Story of the Line

The Pundits

Every fanbase has these six voices. We turned each one into a model that picks every game, froze their lines at the opening whistle, and grade them in public against the closing market. These are their stories.

Bandwagon

the crowd

“A lot of people looked at the same game and landed on the same side. You sure you know something they don’t?”

Bandwagon reads the public betting splits — the share of tickets on each side at a major US sportsbook — and takes them literally. If 85% of bets are on the favorite, Bandwagon says the favorite wins 85% of the time. No shading, no shame.

This makes Bandwagon the boldest voice on every card: when the crowd is right, it posts the closest read of anyone and collects game balls in bunches; when the crowd is wrong, it finishes last by a mile. High variance is the whole character.

Cash

the money

“Don’t count how many bets came in. Count how much money came with them. Some opinions are bigger than others.”

Cash reads the same public-betting feed as Bandwagon but follows the other column: the share of dollars instead of the share of tickets. A thousand $5 bets on one side and three $10,000 bets on the other look identical to Bandwagon; to Cash they point in opposite directions.

The days Cash and Bandwagon split are the interesting ones — that gap is the crowd disagreeing with its own wallet, and it's the split the daily brief watches closest.

Mover

line movement

“Lines don’t move because the mood changed. Show me where the number went and I’ll show you where the pressure was.”

Mover ignores the teams entirely and watches the number. It compares where the line opened to where it stands now, and leans in the direction of the move — the bigger the move, the harder the lean. A line that steams from 52% to 64% tells Mover that serious money showed up, and Mover rides with it.

Mover inherits the market's wisdom plus its overreactions. On steam days it looks brilliant; when a move reverses, it's the last one holding the bag.

Recency

recent form

“A season average can hide what’s happening right now. When a team finds something that works, I don’t wait for the year-end numbers to notice.”

Recency starts from the market's number and bends it toward whoever is hot: a team's recent results push its probability up or down from the baseline. Win five straight and Recency believes in you more than the market does; lose five straight and it fades you harder than the price says it should.

Streaks carry real signal and real noise in a ratio nobody knows in advance. Its whole season is a referendum on how much momentum actually exists.

Buzz

the hype

“Headlines, hot takes, panel shows. I don’t care whether the pundits are right. I care who they can’t stop talking about.”

Buzz measures attention: it tracks how many people are reading each team's Wikipedia page and leans toward the club drawing the bigger crowd. When one side of a matchup is suddenly being read three times as much as the other, Buzz takes that as information — something happened, and the market may not have priced all of it.

Attention spikes for good reasons and bad ones alike — a trade-deadline star and a clubhouse scandal read identically. Buzz's grade tests whether raw attention carries any edge at all once the market has spoken.

Quant

the full model

“Use all the data, then make every variable earn its place. What helps stays. What doesn’t, gets cut.”

Quant is the house model: a trained forecaster that sees everything the other five see — form, matchups, starters, prices — and learns from history which of it actually predicts games. Features that don't help get cut. It's the only pundit whose opinion comes from a fit rather than a rule.

Quant's bars carry a ring on every card because it's ours — and it gets no grading favors for that. Most nights the honest story of sports forecasting is how hard the closing line is to beat, and Quant's row on the scorecard tells it.

How the grading works: every pundit's line is frozen at the opening whistle (first pitch, tip-off, kickoff) and scored against the final result. Skill is 1 − Brier / baseline, the percent better than a know-nothing guess, so 0% is blind guessing and the closing market usually sits well above it. The chips above show each pundit's live season record by sport; click through for the full scorecard.

Where the lines come from: the Market row is the sportsbook average, devigged, across the books listed under each card's best-price row. Every number on a card is a closing-line snapshot taken before the game starts, not a live in-game price, and the books quoted are whichever ones you have left switched on under “Your books”.