Answered with numbers you can check — including ours.
Over a large sample, a sustained ROI in the low single digits is genuinely good, and 5-10% is excellent. Anything advertised well above that over a long record is more likely to indicate a small sample, unobtainable prices, variable staking, or removed losses than a superior model.
Sample size dominates the interpretation. Over fifty picks, an ROI of 20% is unremarkable noise. Over five thousand, an ROI of 3% is a serious edge. Always read the two numbers together; either alone is close to meaningless.
For a worked example: MalluSports currently shows 10.5% ROI across 93 settled picks. That is a small sample by any standard, and the honest reading is that it is encouraging but not yet evidence. It is published with the per-market breakdown so the weak markets are visible too.
One more caveat: ROI computed at flat stakes is comparable across services. ROI computed with variable or 'confidence-weighted' stakes is not, because the weighting can be tuned after the fact.
MalluSports is a free prediction and analysis service only. It accepts no bets, holds no funds, operates no gambling platform and holds no gambling licence. Betting carries real financial risk. 18+ · begambleaware.org