BENCHMARKING
Are you beating the market, or just surviving selection bias?
Most performance claims are polluted by survivorship bias, cherry-picked periods, and invalid benchmarks. Learn how to compare fairly and recognize the biases that distort the industry.
– ALPHA
To beat the market is to produce alpha: return above what a passive benchmark would have delivered for the same risk. The widely quoted SPIVA finding that most active funds fail to beat their benchmark is technically true but misleading, because it compares returns net of fees against a fee-free index. Gross of fees, skilled managers often do add alpha; fees then capture most of it. A fair comparison matches the benchmark to the strategy and adjusts for risk, rather than looking at raw return alone.
Most performance you see has already been filtered before you ever compare it, which leaves a survivor's view that flatters the average.
Test whether your edge is real alpha or luck with the Deflated Sharpe Ratio calculator.
The SPIVA-style statistic is technically true but misleading: it compares returns net of fees against a fee-free index, over periods and universes affected by survivorship and selection bias. Gross of fees, skilled managers often generate alpha; fees then consume most of it. The headline number says more about cost and methodology than about skill.
It is the distortion that appears when funds or traders that failed are removed from a dataset. Only the survivors remain, so the average looks far better than it really was. It is a major reason self-reported track records overstate true performance.
Passive management tracks an index at low cost; active management tries to beat it through selection and timing. What matters is whether the active manager's extra return (alpha) exceeds the extra fees and risk, which is why benchmarking must be risk-adjusted and net of fees.
Match it to the strategy's asset class, risk, and exposure. A market-neutral crypto strategy should not be measured against the S&P 500. The wrong benchmark can make a mediocre strategy look brilliant or a strong one look weak, so benchmark choice is itself a source of bias.
Over a short window, even a zero-skill strategy can post a high return by chance. The Deflated Sharpe Ratio estimates the probability that a result is genuine rather than the product of multiple testing and a small sample. A statistically significant, risk-adjusted edge against a fair benchmark is what separates skill from luck.
90% of funds don't beat the market. Or do they? Deconstructing SPIVA methodology.
Why self-reported performance is unreliable. How survivorship bias and cherry-picking distort the picture.
Statistical significance in trading. Use the Deflated Sharpe calculator to test if your performance is skill or luck.
NEXT STEP
Don't trust claims. Verify them. See how AuditZK makes performance independently auditable.
Related Reading