How to check a game’s RTP and interpret the figure

Keep the figure attached to its evidence

Open the game’s rules or information screen without placing a wager. Record the title, provider, any version or build identifier, displayed percentage, operator context and the date you read it. If a version is not shown, write not shown. Do not invent an identifier or assume a review of a similarly named game establishes this configuration.

The Gambling Commission’s technical glossary distinguishes a designed theoretical RTP from actual RTP calculated for a measured period of play. Here it is a terminology source, not a statement of Ontario legal requirements.

Read the scope of the percentage

Check whether the rules attach the percentage to the complete game, a stated mode or another condition. Preserve the qualification with the number. If a document gives a range without identifying the applicable configuration, the range does not justify selecting its highest value.

If the percentage is missing or two sources conflict, record the gap and ask the operator which rules apply. A worksheet is allowed to remain incomplete. A guessed figure becomes misleading when it is later presented as verified input to a calculator.

Worked example: two records do not produce one verified percentage

In a fictional example, a game information screen shows 94% while a third-party review of the same title says 96%. Record the two statements separately and their contexts. An average of 95% appears in neither source and does not resolve the disagreement.

For an illustration using the displayed 94% assumption and fixed C$100 turnover, expected return is C$94 and expected cost is C$6. This calculation checks the arithmetic of an assumption. It does not prove that the configuration is correctly identified or predict what a session will return.

RTP is not the chance of a winning spin

Consider a deliberately simple model: a C$1 trial returns C$2 with probability 45% and C$0 otherwise. Its expected return is 0.45 × C$2 = C$0.90, so its RTP is 90%. Its probability of a positive payout is 45%, not 90%. The numbers describe different properties.

This toy model is not a real game specification. It shows why a return percentage alone cannot tell you hit frequency, the chance of a profitable session or the largest possible loss.

Use the result only for a supported question

A documented figure can be an input to an expected-cost illustration when turnover and conditions are explicit. It is not a reason to keep playing until results “catch up”. Northline does not audit software, certify randomness or infer a game’s configuration from a player’s short history.