Max Fried Makes Doubles Disappear

By the peripheral metrics, Max Fried is nothing special. Over the course of his career, his strikeout and walk rates sit somewhere around the league average. By results, however, he looks like a Hall of Famer; his 2.81 ERA since the start of 2020 ranks third among all starters. When teams contemplate whether to offer the left-hander a five- or six-year deal this winter, one question will be top of mind: Who is the real Max Fried?

Last week, Thomas Harrigan at MLB.com argued that Fried’s contact prevention resume might be the strongest of any contemporary pitcher. Harrigan noted that, after setting a cutoff, nobody has allowed a lower barrel rate this decade. That is reflected in Fried’s home run rate and therefore his FIP — other than Logan Webb, no starter allowed fewer home runs on a rate basis. Thanks to that super low home run rate, Fried posted a 3.11 FIP and 15.4 WAR over the decade — strong numbers, but reflective of a very good pitcher rather than one of the game’s elite.

One thing FIP ignores is the number of doubles a pitcher allows. Fried allowed just 16 doubles in 2024, tied for the lowest mark among all qualified pitchers. That low doubles rate is consistent with his performance over the years — during the 2020s, only 3.5% of the hitters Fried faced reached on doubles, ranking sixth-lowest among 173 pitchers in the sample:

Can Fried claim credit for this skill? For FanGraphs readers and writers, the question is largely academic. For teams, however, the answer carries multi-million dollar implications. How much of that doubles suppression is Fried, and how much of it is the Braves’ defense and ballpark, the hitter he’s facing, the weather on any given day, or any other number of external variables?

These questions are particularly relevant for Fried, who ranks as merely decent by the two components of pitching most ostensibly under his control. If a team offers him a hefty deal, they’re betting that his ability to prevent extra base hits on balls in play is a predictable skill. I believe that bet would be a good one.

Last week, I wrote about Nick Martinez, who leverages his six-pitch arsenal to induce weak contact to both lefties and righties. In that piece, I outlined the pitch-by-pitch theory of the case — sinkers into righties, cutters into lefties — as well as the multiplicative effect of his pitches in concert. (Marek Ramilo of Driveline uses the example of bishops in chess: A bishop alone on the board is worth roughly three pawns, but two active bishops are worth more like nine pawns, or 4.5 pawns each.) But where Martinez sows confusion by producing pitches that look the same out of hand, Fried takes the opposite approach.

The “similarity of pitches out of the hand” effect, measured by the Driveline model Match+, grades Fried as roughly league average. (I wrote about Match+ in the context of both Martinez and Tarik Skubal last month, who both excel by this measure.) But there is another model called Mix+ that Driveline uses to quantify the breadth of an arsenal. Mix+ calculates the pairwise distance between a pitcher’s offerings in three-dimensional space when the faster pitch reaches the plate and then adjusts for the frequency of each pitch:

The theory behind Mix+ is two-fold. First, pitchers who rank higher in Mix+ should receive more takes owing to the spread of potential pitch movement patterns. But the positive effects of a wide arsenal also extend to the quality of contact that a pitcher allows, according to the findings of Marek and Jack Lambert. Because hitters must account for a variety of potential shapes at varying speeds, they’re less likely to unleash an “A” swing:

As you can see, only Chris Sale ranked higher in Mix+ than Fried in 2024. He gets there in large part because of the shape of his curveball and how frequently he uses it. He throws his curveball roughly 20% of the time, with a slight uptick in frequency to right-handed hitters. At -17 inches of induced vertical break, it gets more “drop” than all but five other pitches in all of baseball:

It’s a true two-plane arcing beauty, and Fried uses it to drop in called strikes and get whiffs from unsuspecting hitters. Its 38.1% called-strike-and-whiff percentage in 2024 ranked among the best for any pitch thrown as frequently as Fried threw his curveball:

But it isn’t just the performance of the curveball in a vacuum. The possibility of the curveball in batters’ minds affects the rest of Fried’s pitches, most notably his four-seam fastball, which bores in on right-handed hitters like a lefty version of Corbin Burnes’ cutter. Fried’s fastball and sinker both average 94 mph; his curveball averages 75 mph. The high likelihood of seeing either a hard fastball running in on the hands or a curveball with significant loop moving nearly 20 mph slower leads to uncomfortable at-bats, uncertain swings, and unsightly batted ball quality.

This is evident in the behavior of batters. It’s a pretty tight distribution — the leader and the laggard are separated by just three ticks — but among all pitchers who faced at least 400 hitters in 2024, only Fried’s Mix+ peer Sale induced slower swings. And of course, when a batter swings slower, their contact quality suffers. (These bat speed stats are courtesy of Alex Chamberlain’s Pitch Leaderboard, which just introduced some sweet updates.)

Observing exit velocities and launch angles, measuring bat speed, examining three-dimensional pitch trajectories — these tools were not available to Voros McCracken when he published his seminal work on Defense Independent Pitching Statistics theory, or DIPS theory for short, in the early 2000s.

Many (including me, on this website, last week) misrepresent the fundamental principles of DIPS theory. DIPS does not say that pitchers lack control over batted ball outcomes, as Mike Fast pointed out on Bluesky last week. What DIPS posited was that a pitcher’s ERA depends on a lot of things outside their control, and so the baseball statistics community ought to undertake an effort to measure what a pitcher is responsible for.

Fielding Independent Pitching, or FIP, was the first of these efforts, a creation of the sabermetric giant Tom Tango. It remains influential to this day, and in fact continues to perform well at predicting a pitcher’s next season ERA despite the fact that it was not designed to be predictive. It looks only at home runs, strikeouts, and walks, the events Tango deemed the most in a pitcher’s control.

As novel forms of data came online, from PITCHf/x in 2007 to the debut of Statcast in 2015, new approaches were developed to refine the ideal that DIPS outlined and that FIP put into practice. xERA replaces home run rate with expected damage from exit velocities and launch angles; over at Baseball Prospectus, DRA incorporates a mixed-models approach to account for those external variables like batter quality and ballpark, assigning credit and blame for every given plate appearance.

DIPS theory will never die, because the essential insight will always remain true — the first-generation pitching statistics relied too heavily on the park and the defense behind the pitcher, and there ought to have been a correction to account for this fact. But as Fast wrote, all DIPS claims is that batted ball results are correlated with the DIPS stats like FIP. As we sit here in the waning days of 2024, we can say for sure that there are statistics that are even more strongly correlated with batted ball outcomes. There exists an opportunity to better understand the relationship of the pitcher to batted ball outcomes than ever before, thanks to these new-generation modeling techniques as well as ever-new sources of data — bat speed, arm angle, future biomechanical outputs.

All of these new models and inputs enhance our understanding of what a pitcher can control, but the problem is far from solved. It’s an extremely difficult problem, after all. But there is reason to believe a pitcher like Max Fried can contain doubles at a rate that FIP declines to give him credit for; DRA might not even fully credit Fried for his contributions. These metrics are designed to apply to all pitchers, but Fried is a true outlier. Mix+ provides a compelling theory for why that might be the case.