The offseason is the time for the best descriptive stats — how a player actually performed in the prior season beyond rote fantasy numbers. These stats are the foundation for our projections and rankings, and they are far more important for pitchers than hitters because we generally agree on hitters.
Sure, there is minor variance with hitters. There is quibbling. But there is industry-wide convergence in hitting rankings (which is why we have to pay more for hitters).
Conversely, there is a divergence in pitcher rankings. Some of us are looking at traits, some are looking at the conventional fantasy stats, and others are looking at the more advanced stats that put a fine point on how the pitcher should have fared the previous season. For a generation, we used FIP ERA for the latter, which adjusts all balls in play to the league average rate of hits. I never liked that because it didn’t factor in trajectory and contact quality. And xFIP is even worse because it normalizes home run rate. Maybe hitters control some of that, but certainly not all of it, especially given fly ball-prone pitchers.
But now we have elite tools due to Statcast radar technology that puts everything in one easily digestible bucket. All the Statcast data for a pitcher, including his walks and strikeouts, is calculated and translated into an expected ERA we can compare to his actual ERA. Which is more accurate in describing how the hurler actually pitched? Ironically, I hate actual ERA as a statistic because it can be so non-descriptive and omits a lot. But xERA includes all data in the perfect ratios and allows for looking at the outliers, meaning the pitchers who were significantly misaligned in expected vs. actual ERA.
As the number of balls in play piles up, the variance between the two stats decreases. During the season, it can be much more radical. But when analyzing at least 350 balls in play for the entire prior season, 60-to-70% of the pitchers have an xERA within 0.50 of their actual ERA (63% this past year). So, we’re generally focusing on the other 37%. And of those, we don’t care about guys who were bad but could have been worse or good but could have been better. We’re looking for a margin that turns a pitcher from a useful asset to a harmful one or vice versa.
Let’s first look at the pitchers who were better in xERA than actual. In other words, they pitched better than their rankings indicate — here’s the 2024 list (min. 350 BIP):
Brandon Pfaadt (0.93 better in xERA; starting pitcher No. 52, overall pick 167): He deserved a 3.78 ERA and was above average in strikeout rate and well above average in walk rate. Unlike hitters, pitchers are always a minor adjustment away from radically transforming. Looking at his pitch data, he could just junk the four-seamer that hitters crushed, though maybe it helps the efficacy of his changeup. Based on xERA and pitch quality (his changeup and sweeper are top offerings), I would project Pfaadt to have an ERA in the 3.25-to-3.75 range.
Bailey Ober (0.76, 24, 81): His xERA was 3.22. There’s so much red ink on his Statcast page. I don’t know where to begin. He should be starting pitcher 15-20, not 25-to-30 — that’s a massive discount. He was 10th in xERA last year, between Logan Gilbert and Corbin Burnes (ahead of Burnes).
Joe Ryan (0.73 33, 103): Maybe the Twins are just a terrible defensive team, given they have two starters on the list? Nope, they were almost exactly average. The radar technology says defense is a very marginal consideration in our pitcher projections. The range is about 6% from top to bottom. In other words, the difference between the best and worst teams in defense is about a quarter of a run. For about 80% of the teams, defense should be ignored. Ryan’s xERA was an eighth-best 2.87, and he’s currently the 33rd starter off the draft board. Sign me up at that price.
Nick Pivetta (0.63, 71, 240): OK, we didn’t know where he’d be pitching for most of this early ADP data. But this is a total giveaway price. Now that he’s with the San Diego Padres, for shockingly little money, the cost may increase somewhat. I’m still buying. His xERA was 3.51 (21st), ahead of Jack Flaherty and Michael King. He was 88th percentile in K% and 80th in walk rate. That’s a solid foundation for elite WHIP. Moving from Boston will help his stats given the hitter’s haven that is Fenway Park.
Yusei Kikuchi (0.35 47, 149): His xERA was only 0.35 better for the year, but this price is ridiculous given how he pitched once the Astros tinkered with him after the midseason trade last year. Then his actual ERA was 2.70, with a 0.933 WHIP, 31.8% K rate and 5.9% walk rate. I bet he doesn’t forget what the Houston coaches and analytics people told him. He’s a potential mixed-league SP1 at a SP4 price.
Generally, I advocate taking a glass-half-full look at pitchers. We’re looking for reasons to say, “Why not?” But consider yourself warned on Kevin Gausman (xERA was 82nd at 4.71) and Max Fried (30th at 3.72). There is no way they should be drafted SP43 and SP21, respectively. Remember, Fried, now with the New York Yankees, is moving to a much tougher pitching park.
Some pitchers are viewed as so overrated that they’re underrated. Ronel Blanco should be SP48 according to xERA (4.00). But he’s SP58 in drafts despite his sub-3.00 ERA in 2024. Was he lucky last year? Sure. But this is an overcorrection. He was 18th in expected batting average allowed. I will definitely take Blanco as my SP5 or SP6, hoping he finishes at an SP4, even assuming normalized luck.
The bottom line is that you can’t just look at xERA and decide whether or not to draft a player. Draft day price is the more paramount consideration. Even “lucky” pitchers can be bargains and vice versa for those who underperformed relative to their expected stats. We’re using this data in the context of market consensus, searching for plausible reasons why the general view on a pitcher likely is wrong.
(Top photo of Bailey Ober: Stephen Maturen / Getty Images)
