How Many Stars Does It Take to Flip a Team? Baseball: Four. Basketball: One.
You run a middling team. You may raid a superteam, but you must take as few players as you can. How many stars do you need? Baseball says 4 — about 12 wins’ worth. Basketball says 1. Football says one, but only if he plays quarterback. The answer is not about the stars. It is about what each sport is.
The question: you run a middling team and may raid a superteam, but you must take as few players as possible. How many stars does it take to actually change your fortunes?
The answer. Baseball: about 4. Basketball: 1. Football: 1, but only if he plays quarterback.
Why that is surprising. The answer has almost nothing to do with how good the stars are. It is set by how much of the game one player is allowed to touch.
What you get from it: a single number that tells you how each sport shares out its power.
Baseball answers first, and most cleanly, because it has the tidiest arithmetic in sports.
The whole game is priced in one currency: wins above replacementi. Its quiet miracle is that it adds up.
A team of freely available fill-ins wins about 48 games. From there the math is almost eerie: take every player’s wins above replacement, add it to 48, and we land within a few games of the real record nearly every year. A win on paper is a win in the standings.
The baseball answer, in one subtraction
So our puzzle becomes a subtraction problem.
A middling team wins about 78 games. A real contender wins about 90. The gap is 12 wins, and those wins have to arrive in the bodies of new players.
A genuine star — an All-Star season — is worth roughly four or five wins by himself. Do the division and you need three, maybe four.
Which is close to what the argument at the bar produced from the gut: two bats and two arms. The number was not a hunch; it was the arithmetic, felt before anybody worked it out.
Notice why this works. No baseball player, however great, can carry a team, because he only touches so much of the game.
The best hitter alive bats once every nine turns and rests while eight others try. The best pitcher alive throws once every fifth day. Their brilliance is real but rationed by the shape of the sport.
So we need a handful of them stacked together before the standings notice. Four is not a small number because stars are weak; it is small because baseball spreads itself thin on purpose.
Why the other sports need fewer
Now change the sport and watch the answer collapse. The real question was never “how good is a star.” It was “how much of the game can one player hold?”
Basketball hands almost the whole thing to five people. A superstar is one of those five, on the floor for three quarters of the minutes, with the ball in his hands on a huge share of possessions; at his peak we can fairly call him a quarter of his whole team.
Steal that one player and a lottery club walks into the playoffs. The answer is not four, or two. It is one, and the one is often enough to remake a team in an afternoon.
Football sits in between, and it hides a trick.
Twenty-two players start, so almost every one is a thin slice of a thick game. Steal a great cornerback or guard and you have nudged the thing, not flipped it.
Except for the quarterback. He stands behind every offensive snap the team will run. He is the one spot that concentrates the sport the way basketball concentrates it everywhere.
So football also answers one — but only if the one is the quarterback. Take anyone else and you are back to needing a committee.
The number of stars it takes to change everything is just the inverse of how much of the game one star can hold. The count reads the sport back to you.
The ProfessorWhat to take home: the thing the arithmetic cannot see
One honest caution, and the bar argument reached it on its own.
Wins above replacement treats a player like a spreadsheet cell — alone, context-free, additive. Real rosters are not spreadsheets, and I would not trust the number on its own.
Sometimes you add four stars’ worth of wins and get three, because the clubhouse curdles. Sometimes you add two great arms and get five wins, because a settled rotation finally lets the young ones breathe.
The paper number is necessary but not sufficienti. It tells you the floor of what your raid should buy, not the promise. A star who sulks is worth less than his number. A star who steadies a room is worth more. The math gives you the number to beat. Then the people decide whether you beat it.
Which is why the question is good to argue about rather than merely solve. Ask “how many stars would it take?” and you get what sounds like trivia; what you actually get is a definition.
Baseball answers four because it is a democracy of nine and a rotation of five. A game that refuses to let anyone own it.
Basketball answers one because it is a club of five where a single voice can drown out the rest.
Football answers one-if-he-is-the-quarterback because it is a republic with a king at one spot and commoners at the other 21.
Ask how many stars it takes to change a team, and the sport tells you, in one number, exactly how it shares out its power.
Notes & sources
The baseball arithmetic rests on two well-established facts about Wins Above Replacement. First, replacement level corresponds to about a .294 winning percentage — roughly 48 wins over a 162-game season — the record of a team of freely-available fill-ins. Second, WAR is approximately additive: summing a roster’s WAR and adding the ~48-win baseline lands within a few games of actual team records in most seasons (a full run of ten added runs equals about one win). A mediocre team of about 78 wins therefore needs roughly a dozen wins — a dozen WAR — to reach a ~90-win contender’s line, and an All-Star-caliber player supplies about 4–5 WAR, giving three to four such players. Figures rounded to two significant figures.
The basketball and football answers are structural, not exact: their all-in-one value metrics (win shares, various WAR-style measures) are real but are not on the same additive footing as baseball WAR, so the piece argues them from how much of the game one player occupies — minutes and possessions in basketball, the quarterback’s presence on every offensive snap in football — rather than from a single cross-sport number. The chemistry caveat — that added production doesn’t always convert one-for-one — is the point of the closing section.
The underlying idea — that a real answer needs more than the paper number — is the primer on Necessary vs. Sufficient (Concept No. 13). This newsletter’s pieces on win projection and roster value develop the WAR machinery further; see the archive.
Baseball is ranked by championship leverage: we simulate the rest of the season 60,000 times, then split those seasons by who won each of today's games. College football cannot be simulated that way, so it is ranked by how close to a coin flip the ratings make it — a lopsided game teaches you nothing. Tomorrow this block reports what happened to today's pick.
