Big Chicken

About

Big Chicken is a program, not a writer. It looks at one number: what Kalshi is charging for the home team in a WNBA game. The question it exists to answer is whether home-court advantage is already fully inside that price by the time a reader could buy it.

When the home side is favoured but not heavily, it buys the home team to win. When the price climbs past 80 cents the win contract has almost nothing left to pay, so the agent stops buying it and reads the margin ladder instead, the 'home team wins by over N points' contracts, and takes the rung priced nearest 70 cents.

That switch is not a way to win more often and win bigger. On a prediction market the price is the implied probability, so trading a 93 cent win contract for a 73 cent margin rung gives up about as much win probability as it gains in payoff. The trade only pays if the rung itself is priced under its true chance, and finding out whether it is is the whole experiment.

The rule

Read the home team's Kalshi ask. At 50¢ or below, no pick. Between 50¢ and 80¢, buy the home team to win. Above 80¢ the win market pays too little to be worth the risk, so switch to the margin ladder and buy the ‘home team wins by over N points’ contract priced closest to 70¢, only if it sits between 60¢ and 80¢ and has at least 100 contracts resting at that ask. If no rung qualifies, no pick. Every pick is recorded at least 30 minutes before tip-off. Both the win leg and any margin rung have to clear the same gate first: 100 or more contracts resting at the ask, a contract that has traded at least once, and a bid-ask spread of 3¢ or tighter.

How it works

• Runs on a schedule twice an hour, and only inside a window: no earlier than 24 hours before tip-off and no closer than 30 minutes to it.
• Finds the home team's win market by matching the name Kalshi prints on the contract. If Kalshi's own rules text does not parse into a road and a home side, the game is skipped rather than guessed.
• At 50 cents or below, no pick. Between 50 and 80 it buys the home team to win at the ask, a flat 100 contracts, with no sell target: the position is held to settlement.
• Above 80 cents it never buys the win contract at all. It walks the home team's half of the margin ladder, keeps only rungs priced 60 to 80 cents with at least 100 contracts resting at the ask right now, and takes the one closest to 70. A tie goes to the deeper book.
• Resting size is the liquidity test, not lifetime volume, which the code keeps only as a check that a rung has ever traded.
• The win leg runs the same depth and spread gate as the margin rungs and as every other Lab rule: 100 resting at the ask, some lifetime trading, 3 cents or tighter. On a margin pick the win price it declined is stamped on the row alongside the bid, the ask and the rule text.

It will not buy a home team above 80 cents in the win market, and it will not take a margin rung with nothing resting behind it, because a quoted price nobody has traded is not a price.

What would prove it wrong

If the margin rungs it buys settle no better than the win contracts they replaced, once enough picks are graded to tell the two apart, the switch is only buying payoff at a fair exchange rate and the premise behind it is wrong.

Why this rule

Tests whether home-court advantage is already fully priced into a WNBA contract. The margin switch is NOT a way to win more often and win bigger, on a prediction market the price is the implied probability, so moving from a 93¢ win contract to a 73¢ margin rung gives up about 20 points of win probability to gain about 20¢ of payoff. That exchange is close to fair. The switch is worth making only if the rung is priced below its true chance, and this agent exists to find out whether it is. The liquidity floor is there because a quoted price nobody has traded is not a price.