A designed percentage is not a personal receipt
The Gambling Commission distinguishes theoretical RTP from observed performance. The theoretical figure is the designed return displayed in the game rules; observed results can differ over a limited sample. A player’s session is not a controlled demonstration of that theoretical percentage.
The figure should also belong to the actual version. Push Gaming’s original Jammin’ Jars page lists both 96.83% and 94.25% configurations. A review which prints just one number without identifying the configuration cannot establish the value in every operator’s version.
Variability is a different property
Volatility concerns the spread of results. The Commission’s technical terminology notes that standard deviation is commonly used to represent it. A broad high or low label does not itself give the chance of a win on a particular round or the expected time between features.
Two games can have a similar theoretical RTP and different patterns of awards. Conversely, two games described by the same broad volatility label need not have identical rules, hit frequencies or maximum awards.
| Term | What it describes | What it does not promise |
|---|---|---|
| Theoretical RTP | A designed return property. | The amount returned to your balance. |
| Observed RTP | Results from a defined body of play. | That the same result continues. |
| Volatility | Variability in outcomes. | A particular next award. |
Check the help rather than decoding a title
A market identifier does not supply every numerical property of a game. Relax Gaming lists an alternative UK GameID for Money Train 4, but the suffix alone does not verify an RTP percentage. Global euro stake figures also do not establish GBP stakes or GB limits.
Read the actual information panel for the configuration, mechanic and award rules. A connected-cluster game and a Scatter Pays game are different rule systems even if both display an RTP. The percentage is not a substitute for understanding how the game resolves a round.
Three common readings to avoid
First, a losing sequence does not make a compensating win due. Second, a hypothetical average is not a prediction of the cost of completing a bonus requirement. Third, a high possible award does not establish how often a particular player will reach it.
The Commission requires testing of regulated online RNG games under its testing strategy. Testing concerns fairness and the designed behaviour; it does not guarantee an individual return. Use the published figures to understand the product, not as an instruction to keep playing until a percentage appears to balance out.
A few clear answers
Does a 96.83% game give that percentage back in one session?
No. The percentage is a theoretical property of an identified configuration, not a promised session result.
Are RTP and volatility interchangeable?
No. Return and variability describe different aspects of the game.
Sources behind this note
Public documentation checked on 10 November 2026. Account-specific conditions and game versions may differ. This is not a record of a test deposit or played round.