Notes on fantasy sports, data, and the decisions between.
Bo Nix dropped back to pass more often than any quarterback in football, and the season around it looked like an arrival: just under 4,000 yards, 25 touchdowns, fourteen wins, the top seed in the conference, a second-year quarterback whose team finished a game short of the Super Bowl. On a per-dropback basis he ranked fourteenth of thirty-two qualified passers. Both pictures are his, and the…
Jameson Williams played nine of every ten Detroit snaps in 2025, every game, the whole season. He was thrown to six times a game. The offense wanted him on the field for almost everything it ran and fed him like a complementary piece once the play began, a strange way to use a receiver who, on the throws he did get, produced more value per target than all but six players in the WR/TE field.
JJ Bleday is the hundred-and-ninety-second-ranked player in fantasy baseball, and by the quality of his contact he is hitting like a top-ten-percent bat. Both numbers are current, both describe the same player, and the distance between them is not an accident. A ranking is sold as a measure of production. What it often measures is reputation: where a player was drafted, which team signs his…
Jordan Love finished the 2025 season second among qualified quarterbacks in expected points added per dropback, behind only Drake Maye and ahead of Matthew Stafford, Brock Purdy, Dak Prescott, Josh Allen, and Patrick Mahomes. In the same months, the market that prices quarterbacks moved him the other way, down toward a back-end starter, the kind of name you wait on late rather than reach for. Two…
In 2021 and 2022, Josh Allen averaged six hundred pass attempts a season and Buffalo threw over its situational expectation more than all but two teams in football. In 2025 Allen threw 460 times, the offense ran the ball on 49 percent of its plays, and the rushing title belonged to his running back. Then in January the Bills fired a 12-5 head coach and gave the franchise to the coordinator who…
Marvin Harrison Jr. was the most certain receiver prospect in a decade. He went fourth in the 2024 draft and first in most dynasty rookie drafts, the son of a Hall of Famer, scouted as a finished product. In two NFL seasons he has separated from coverage less than almost any qualified receiver in football. As a rookie, he separated less than all of them.
On Wednesday at Petco Park, Cristopher Sánchez threw seven scoreless innings against the Padres. Six hits. Nine strikeouts. The Phillies won 3-0. The seven innings pushed his scoreless streak to 44 2/3 innings, past the 41-inning Phillies record Grover Alexander set in 1911.
Drake Maye finished 2025 at +9.14 CPOE. Caleb Williams finished at −6.87. Same draft class, same season of measurement, opposite ends of the same chart. The chart is the thing that knows whether either number means what people think it means.
Mike Trout is hitting .238. His barrel rate is the highest of his career.
Caleb Williams took twenty-four sacks in 2025, down from sixty-eight as a rookie. He threw to a rookie tight end whose receiving EPA was +49.0 and a rookie wide receiver whose was +30.8. Ben Johnson installed a play-action offense around him and the Bears won eleven games and the NFC North. Every condition for a year-two breakout was present. His completion percentage above expectation finished…
Drake Maye held the ball 2.97 seconds before throwing in 2025. He threw it 9.1 yards downfield on average. He led the NFL in completion percentage above expectation by four full points. The first two numbers usually rule out the third.
A rookie’s first month of major-league baseball is the one stretch in his career where the contact data is the only signal that matters. There is no prior major-league sample to weigh against it. What he produces is what he produces. By the time he has put a hundred and twenty balls in play, or pitched thirty innings, the contact data has stabilized enough to read. The 2026 class is at that line.…
Justin Wrobleski threw six shutout innings against the Cardinals on May 3. He allowed six hits and one walk. He did not strike out a single batter.
Cam Schlittler throws three fastballs. Two of them sit above 97 mph. The third sits at 94. They are 89 percent of his pitches. He has walked six batters in 41.2 innings, which is one walk every seven innings he has pitched. The league averages one every three.
Through 29 games, Fernando Tatis Jr. has the highest hard-hit rate in baseball. He has not hit a home run.
Justin Jefferson generated more separation in 2025 than he has in three years. His EPA was −13.8. George Pickens has never generated meaningful separation in his career. His EPA was 88.7.
In 2025, a tight end in the bottom quarter of separation at his position averaged 2.78 yards of space at the catch point. In 2020 that number was 2.26.
Pete Fairbanks has issued one walk in eight innings. The ERA is 7.88.
Ke’Bryan Hayes is batting .068. His contact quality says .249. At 48 plate appearances, both numbers are provisional. But the distance between them is not small, and it points in one direction.
Saquon Barkley rushed for 1,140 yards in 2025 and produced the largest year-over-year rush EPA collapse at the position since the current EPA model was implemented in 2020. Both numbers describe the same backfield. They disagree about the season.
Contact quality in Jordan Walker’s first 12 plate appearances can tell you a direction. It cannot tell you a magnitude. The Week One Truthers piece was explicit about the limit: no responsible analyst would draw conclusions from the sample it was drawing conclusions from, the method was the point rather than the prediction, the gap between the box score and the contact data was most visible when…
Chandler Simpson has forty-nine plate appearances, zero barrels, and an average exit velocity of 80.6 miles per hour. He is batting .391.
Most fantasy football rankings for 2026 will be built from 2025 full-season averages, because full-season averages are the easiest number to pull. They also weight the first fourteen weeks at more than three to one against the last four, which is bad math for any season where the meaningful role changes happened in December. Most seasons.
Sandy Alcantara threw a complete game shutout against the White Sox on 93 pitches. Three hits. Zero walks. Seven strikeouts. Ninety-three pitches is barely ten per inning, and the White Sox are on pace to lose a historically meaningful number of games, so the natural instinct is to discount it. But the pitch count is the interesting part. It tells you how the outs were made.
Week One Liars caught pitchers whose Statcast profiles disagreed with their box scores. This is the companion piece, applied to hitters, and the premise is simpler: these are batters where the contact data agrees with what you saw, or says what you saw was the understatement.
One start. That’s the dataset. One start per pitcher, roughly twenty batters faced, and the hot takes have already been filed. This is, by any statistical standard, a completely meaningless sample. What follows is a completely meaningless analysis of that sample. But Statcast doesn’t know it’s Opening Week, and it tracked every batted ball anyway, so we might as well look at what the contact data…
The barrel rate moves first. Before the expected wOBA improves, before the results follow, a hitter starts barreling the ball at a different rate. The contact quality catches up to the barrel rate. The box score catches up to the contact quality. The gap between where the contact quality already is and where the results still sit is where the 2026 edge lives.
A hitter’s batting average tells you what happened. His expected batting average tells you what should have happened, given how hard he hit the ball and at what angle. The gap between the two is where mispricing lives. A positive gap means the bat is better than the stat line. A negative gap means the stat line is flattering the bat. Multiply that logic across three dimensions of contact quality…
Ranked lists sell certainty. This one doesn’t. This is a list of 20 starting pitchers ordered by how good their stuff actually was, not by how good their results looked. The distinction matters because results include sequencing, defense, and luck. Stuff is just stuff. Exit velocity, launch angle, and batted ball quality on every pitch that was put in play, measured by MLB’s Statcast system and…
Every pitcher in baseball has two stat lines. One lives on the box score. The other lives in the quality of contact he allowed, computed from exit velocity and launch angle on every batted ball that left a hitter’s bat. MLB’s Statcast system tracks both. The gap between the two is among the most exploitable signals in fantasy baseball.
On the page, a roster is a list of names. The analytics engine reads it differently. The two versions often disagree on what the team is doing well, and on what to do about it.
The best team in a fantasy league is often the one that read what the numbers were measuring, not the one with the most talent.