One popular stat lying to New York Mets fans about Devin Williams signing

Anthony Licciardi

One popular stat lying to New York Mets fans about Devin Williams signing image

New York Mets fans' reaction to the Devin Williams signing probably correlates with the number of New York Yankees fans in their life.

Williams' 2024 season ended in infamy, surrendering the biggest home run of Pete Alonso's career. His 2025 didn't offer as sharp a pain, but after a poor first impression, Yankees fans never got to wrap their arms around one of the best relievers in baseball.

Williams posted a 4.79 ERA in pinstripes, easily the worst mark of his career and far short of anyone's expectations. And yet, after the Mets signed him to a three-year, $51 million deal -- a clear indictment on the stickiness of ERA -- the stat remains at the center of his discourse.

Devin Williams' ERA is lying to you

It goes without saying that neither Williams nor David Stearns can guarantee a bounce back in 2026. Relievers are inherently volatile, frequently weird, and liable to tumble down a steep age curve. But that's also true for Edwin Díaz, regardless of how willing the fanbase is to reset the closer market.

Just about every piece of Williams' profile suggests that 2025 was a statistical anomaly. Whether it be bad luck, variance, or an extenuating circumstance like pitch-tipping, Williams wasn't himself.

We can start with ERA estimators like xERA, FIP, xFIP, and SIERA. Each of them are more predictive than ERA year-over-year by adjusting for things like defense and home run luck. I'll use 2023 for comparison's sake because an injury limited him to 21.2 innings in 2024, but the numbers tell a similar story.

Data via FanGraphs
 ERAFIPxFIPxERASIERA
20254.792.682.953.112.67
20231.532.662.792.662.93
Career2.452.452.692.542.66

One number here stands out, shaping the narrative and the bad taste in the Yankee faithful's collective mouth. 

This can be explained by Williams' third-percentile left-on-base percentage (55.2%, via Pitcher List). The league average for relievers was 72.4% in 2025. Partly a measure of sequencing luck, Williams is a bit of a victim here, especially while posting his highest BABIP (.299) since 2021. A few more balls found holes, a few more guys crossed home plate, and his ERA subsequently skyrocketed, without the starter-level volume to experience regression to the mean.

MORE: Mets linked to trade rumor for $43 million starting pitcher

Moving forward, Mets fans can lean on Williams' key performance indicators as reasons for optimism. His strikeout rate remains elite, and his K-BB% (arguably more predictive than any ERA estimator!) Ranked 13th among relievers last season. 

Williams still has an outlier changeup that is unhittable in the zone, and his stuff hasn't taken a step back, unlike Díaz. Williams has a track record of success, a bid for bad luck, and is yet to show physical signs of regression. That bodes well for quality innings in Queens, and a decent contract makes the signing even sweeter.

New York didn't break the bank at $17 million per year (fewer when accounting for deferrals), nor did it pay with years like it may have to for Díaz. He's open to a set-up role should his predecessor return for the ninth inning, too. 

The markers are there for Mets fans hoping to feel good about the move, and a seemingly robust market suggests that many teams felt the same way. Whether he's pitching the eighth or the ninth, Williams should look much closer to the reliever Stearns remembers from his time in Milwaukee.

More New York Mets news

Contributing Writer