Every so often you'll hear someone say "I beat the market this year" — and it's a genuinely tricky claim to check. If your portfolio returned 20% while the ASX 200 returned 12%, did you actually do something clever? Or did you just hold a riskier mix of stocks that was always going to swing harder in both directions, and this year it happened to swing up?
Jensen's Alpha is the number built specifically to answer that question properly.
What it actually measures
Jensen's Alpha compares your actual return to the return you'd have been expected to get, given how much market risk your portfolio was carrying. If your actual return is higher than what was expected for your risk level, that gap — the alpha — is the part that isn't explained by simply taking on more risk. It's the closest thing on this dashboard to a genuine "skill" number.
The "expected return for your risk level" part relies on a concept called beta, which measures how sensitive your portfolio is to market moves (we'll cover beta properly in the next post). For now, the short version: a beta of 1.0 means your portfolio tends to move in line with the market. A beta of 1.5 means it tends to swing 50% harder than the market in both directions. Jensen's Alpha uses beta to work out what return you should have gotten, then checks whether you actually beat that.
For the technically curious
Where:
- Rₚ — your portfolio's actual return
- R_f — the risk-free rate
- βₚ — your portfolio's beta (its sensitivity to market moves)
- Rₘ — the market/benchmark's return
The part in the square brackets — R_f + βₚ(Rₘ − R_f) — is the expected return under the Capital Asset Pricing Model (CAPM): the risk-free rate, plus your beta multiplied by however much the market beat the risk-free rate. Alpha is simply your actual return minus that expected figure.
How to read the number
- Positive alpha — you outperformed what your risk level would predict. This is the number people mean when they say they "beat the market."
- Zero alpha — you got exactly what your risk level predicted. No better, no worse than a passive exposure to the same amount of market risk.
- Negative alpha — you underperformed what your risk level would have predicted. You took on the risk without getting properly compensated for it.
Unlike Sharpe or Sortino, alpha is measured in the same units as return itself (a percentage), which makes it fairly intuitive to read directly — a +5% alpha means you beat your risk-adjusted expectation by 5 percentage points.
Why it's a genuinely different question than the other ratios
Sharpe, Sortino, and Calmar all measure efficiency — how much return you got for the risk you took, using your own portfolio's numbers in isolation. Jensen's Alpha does something different: it explicitly compares you against a benchmark, using a formal model of what "fair" compensation for your risk level should look like.
That makes alpha the number best suited to answering "am I actually good at this, or did I just get lucky/take on more risk than the index?" — which is a fundamentally different (and arguably more interesting) question than "was my ride smooth relative to my return."
The one big thing to watch out for
Alpha is only as good as the beta and benchmark it's built on — and both of those can be shakier than they look, especially for a concentrated or unusual portfolio.
Beta assumes a stable, linear relationship between your portfolio and the market, which works reasonably well for a diversified, broadly-market-like portfolio, and much less well for a small, concentrated portfolio dominated by a handful of high-conviction bets in one or two sectors. If your portfolio doesn't really behave like "a leveraged or de-leveraged version of the index," the whole "expected return" calculation alpha is built on becomes a rougher approximation — which means a large alpha number, positive or negative, should be read as "this was a genuinely unusual period relative to a simplified model," not as hard proof of skill or its absence.
A big positive alpha over a short period, especially from a concentrated portfolio, is usually telling you more about a handful of stocks having a great run than it is about durable, repeatable investing skill. The longer the period and the more diversified the portfolio, the more the number actually starts to mean what it claims to mean.
This is part 5 of a series on the risk numbers behind your portfolio. Next up: Beta itself — the number that measures exactly how jumpy your portfolio is compared to the market.