Game 6 is tonight. The Issue #56 Carolina historical comp went from 75% to 31% to 75% again over eight days. A Bayesian prior that encodes the structure of a situation does not disappear when things get uncomfortable. It updates, and it waits. This week, it was right to wait.

In 1748 the Scottish philosopher David Hume published An Enquiry Concerning Human Understanding, in which he warned, with the calm exasperation of a man who knew his warning would be politely ignored for several centuries, that human beings habitually mistake temporal sequence for causation. When one billiard ball strikes another and the second one moves, Hume wrote, we do not actually see a cause. We see a sequence. The leap from “A happened, then B happened” to “A caused B” is something the mind adds, on its own, after the fact. Sports media commits a version of this error roughly daily, including by this newsletter, including yesterday morning. This issue inaugurates a new recurring thread: The Reader’s Defense. The goal is to send readers home with vocabulary they can use to spot the error the next time it walks past them in a headline. We start with the version we just published.

A piece making the rounds this week declares Juan Soto the least-supported hitter in Mets history — his nearest teammate sits 94 places below him on the league leaderboard. The claim is built on a real and genuinely bleak chasm. But “94 spots” is the wrong number to be frightened by, and the reason why is the most useful thing a baseball leaderboard can teach you: rank gaps and skill gaps are not the same animal, and on a leaderboard the two come apart most violently exactly where Soto is standing.

Lucius Quinctius Cincinnatus was a Roman patrician who, in 458 BC, was called from his farm to be appointed dictator for the purpose of rescuing a Roman army trapped on Mount Algidus. Sixteen days later, having defeated the Aequians, he resigned the office, returned the powers, and went back to plowing his field. The story made him the moral exemplar of the early Republic: the leader who voluntarily relinquishes power once the job is done. The American city of Cincinnati was named, indirectly, after him — through the Society of the Cincinnati, the post-Revolutionary War officers’ club whose founders saw George Washington as their own Cincinnatus. The football program at the University of Cincinnati is in this issue because, across the past two decades, it has run the exact opposite of the Cincinnatus play. Each of its last five head coaches, when offered a bigger job, has taken it. The two who built the program to its highest peaks — Brian Kelly in 2009 and Luke Fickell in 2021 — were precisely the two whose work made the next offer impossible to refuse.

Aristotle, in a passage of the Politics still quoted today, wrote that “what is common to the greatest number gets the least amount of care.” He was arguing against Plato’s plan for collective property. The line endures because the principle is older than the argument: when individual incentives diverge from the collective good, individuals optimize and the commons decays. This is the only issue in this series where the Pyrrhic victory is paid for not by one team but by an entire league. The 2017 Houston Astros are the team that won. Major League Baseball is the commons that lost.

In Greek mythology, Tantalus is the king condemned to stand forever in a pool of water beneath a fruit tree. When he stoops to drink, the water recedes. When he reaches for the fruit, the branches lift. The prize is always exactly in view and never quite obtainable. His name became the root of the English word tantalize. In May of 2012 the Charlotte Bobcats finished a lockout-shortened season at seven wins and fifty-nine losses, the lowest winning percentage in the history of the National Basketball Association, and entered the draft lottery with the highest possible odds of selecting the consensus best prospect of his generation. Then they reached.

The Romans built temples to Fortuna because they understood, as we have mostly stopped admitting, that the best-laid plan can be undone by a roll. Sallust wrote that Fortune ruled in all things and gave success or failure not according to the real truth but according to her own caprice. The 2015 NHL draft lottery was a temple of that same goddess. The Buffalo Sabres arrived at it with the worst record in hockey, the best odds, and what the tank-or-not-tank discourse called “the most successful tank in NHL history.” They left with the second pick. The team that won the lottery had spent the previous five years winning it.

Vegas leads the 2026 Stanley Cup Final 2–1 after three consecutive overtime thrillers. The Carolina historical comp from Issue #56 gets its promised grade: PENDING, series live, mathematical path narrowed from 75% to 31%. Five open predictions. Zero resolved. Full accounting inside.

Six programs are now valued north of $40 million for 2026, and the Irish are one of them. But the team that won last year’s title wasn’t the biggest spender — and the No. 6 roster on this very list couldn’t even make the 12-team playoff a season ago. At the top of the market, money doesn’t buy the trophy. It buys a ticket to the lottery.

Down 2–0 in Carolina, the Golden Knights stormed back to win Game 1, 5–4 — the first road team in NHL history to erase a multi-goal deficit in a Stanley Cup Final opener. History says the Game 1 winner lifts the Cup 77% of the time. The math of a single 1–0 lead says something more modest. Both are true.

Cal Raleigh showered in his uniform on May 12 and snapped a 38-at-bat hitless streak the next night. Wade Boggs ate chicken before every game of an 18-year Hall of Fame career and slashed .328 lifetime. Moisés Alou famously refused batting gloves and reportedly “toughened” his hands by an unsavory pre-at-bat ritual that we will not describe in print. The 1986 Mets turned their caps backwards in the seventh inning and beat the Red Sox in seven. Every one of these stories ends with the apparent confirmation that the superstition worked. The deeper question is why they all always seem to work, even though most of them, on rigorous examination, do not work at all. The answer comes from three statistical concepts the newsletter has now mentioned in three consecutive issues: regression to the mean, confirmation bias, and base rates.

On October 3, 2023, Mariners president of baseball operations Jerry Dipoto told reporters his organization operates on a ten-year plan to win 54% of its games. He said this two days after the Mariners missed the playoffs for the 21st time in 22 years. Seattle fans were not amused. The math, however, deserves more than the headline outrage allowed. Dipoto's number is approximately right — the historical American League Wild Card 3 cutoff has indeed sat around 86 to 89 wins, very close to 54%. The deeper problem is what happens when a team sets that number as its ceiling: the Mariners have hit it on the nose twice in three years and missed the playoffs both times.

Most pre-season predictions hide their uncertainty in fake precision. ESPN's experts will rank thirty-two NFL teams to one decimal place; nobody can defend the difference between the seventeenth and eighteenth ranked team. Today we introduce a recurring format — the Pre-Season Simulation — that goes the other way. It uses consensus pre-season expectations as the prior, applies a just-noticeable-difference threshold cribbed from the Significance Series, and forecasts every regular-season game by one of two rules: higher-projected team wins, or coin flip. There is no middle ground, on purpose. The point is to lock in only what the data justifies and to explicitly mark the games that pre-season expectations cannot tell us about. The first sport up: NFL, in late August.

In 279 BC, King Pyrrhus of Epirus beat the Romans at Asculum and lost so many of his irreplaceable officers that, as Plutarch tells it, he said: “One more such victory and we are utterly ruined.” A win that costs more than a loss has carried his name ever since. This new series looks for Pyrrhic victories in sports — wins that quietly cost a franchise far more than the defeat would have. We begin with the cleanest example of the modern era: the two games the 2020 New York Jets won, after starting 0–13, that cost them Trevor Lawrence.

Week 8 recap: Devin Williams' Bayesian model upgrades from PARTIALLY HIT to HIT — zero earned runs all of May, ERA migrating from 10.29 to 4.32. Two timing misses on Lindor and Alvarez, who remain on the IL despite "before end of May" projections. Carolina sweeps Montreal and enters the Stanley Cup Final. The scorecard is honest about all three.

Between April 30 and May 12, Seattle Mariners catcher Cal Raleigh went hitless in 38 consecutive at-bats — the longest slump of his career. Teammate Logan Gilbert convinced him, on the night of the 12th, to shower in full Mariners uniform to wash off the baseball juju. On May 13, Raleigh went 2-for-4 with a walk and three runs scored, the slump broken at last. Two days later, he was on the injured list with an oblique strain. The shower was a great story. The math says the streak was going to end anyway. The math also says the oblique was probably the reason it started.

When two teams finish tied at 89–73, as the Mets and Braves did in 2024, the league reaches for the head-to-head record to break the tie. It looks rigorous. It is settled in advance, by formula, with no ambiguity. But the head-to-head record between two teams is built from a sample so small that, under statistical scrutiny, it cannot reliably distinguish the best team in baseball from the worst. The Braves won the 2024 NL tiebreaker over the Mets 7–6. The probability of that outcome under the assumption that the two teams were equally good is, to four decimal places, 1.0000.

Three teams in New York — the Mets, the Jets, and the Rangers — have a combined 129 years since their last championships. Three teams in Washington, D.C. — the Nationals, the Capitals, and the Commanders — have a combined 49 years. The gap is eighty years of cumulative drought. The wrinkle is what happened between June 2018 and October 2019, when D.C. won two titles in eighteen months and broke a regional drought of its own that had stretched twenty-seven years. The math says D.C. fans have less misery than New York fans by every available measure. The harder claim is what to do with that.

Yesterday we used Bill James’ 1981 Pythagorean expectation to ask whether the Mets’ 20–26 record matches their underlying play. The answer was: close enough. Their 176 runs scored and 186 runs allowed predict a 22–24 record — two wins under, but inside the noise band. Pythagorean treats runs scored as a fact about the team. BaseRuns, the formula proposed by David Smyth in the 1990s and now standard at FanGraphs and Baseball Prospectus, treats runs scored itself as a thing that can be unlucky. It goes one level deeper. Today’s issue is about why that matters — and what it tells us about a Mets offense that has scored fewer runs than 23 of the 30 teams in baseball.

In 1981, Bill James proposed a formula so simple a 12-year-old could compute it: take a team’s runs scored, square them, and divide by the sum of (runs scored squared) plus (runs allowed squared). The result, multiplied by games played, is the team’s expected wins. The Mets, through 46 games, have scored 176 runs and allowed 186. James’s formula says they should be 22–24. They are 20–26. They have been, in close-game outcomes, a touch unlucky. The harder problem is that the same formula, projected forward, says this team is on pace to win 76 games — not the 87 the wild card has historically required.

Notre Dame signed the #2 high school class in the country in February, added four top-100 transfers in the spring portal window, and has spent the past six weeks building a 2027 commitment list whose composite score, per On3, is already higher than the class that finished second in America. Most programs trade off one front for the other. Notre Dame is winning at both. One-third of the 2026 roster will be either a freshman or a first-year transfer, and the 2027 class is on pace to repeat the trick.

Welcome to Sunday Edition No. 007. The Mets have slid to the bottom of the NL East and are running well off the .578 wild-card pace from Issue #52. The Hurricanes’ 8–0 playoff streak ended Thursday in a four-goal Canadiens first period. Devin Williams, on the other hand, has not allowed an earned run all month, quietly vindicating Issue #45’s xERA-driven Bayesian projection. The week’s corrections, including a substantive dateline error in Issue #54, get their own section further down.

The Hurricanes have swept both Ottawa and Philadelphia to open the 2026 Stanley Cup Playoffs — the first team to sweep its first two rounds in the four-round, best-of-seven era introduced in 1987. The closest historical comparable is the 1985 Edmonton Oilers, who also started 8–0, in a different playoff format. The team before them in any format was the 1969 St. Louis Blues. The thing Carolina is doing has not been done in the lifetime of most NHL coaches, and never in the bracket fans know.

A primer on the chi-square test, the oldest and most useful tool for asking the question every fan asks all summer: are those two teams actually different, or did the dice just roll their way? Today we set up the simplest possible case — the Mets and the Yankees, head to head, at four points in a hypothetical season — and discover the surprising number of games one team needs to be ahead before statistics will let us call them different. The answer, every quarter, is bigger than you think.

Last night, in the eighth inning of a 2–2 tie, Craig Kimbrel walked to the mound to face Jake McCarthy and the Colorado Rockies. He gave up a single, walked a batter, gave up another single, and then gave up a grand slam. The Mets lost, 6–2. This issue is about how a $250 million payroll has produced the National League’s most expensive late-inning catastrophe.

A neighbor of the editor, Chris Kelly, writes: “I’ve noticed over the years that teams with more patience — more pitches taken per at-bat — tend to win more often than free-swinging teams. But I could be wrong.” Today’s issue is about that last clause. The intuition is reasonable. The team-level data does not, in fact, support it. The correlation between team pitches-per-plate-appearance and team winning percentage, across the modern era, hovers around r = 0.07. That number is, in practical terms, indistinguishable from zero.

After Sunday’s 7–6 win over the Yankees, capped by a Tyrone Taylor home run in the ninth and a Devin Williams hold-and-win in the tenth, the Mets are 20–26. To reach the average National League wild card cutoff of 87 wins, they need to play .578 baseball the rest of the way. That is roughly what PECOTA and FanGraphs projected them to play in March. The famous comeback teams — the 2019 Nationals, the 2005 Astros, the 1973 Mets — were all in worse holes than this when they started climbing out. The math is not the problem. The next fifty-eight days are.

A companion piece to last week’s issue on cost-per-win. Same payroll. Same wins. Same thirty teams. Three different charts — each one technically honest, each one telling a story the data does not, in fact, support. Consider this a guided tour of how a sports columnist could lie to you with statistics without saying a single false sentence.

The Semicentennial Edition: six issues recapped, the final resolution of Issue #38’s log5 forecast (model said 6 wins, Mets have 8 with four games left), the Clay Holmes injury and what it means for everything that follows, and a methods retrospective on what fifty issues have taught the editor about predictions.

Since 1995, the Atlanta Braves have won the National League East seventeen times. The New York Mets have won it twice. The all-time head-to-head between the two clubs is 510 wins for Atlanta, 419 for New York. This is, on paper, one of baseball’s great rivalries. In practice, only one team has been showing up to it.

The Mets are paying $17.7 million per win. The Marlins are paying $2.0 million per win. One of these teams is in last place. The other is also in last place. Today’s issue is about why the most popular ratio in sports economics tells you almost nothing.

Hate, in baseball, is supposed to be reserved for villains. The Colorado Rockies have spent thirty-three years failing in slow motion at high altitude, and they cannot make a single rival fanbase angry. The Mets, sitting in last place at $248 million, are the fourth-most-hated team in the sport. This issue is about the difference, and what it says about why we hate who we hate.

For the first time in the transfer-portal era, college football has only one window. It opened January 2 and closed January 16. There is no spring window in 2026 — it was eliminated last fall. Rosters that were assembled in those fifteen January days are now locked through bowl season. This piece is about what that means, mathematically, for the teams that bet right and the players who didn’t make it onto a roster before the door shut.

In 2020, Devin Williams posted a 0.33 ERA — the lowest single-season ERA with twenty-plus innings since earned runs were first recorded in 1913. In April 2026 he is 9.95 in the same role, on the same arm, in front of a Mets fanbase that paid him to be the 2020 version. The math on what happened, and why it was always coming, is unkind to free-agency optimism.

Part 4 of the After-the-Jets series tests a long-held belief: that scouting departments distinguish themselves not in the early rounds, where the talent is obvious, but in the late ones, where it isn’t. The data confirms it. The relative spread of late-round hit rates is roughly three times the early-round spread — and the team rankings on the two metrics tell entirely different stories.

A clean six-issue week — two on the Jets, two on MLB, one on owners, one on coaches. The big prediction piece, Issue #38’s log5 forecast of the Mets’ next seventeen games, is now half-resolved. Through the first eight of those games, the Mets are running 1.3 wins ahead of pace, and the model’s pessimism about a 6-of-17 finish looks slightly too dark.

Most owner rankings rest on a single number — championships, usually — that overweights the lucky and underrates the patient. This piece builds an Owner Index that respects variance: three metrics, z-scored within league, averaged. The top and the bottom of that ranking are not who the headlines say.

Part 2 of the After-the-Jets series ranks all 32 NFL franchises on the three Issue No. 25 bust definitions. The Jets are below average. They are not the worst. The worst is Cincinnati, by a margin that is genuinely surprising. And the Jets have a paradox of their own.

A headline at The Athletic this week asked whether the Padres’ closer is the most unhittable pitcher in baseball history. The answer is no, and the reason is not Mason Miller. The reason is the size of the sample required to use a sentence like that.

Issue No. 25 noted that the Jets had produced exactly one first-round Hall of Famer since 1976. The number sounded damning. The number is, across the league, the modal outcome. Part 1 of a four-part series on what the Jets’ draft history looks like once you put it next to everyone else’s.

Six series. Three home, three away. One opponent below them in the standings, one currently the second-best team in baseball. The log5 model says the next two and a half weeks should produce six wins, give or take. That is not a recovery pace. That is the pace of running out of road.

Fans want patience. Boosters want results. The math says the boosters are closer to right: a single 12-game season explains 60% of a coach’s true ability. By Year 3, the picture is 82% clear. The “5-year plan” is a $20 million way to learn what 12 games already told you.

Six issues in seven days: four methodology tools, one retrospective, one conditional-probability case study. We score every outstanding prediction, close the Skenes ERA call as a HIT, and explain why the Mets beating the Angels on Friday is not evidence of anything.

A 10-21 record is a fact. The probability of recovering from it is something else entirely — and the way you ask the question changes the answer by an order of magnitude. A short course in conditional probability, with the Mets as the patient.

Since 1972, a fan of the Mets, Jets, Rangers, and Notre Dame has watched 5 championships — and none since 1994. A Boston fan has celebrated 14. A Buffalo fan has celebrated zero. A 32-year drought across 4 teams is a 1-in-185 event. Here’s the fan misery map — who has it best, who has it worst, and where you land.

The executive order limits athletes to one free transfer. But junior college players aren’t transferring — they’re arriving. The JUCO-to-D1 pipeline doesn’t trigger the portal’s one-transfer cap, and smart programs are already loading rosters with physically mature, battle-tested players who never burned their portal move. If you want to see the future of college baseball under this EO, look at the player who led the Big 12 in hitting last year: he came from a community college in Las Vegas.

Every April, a hitter on a hot streak hits .450 and fans decide he is the MVP. Every April, the same hitter regresses and fans decide he is finished. Both readings are wrong for the same reason: the denominator is too small. Here is the reference card.

Every Saturday during the college football season, fans argue about whether their team should be ranked #8 or #11. Thirty years of data say those arguments are mathematically indistinguishable. Here is what the ranking can, and cannot, tell you — and the surprising way this changes across the season.

The NFL’s polite convention is that a first-round pick gets three years, a head coach gets four. The statistical convention should be shorter. Here is what the numbers say about how quickly a career shows its ceiling — and why front offices persist in not believing them.

A 12-game losing streak ended the same night Juan Soto returned from the IL — and the same night Francisco Lindor injured his calf and walked off. We made one strong prediction this week and it landed precisely. We also told you, in print, that the Jets would pick seventh on Thursday. They picked sixteenth. Owning that.

Since 2023, the Mets have parted with a meaningful number of players. The roster that replaced them is currently 7-14. The honest accounting is neither "they let everyone go too cheap" nor "every departure was correct." It is somewhere uncomfortable in between, and the replacements are the part that hurts.

The two teams share a market, a building, and as of this Thursday evening an actual draft class picked in the same city. Their first-round histories tell a less obvious story than the standings do — and not the one most fans would guess.

Twenty-one games in. An eleven-game losing streak. The worst start the team has produced since 1983. What the historical base rate says about where this ends.

The first round is where every team puts its most confident bet. Since 1976, the Jets have placed 57 such bets. Twenty-eight percent of them — more than one in four — produced careers that failed every reasonable definition of success. The draft is about to happen again.

We computed the phi coefficient between the S&P 500’s daily direction and six MLB teams’ win-loss records. The Mets came back with φ = −0.567, p < 0.05. On days the market goes up, the Mets lose 90% of the time. The correlation is statistically significant. It is also the most dangerous kind of number in statistics: one that looks real and isn’t.

The popular comparison pits the WNBA against the NBA — an 80-year-old league against a 30-year-old one. But there’s a better comparison right next door: Major League Soccer, founded the same year, facing the same growth challenges, in a similar position as a “secondary” league in its sport. The salary gap between them? Eight percent.

The Lady Vols’ entire roster entered the transfer portal — zero returning players on a Pat Summitt-legacy program. But the deeper question: in which sport can losing just one superstar crater an entire season? A Bayesian model across four sports runs the numbers. One LeBron is worth 42 wins. One Mike Trout is worth 10. The math is structural — and it starts with counting to five.

We gave the Mets’ .500 start a pass two weeks ago. We shouldn’t have. They’re 7-14 with a 10-game losing streak — tied for the worst record in MLB. Meanwhile, Skenes recovered exactly like we modeled, the Rangers missed the playoffs, and the WNBA’s new CBA landed three days after we called the league “ahead of schedule.” Three hits, one miss, one partial. The scoreboard doesn’t lie.

The popular framing compares a 30-year-old league to an 80-year-old one and calls the gap a scandal. But when you compare the WNBA to where the NBA was at the same age — same maturity, same trajectory — the picture changes completely. The WNBA generates 3–4x more revenue, pays comparably after inflation, and is the most selective major league in America.

Transfer limits, NIL caps, and eligibility rules all favor programs that were already winning. Blue bloods won 85% of titles pre-portal. The portal briefly opened the door. The EO closes it. But threatening $380 billion in federal research funding over sports rules is a precedent that should worry everyone. Part 5 of 5 (Final).

Five starters with ERAs under 3.10. A bullpen at 2.84. And yet the Mets are playing .500 ball. The culprit isn’t on the mound — it’s in the batter’s box, where a $765 million outfielder is on the injured list, a $341 million shortstop is hitting .135, and the franchise’s most expensive offseason addition is posting an OBP that starts with a “2.”

Curt Cignetti used 18 incoming transfers to turn 3–9 Indiana into an 11–2 CFP team. Under the executive order’s transfer limits, that instant rebuild becomes nearly impossible. The drawbridge rises after the king is inside — and mid-tier programs may never get that chance again. Part 4 of 5.

The executive order’s sports provisions help Notre Dame on every front: transfer limits favor their recruiting, NIL caps compress spending toward their level, and the 5-year eligibility rule suits their young roster. But the enforcement mechanism — threatening federal research funding — puts $200 million at risk for a private university already dealing with grant cancellations. Part 3 of 5.

The first report card. We graded every prediction the newsletter has made against fresh data: Soto’s injury model nailed it, the Rangers collapse confirmed, and Skenes’ recovery is slower than our Gamma-Poisson model projected. Three clean hits, two partials, two pending.

Texas A&M spent $51.4 million on NIL. The average Group of 5 school spent $3 million. That’s a 14× gap. The executive order aims to compress it — and the math says it would push the entire sport toward Notre Dame’s spending level. Part 2 of 5.

On April 3, the President signed an executive order limiting college athletes to one free transfer. The portal grew from 1,946 entries to 3,100 in three years — now it’s about to shrink by a third. The schools that built dynasties through recruiting will thrive. Schools that built rosters through the portal — including Cignetti’s Indiana — may never get that chance again. Part 1 of 5.

The Raiders signed a quarterback who reportedly refused to help his replacement last year, then plan to draft his replacement this month. History says supportive mentors produce stars 80% of the time. Hostile ones? 25%. Here’s what the data — and every other sport — says about teaching someone to take your job.

The Rangers dropped 36% in two seasons — the biggest collapse from a Presidents' Trophy in NHL history. The Sabres are one win from ending a 15-year drought. And the Golden Knights, six years removed from their Cup, are aging into irrelevance. One chart tells the whole story.

Sean asked: will the A's get better? The relocation scorecard across 10 moves in 4 sports says 60% improve — but the average gain is just .050. The real variable isn't the city. It's the owner. And Gruden's emails cost the Raiders ~12 wins.

If you assembled a team from every player the Mets traded away or let walk in the last five years, would it beat the actual 2026 Mets? Pete Crow-Armstrong alone — traded for a 2-month rental — is worth more than everything they got for him.

Lindor is at .143. Bichette is at .111 with a 53% strikeout rate. Both slumping simultaneously is a 1-in-77 event. But 15 PA is 10% of the evidence needed to judge. Mets fans booed at 10% of the evidence.

Soto left Friday's game with right calf tightness. MRI pending. He was hitting .333 and carrying 28% of the team's hits. A Bayesian model says 54% chance he misses 10+ games. The lineup without him hits .211.

A 29% hit rate on first-round picks. Three first-round QBs with zero Pro Bowls. The longest playoff drought in professional sports. And they bust on higher picks than they hit on. Part 1 of the Jets Misery Series.

The 2026 NFL Draft is three weeks away. The math of combinatorics guarantees that virtually every mock draft will be wrong. The best analyst in the world nails maybe half. Here's why 32! is a 35-digit number, why trades break everything, and what your Jets, Raiders, and Notre Dame should expect.

The Mets are 3-3 with 50% of their games going to extras — a 1-in-99 event. They went 1-for-29 with RISP in St. Louis. But .500 after 6 games tells you exactly as much about October as the 2025 Mets' 45-24 start did: nothing. A Beta-Binomial model, PTSD therapy, and the new Sports Page layout.

Since Thanksgiving, the market swallowed 14 cents of every dollar. Tariffs, war, and oil did the damage. But it turns out the NY Rangers are declining at almost exactly the same rate. A cross-sport, cross-market decline comparison — plus a Monte Carlo model for when you break even.

Pete Alonso is 1-for-6 in Baltimore. The Mets are 2-0 without him. A Bayesian model says there's a better-than-coin-flip chance he never cracks .250 this year. The decline curve, the contract math, and why the Orioles bought the audition tape while the Mets had seen the whole movie.

5 earned runs in two-thirds of an inning gave baseball's brightest young arm the ugliest number imaginable. But the denominator is a liar — here's the math of why Opening Day disasters disappear, and how long it'll take Skenes to make this one a footnote.