Benchmark: choosing a fair comparison for a result
Separate a return, a percentage-point gap and a ratio of ending wealth.
A benchmark is a reference selected in advance. Growth by itself does not show how a result compares with the relevant market. Choose a reference that addresses the same task, rather than choosing it after seeing which comparison wins.
Align the comparison conditions
Record market and risk exposure, start and end dates, currency, reinvestment and cash flows. A price index and a total-return index are different references. Another market’s chart can provide context without automatically being a fair performance benchmark.
Identify costs explicitly. An account after fees may be compared with a theoretical cost-free index, but the different assumptions must be visible. An index is usually not an executable portfolio. With deposits, a simple change in account balance is not the investment return.
Two descriptions of one shortfall
Use a hypothetical initial 100, one period, one currency and no external flows. Ending 109 is already after included costs. The selected theoretical benchmark ends 112, reinvests distributions and has no execution costs. The comparison states these assumptions openly.
Returns 9% and 12% differ by−3 percentage points. Ending-wealth ratio 109/112−1≈−2.6786% answers another question: how much lower the final wealth is relative to the benchmark. Calling both values simply “minus 3%” is imprecise.9/12−1 would compare gains rather than ending wealth.
Our own teaching example; all figures are hypothetical.
9% − 12% = −3 pp; 109 / 112 − 1 ≈ −0.0267857143 = −2.67857143%
| Series | Start | End | Return |
|---|---|---|---|
| Result after costs | 100 | 109 | 9% |
| Theoretical benchmark; no execution costs | 100 | 112 | 12% |
| Ending-wealth ratio | 1 | 109/112≈0.973214 | −2.6786% |
Limit the conclusion and preserve the method
One period proves neither skill, a lasting edge nor future returns. Risk, drawdown and data availability also matter. Do not replace the benchmark retrospectively without recording why; otherwise the comparison can become selection of a convenient answer.
Record method, dates, currency, index type, flows and included costs before evaluating results. Elliott does not measure brokerage-portfolio performance or promise to beat a market. The journal helps keep the original hypothesis separate from later observations.
Account 100→95, benchmark 100→90: the account is ahead by 5 percentage points. Does that mean a profit?
No. The account return is−5%, the benchmark−10%; the gap is+5 percentage points. Relative outperformance does not remove an absolute loss.
Sources
Sources checked: 2026-10-21
This material and its translations were prepared with AI assistance. Educational calculations were checked programmatically. Sources explain concepts; examples and files were created for this article.
Related reading
Next step
Record the chosen reference and comparison conditions before the next result becomes known.
A market-scenario journal: test an idea without hindsight →