Benchmark-Relative Versus Absolute Evaluation

A long-only equity strategy returns 8% for the year. Is that good? An absolute evaluation says yes — positive return, money made. A benchmark-relative evaluation asks a different question: the S&P 500 returned 15% that year, so the strategy lagged the market by 7 percentage points despite the positive headline number. Both framings are correct answers to different questions, and mixing them up is one of the most common ways a performance review misleads its audience.

Two different questions

Absolute evaluation measures the strategy's own return, volatility and drawdown in isolation — did it make money, and how bumpy was the ride getting there? This is the natural lens for a strategy that could otherwise be cash: a market-neutral fund, an absolute-return mandate, or a retail trader comparing a strategy to simply not trading.

Benchmark-relative evaluation measures active return — the strategy's return minus a chosen benchmark's return — and risk-adjusts it with the information ratio, active return divided by the volatility of that difference (tracking error). This is the natural lens whenever the alternative to the strategy is holding the benchmark itself, which is the normal situation for a long-only mandate: an investor comparing your fund to an index fund cares about the difference, not the absolute number in isolation.

$$\text{active return} = R_{\text{strategy}} - R_{\text{benchmark}}$$

In plain English: how much extra return did the strategy generate for taking on the job of active management, over and above what a passive index holding would have delivered for free.

Worked example: same 8%, two verdicts

Strategy A returns 8% against a benchmark that returned 15%: active return is $8\% - 15\% = -7\%$ — underperformance, despite a positive absolute number. Strategy B, a market-neutral strategy with no natural benchmark beyond cash, returns 8% against a 0% cash rate: active return is $+8\%$ — clear outperformance of the relevant alternative. The absolute number is identical; the relative verdict is opposite, because the two strategies are being judged against the correct alternative use of the money in each case.

What this means in practice

Choose the evaluation frame based on what the capital would otherwise be doing, not habit. A mandate that promises to beat an index needs benchmark-relative evaluation as the primary lens, with absolute numbers as context. A mandate that promises an absolute-return profile — hedge funds marketed on Sharpe ratio, market-neutral books, most quant prop strategies — should be judged absolutely, with a benchmark comparison used only to check the strategy isn't secretly just replicating market beta. Reporting the wrong frame, deliberately or not, is a classic way to make a mediocre result look better than it is: a strategy that merely tracks the market with extra fees looks fine in absolute terms and terrible relative to a free index fund.