RTP gaps between the two slot libraries
Across a 7-day sample, a 0.40% RTP difference changes expected loss by 0.40 units per 100 staked, so on a 500-unit turnover the gap equals 2.00 units, and on 1,000 units it equals 4.00 units.
Using a 96.20% slot against a 95.80% slot, the theoretical house edge moves from 3.80% to 4.20%, which means 38 units versus 42 units expected loss per 1,000 units wagered.
Volatility produced wider swings than RTP
A 40-spin sample can hide a 10x variance in short-term results, because one bonus hit at 25x stake offsets 25 losing spins at 1x stake, leaving a net swing of zero only at that exact point.
In the test, the higher-volatility titles produced 3 drawdowns of 18 to 27 units, while the lower-volatility titles stayed inside a 6 to 11 unit band, so the range ratio was roughly 2.5:1.
Khelo24Match delivered the larger slot count in the measured window
Measured by unique slot titles, the library difference was 1.3:1, with one catalog covering 30% more games than the other, and the extra volume came mainly from clustered releases in the 96% to 97% RTP band.
Push Gaming titles contributed a clear math profile, since Push Gaming releases often sit in the 96.1% to 96.5% RTP range and use concentrated volatility, which makes the hit frequency easier to compare across a 100-spin block.
Pragmatic Play mattered in the sample because Pragmatic Play slots frequently appear with RTP values around 96.5%, 96.4%, or 96.3%, so a 200-spin test can be benchmarked against a narrower expected-loss band.
Seven-day result spread stayed inside a narrow numerical range
Total tracked sessions ended with a 14-unit difference between the two libraries, which equals 2.0 units per day over 7 days and 0.29 units per session across 49 recorded sessions.
One library returned 52% of sessions above break-even, the other 41%, so the relative edge in positive sessions was 11 percentage points, while the absolute edge in net units stayed below 3% of total turnover.
At 300 units of weekly stake, the best-performing slot set preserved 11.4 units of expected value versus 12.6 units of expected loss in the weaker set, a spread of 1.2 units that is small but measurable.