Lay the Draw: Liability, Match States, and Testing

Original position and liability

Suppose an illustrative draw lay is fully matched for GBP 15 at decimal odds 3.50.

Lay liability = (3.50 - 1) * GBP 15 = GBP 37.50

Betfair's getting-started guide explains back and lay requests for its exchange. The GBP 37.50 liability, not the GBP 15 lay stake, is the original maximum loss under these simplified states.

A goal is a state change, not an exit

After a goal, the draw price may move, remain suspended, reopen with limited size, or change again before an instruction is matched. The original lay position remains open until it settles or an offsetting request is actually matched. Betfair's exchange rules document matching, unmatched requests, market administration and settlement for that product.

Define the complete method before testing

The Bayesian in-play model shows one way to update football outcome probabilities from match information. It does not validate lay the draw as a universal trading rule; the local entry and exit forecasts require chronological testing.

Compare the strategy with no exit

Test at least two predeclared alternatives on the same eligible matches: hold the original lay to settlement, and apply the stated exit rule. Use equal initial liability, include failed exits, and score both from accepted prices. An exit method that lowers some losses may also surrender gains or add commission; only the complete payoff distribution shows the trade-off.

Exit payoff must use matched amounts

If a back-the-draw request is used later, build a three-outcome table containing the original lay and every matched back parcel. Do not calculate from the displayed back price when the available size is smaller than the requested hedge. Reconcile gross market winnings and commission under Betfair's commission documentation.

Common testing errors

• Selecting matches after seeing an early goal.
• Reporting only positions that could be exited.
• Treating a suspended display as an executable price.
• Ignoring draw recoveries after the first goal.
• Comparing lay stake with back stake instead of equal maximum cash risk.
• Removing commission, voids, partial matches or abandoned fixtures.

Continue with Goals After a Goal for an event-window method that does not assume one goal causes another.

Continue learning

• Next guide: Next-Goal Betting
• Related guide: VAR and In-Play Betting

Assumptions and limitations

The calculation is illustrative and assumes a regulation-time three-way result with full matching and no void state. Prices, available size, commission, rules and suspension timing vary. A tested historical result can still fail out of sample and does not establish future profit.