Audit the failure class
| Failure class | Pre-bet question | Evidence to retain |
|---|---|---|
| Contract | Is every event, market, line and period exact? | Ticket and market wording |
| Probability | Is each estimate a probability rather than a trend or price? | Model, version and uncertainty |
| Dependence | Can one leg change another's chance? | Conditional or joint calculation |
| Information | Was every input available before the decision? | Timestamp and source snapshot |
| Price | Does the accepted quote match the analysed quote? | Receipt and precision |
| Settlement | What happens on a draw, push, void or non-participant? | Current house and sport rules |
| Exposure | Which other open tickets share this outcome? | Event-level concentration map |
| Behaviour | Was stake or leg count changed after a loss? | Pre-committed budget and log |
Mistake 1: an undefined soccer market
"Team A moneyline" can mean a three-way regulation-time selection or a two-way advance/winner market. DraftKings' soccer rules illustrate why qualifying period, draw, extra time, shoot-out and player-participation wording matters.
Write the settlement event in one sentence before assigning a probability. If the sentence cannot be written, the leg is not ready.
Mistake 2: adding conditions for a target payout
An added favorite is still another required event. Under an illustrative independence assumption, four 0.70 legs have:
Joint probability = 0.70^4 = 0.2401, or 24.01%
OpenStax permits that multiplication only for independent events. A large possible return does not show that the offered price compensates for the lower joint probability.
Mistake 3: multiplying related probabilities
Home win, home team over 1.5 goals and home scorer are linked match events. Use a score-state model, conditional probabilities or a direct joint forecast. The sportsbook accepting a same-game combination does not validate the reader's probability estimate.
Mistake 4: treating form as a forecast
Last-five results, head-to-head records and table position are observations. They become model inputs only after a defined transformation, cutoff, baseline and later validation. Do not turn "won four of five" into an undocumented confidence percentage.
Mistake 5: using a displayed price that was not accepted
Record the operator, timestamp, signed American price, market line and final receipt. A later screenshot cannot prove what was available or accepted at decision time. A price improvement is useful only if the same contract and stake can actually be placed.
Mistake 6: assuming universal push and void rules
DraftKings' market rules provide one current example of standard and same-game parlay settlement. Another operator, state, market or promotion can differ. Save the rule version beside the ticket.
Mistake 7: measuring only hit rate
A ticket's result does not reveal whether the decision had positive expectation. OpenStax's expected-value framework requires probabilities and net payoffs. Review accepted prices, probability quality, settlement and cash return separately.
Mistake 8: learning from the result alone
The original outcome-bias research shows why knowledge of an outcome can change how a decision is judged. Freeze the forecast and reasons before kickoff, then ask whether the process was reproducible even when the ticket won.
Mistake 9: hiding shared exposure
Two tickets can both rely on the same club, scorer or weather state. Aggregate maximum loss by event and selection, not by ticket name. Do not count unsettled potential returns as bankroll.
Mistake 10: chasing
Increasing stake, adding legs or switching live markets to recover a loss is a stop signal, not a research method. The AGA responsible-play guidance provides account-control and support routes.
Five-minute pre-bet audit
- Read the exact ticket aloud, including period and line.
- Recalculate the combined price and possible cash return.
- Mark every dependence link between legs.
- Confirm all evidence predates the decision.
- Save current push, void, participation and postponement rules.
- Add the stake to event-level open exposure.
- Reject any leg or stake added to reach a payout or recover a loss.
Next step
Use How Soccer Parlays Work for the next part of this topic.
Continue learning
- Next guide: Soccer Prop Bet Parlays
- Related guide: Soccer Parlay Markets
Assumptions and limitations
The audit cannot prove that a forecast is accurate or a ticket has positive expected value. It is designed to remove avoidable contract, evidence, calculation, settlement, exposure and process errors before the result is known.

