runs based on league-average HR/FB rate applied to the pitcher's fly balls	Everything FIP excludes, plus the randomness of how many fly balls leave the park	Assessing underlying skill with home-run volatility removed

The FIP-to-xFIP distinction deserves a beat of attention, because it's where most explanations go vague. FIP charges a pitcher for the home runs he actually allowed. xFIP charges him for the home runs he should have allowed given his fly ball count, using that season's league-average HR/FB rate. MLB.com's glossary illustrates with Randy Johnson's 2002: a 2.66 FIP against a 2.44 xFIP, the gap explained by his 12.9 percent HR/FB rate versus the league's 10.7 percent.

Why bother? Because HR/FB ratios are noisy. FanGraphs' library notes that a pitcher might allow home runs on 12 percent of his fly balls one year and only 7 percent the next. xFIP is predicated on the notion that pitchers control how many fly balls they allow far more than they control how many of those fly balls clear the fence. So xFIP swaps the actual homer tally for an estimate — and in doing so, removes one more layer of luck from the picture.

Applying FIP and xFIP in Fantasy Baseball

Here's the practical payoff, and I'd argue it's the single most useful habit a fantasy owner can build: compare a pitcher's ERA to his FIP and xFIP, and treat the gap as a diagnostic.

ERA well above FIP/xFIP: the pitcher has been unlucky, poorly defended, or both. His past results undersell his skill. That's a buy-low candidate.
ERA well below FIP/xFIP: the pitcher has been fortunate — balls found gloves, homers stayed in the yard. Expect the ERA to rise. That's a sell-high candidate.
ERA, FIP, and xFIP clustered together: what you see is what you get. The performance is probably genuine.

The logic behind this comes from the metrics' predictive nature. As RotoBaller's Rick Lucks explains, both FIP and xFIP predict future ERA more reliably than ERA itself, which makes them "a good go-to stat to determine if an early breakout may be for real or if a struggling superstar is likely to rebound." All things being equal, a pitcher's ERA is expected to regress toward his current FIP and xFIP over a long season.

One honest caveat, because I refuse to oversell a stat: some pitchers are consistently more or less homer-prone than average, and FanGraphs itself concedes that certain pitchers have shown an ability to consistently post lower ERAs than their FIP suggests. The metrics capture most pitchers well — not all of them. Use the numbers; don't worship them.

To judge whether a FIP is actually good, you need benchmarks. FanGraphs publishes this rating scale, based on the 2016 run environment:

FIP rating scale (FanGraphs)
Rating	FIP
Excellent	3.20
Great	3.50
Above Average	3.80
Average	4.20
Below Average	4.40
Poor	4.70
Awful	5.00

So a 3.30 FIP is a great season hiding behind whatever the ERA says. A 4.60 FIP with a shiny 3.10 ERA is a trade offer waiting to be made — in the other direction.