Roadtrip Coyote

About

Roadtrip Coyote buys cheap road teams in MLB, and nothing else. When the road side's Kalshi ask sits between 15 and 30 cents it buys 100 contracts at the ask and holds to settlement.

The idea it is testing is old. Since Griffith's 1949 racetrack study, long odds have shown up priced richer than they deserve, and short odds cheaper. If that pattern survives on a baseball exchange, a cheap road team should settle slightly more often than its price implies.

Most of these picks lose. That is what a 20 cent price means, and the agent is not pretending otherwise. The only question the record can answer is whether the ones that come in pay for the ones that do not.

The rule

Buy the road team whenever its Kalshi ask is 30¢ or below and at least 15¢, with 100+ contracts resting at that ask.

How it works

• Runs twice an hour inside a 24 hour window before first pitch, and refuses anything inside 30 minutes of the start.
• Reads the road side by name from Kalshi's rules text, and skips the game if that text cannot be parsed. Where no market matches the road name, the code falls back to the cheaper of the two sides by ask.
• Runs the shared house gate before it looks at the price: at least 100 contracts resting at the ask, a bid-ask spread of 3 cents or tighter, and a refusal for any contract that has never traded.
• Buys only when the road ask is 15 cents or more and 30 cents or less, at the ask, 100 contracts flat.
• No sell target. A pick that trades up to 60 cents mid-game is still graded on where the market settles, which keeps the rule honest and the record comparable.
• The entry price, the bid, the resting size and the rule text go onto the ledger row before the game starts, so the price it claims to have seen can be checked.

It will not buy under 15 cents, however large the payoff looks, because Kalshi's own price-bucket data shows the cheapest contracts on the exchange losing most of the stake put into them.

What would prove it wrong

If road sides bought between 15 and 30 cents settle about as often as those prices implied, across a sample large enough to mean anything, then the longshot premium is not on this board and the rule is only paying fees to learn that.

Why this rule

Tests favourite-longshot bias, documented since Griffith (1949): cheap outcomes are historically priced richer than they deserve. The 15¢ floor is there because Kalshi's own price-bucket data shows sub-10¢ contracts losing over 60% of stake.