Portfolio Risk Explained: Volatility, Beta, Sharpe Ratio and Maximum Drawdown

Most investors can tell you what their portfolio returned last year. Far fewer can tell you how much risk they took to earn it, and that second number is the one that decides whether the strategy survives the next bad market. Understanding portfolio risk metrics means learning to read four numbers, volatility, beta, the Sharpe ratio and maximum drawdown, and knowing what each one does and does not tell you. This guide explains them in plain language, with the traps that make each one misleading if you take it at face value.

Key Takeaways

Metric	What It Actually Tells You
Volatility (standard deviation)	How widely your returns swing around their average. A rough measure of bumpiness, not of danger.
Beta	How much your portfolio tends to move when the market moves. Above 1 is more sensitive, below 1 is less.
Sharpe ratio	Return earned per unit of volatility taken. Useful for comparing two strategies fairly.
Maximum drawdown	The worst peak-to-trough fall you have lived through. The most emotionally honest number of the four.
Correlation	Whether your holdings actually diversify each other, or just look different on the surface.
Which matters most?	Drawdown, for most private investors, because it is the number that makes people sell at the bottom.

1. Why Return Alone Is a Half-Finished Sentence

Two portfolios both return 9% a year over a decade. The first drifted upward with a worst fall of 12%. The second doubled twice and halved once along the way, with a worst fall of 55%. On a returns table they look identical. In real life they are not remotely the same investment, because only one of them is a portfolio you would still be holding at the end.

Risk metrics exist to complete the sentence. Return tells you what happened. Risk tells you what the range of things that could have happened looked like, and how close you came to abandoning the plan. If you have already read our guide on measuring your real investment returns, this article is the other half of that story.

Return is what you got. Risk is what you were exposed to. Judging a strategy on the first without the second is like judging a driver on how fast they arrived.

2. Volatility: The Bumpiness Number

Volatility, usually reported as the annualised standard deviation of returns, measures how much your returns scatter around their own average. A portfolio with 10% annualised volatility has, very roughly, spent most of its months within 10 percentage points either side of its average annual return.

How to read it

• Under 10% is typical of a diversified portfolio with a meaningful bond or cash allocation.
• Roughly 15% to 18% is in the range broad global equity indices have historically occupied.
• Above 25% usually means concentration, small companies, single-country bets, or crypto in the mix.

The three traps

It treats upside and downside the same. A holding that jumps 20% in a month adds as much to measured volatility as one that falls 20%. Nobody has ever complained about upside volatility.

It assumes a tidy distribution. Real markets have fat tails, meaning extreme moves happen far more often than a neat bell curve predicts. Volatility systematically understates how bad the worst week can be.

It is backward looking. Volatility is calm right up until it is not. Long quiet periods produce low readings that lull investors into taking on more risk just before conditions change.

3. Beta: How Much of the Market You Are Riding

Beta measures how sensitive your portfolio has been to moves in a chosen benchmark. A beta of 1.0 means you have historically moved roughly in line with the market. A beta of 1.4 means a 10% market fall has typically come with a 14% fall for you. A beta of 0.7 means you have taken about 70% of the ride, in both directions.

Beta is useful for one specific question: how much of my result is just the market, and how much is me? An investor delighted with a 30% year should check their beta before concluding they are skilled. If the market rose 25% and your beta is 1.3, the market did the work.

The traps

• Beta is meaningless without naming the benchmark. Beta against the S&P 500 and beta against a global index are different numbers, and a UK-heavy portfolio measured against a US index will produce nonsense.
• It only captures market-linked risk. A portfolio of five unrelated small companies can have a low beta and still be extremely risky, because most of its risk is specific to those companies.
• It moves over time. A portfolio’s beta drifts as its composition and the market’s composition change.

4. The Sharpe Ratio: Return Per Unit of Discomfort

The Sharpe ratio divides your excess return, meaning your return above the risk-free rate such as short-term government bills, by your volatility. It answers a genuinely useful question: how much return did you earn for each unit of bumpiness you accepted?

An illustrative example. Portfolio A returns 12% with 20% volatility. Portfolio B returns 9% with 10% volatility. With a risk-free rate of 4%, A scores (12 minus 4) divided by 20, which is 0.40. B scores (9 minus 4) divided by 10, which is 0.50. B produced less return but did more with the risk it took.

How to read it

• Below 0.5 is unremarkable, and typical of broad equity exposure over many periods.
• Around 0.5 to 1.0 is a decent risk-adjusted result over a full cycle.
• Consistently above 1.5 over long periods for a private portfolio usually means the measurement window is too short, or the risk is hiding somewhere the volatility figure cannot see.

The traps

Because the denominator is volatility, the Sharpe ratio inherits every one of volatility’s weaknesses. It penalises strong upside months. It flatters strategies that produce small steady gains punctuated by rare catastrophic losses, precisely the shape that looks best on a spreadsheet and worst in reality. And it is highly sensitive to the period you measure, so a Sharpe ratio computed over 12 months tells you almost nothing.

Any risk-adjusted measure built on volatility rewards strategies that are quiet most of the time. Ask what the strategy does in the 5% of months that are not quiet.

Sortino: the sensible cousin

The Sortino ratio fixes the most obvious flaw by using only downside deviation in the denominator, so upside moves are not treated as risk. If you have the choice, Sortino answers the question most investors actually meant to ask.

5. Maximum Drawdown: The Number That Predicts Behaviour

Maximum drawdown is the largest peak-to-trough fall in your portfolio value over a period, expressed as a percentage. If your portfolio peaked at £100,000, fell to £62,000 and later recovered, your maximum drawdown was 38%.

Of the four metrics, this is the one to take most seriously, for a simple reason: it is the number that describes the experience that makes people abandon their plan. Nobody sells at the bottom because their standard deviation was uncomfortable. They sell because they have watched a third of their money disappear over eight months and cannot stand another day of it.

Two companion numbers

• Drawdown duration. How long the fall lasted from peak to trough. A fast crash is easier to sit through than a two-year grind.
• Recovery time. How long it took to get back to the previous peak. This is frequently far longer than the fall itself, and it is where most of the psychological damage happens.

The recovery maths nobody likes

Losses and gains are not symmetric. A 20% fall needs a 25% gain to get back to level. A 33% fall needs 50%. A 50% fall needs 100%. This asymmetry is the entire practical argument for controlling drawdown rather than chasing the highest possible return, and it is why concentration risk deserves the attention we gave it in the guide to rebalancing without overtrading.

6. Correlation: Whether Your Diversification Is Real

The four metrics above describe your portfolio as a whole. Correlation explains why those numbers came out the way they did.

Correlation runs from +1, meaning two holdings move in lockstep, through 0, meaning no relationship, to -1, meaning they move in opposite directions. The benefit of diversification comes entirely from holding things whose correlation is meaningfully below 1.

The uncomfortable finding for a lot of investors is that a portfolio which looks diversified on a holdings list is often not diversified at all. Ten different technology-adjacent stocks, a US index fund and a growth-tilted global fund can easily behave as one position with a fancy label. And correlations have a cruel habit of rising toward 1 in exactly the crises when you needed them low.

Diversification is not measured by counting your holdings. It is measured by whether they fall at the same time.

7. Putting the Four Together

No single metric is sufficient, but read together they tell a coherent story. Work through them in this order:

• Start with maximum drawdown. Ask honestly whether you would have held on through the worst period shown. If the answer is no, nothing else matters, because you would not have been there for the recovery.
• Then look at volatility to understand the day-to-day texture, which determines how often you will be tempted to interfere.
• Then beta, to see how much of your outcome was simply market exposure you could have bought far more cheaply.
• Then Sharpe or Sortino, to compare two genuine alternatives on a like-for-like basis.
• Finally correlation, to work out where the risk is actually concentrated and what to change.

Two rules keep this honest. Measure over a period long enough to include at least one meaningful market fall, because a metric computed over a calm three years is a description of the weather rather than the climate. And always compare against something, whether a benchmark or your own portfolio a year ago. A Sharpe ratio of 0.6 in isolation means nothing.

8. How InvestInsight Surfaces Your Risk

The obstacle to any of this is not the arithmetic. It is that the inputs, a complete daily value history across every account you hold, are scattered across brokers that each show you only their own slice.

InvestInsight’s portfolio tracker builds that combined history for you. Because every holding across every account rolls up into one valuation series, the portfolio-level analytics describe the thing you actually own rather than one broker’s corner of it. Drawdowns show on the performance chart as what they were, with the depth and the recovery both visible, which is far more instructive than a single summary percentage.

The allocation and concentration views answer the correlation question in the way most private investors need it answered, by showing weight by sector, geography and individual position. The AI concentration insight goes further and flags when a single holding or a cluster of related holdings has quietly become the dominant driver of your results. That is usually the finding that changes behaviour, because it turns an abstract risk number into a specific position you can do something about.

9. Common Mistakes to Avoid

• Measuring over too short a window. Any risk metric computed over less than a full market cycle is close to noise.
• Comparing against the wrong benchmark. Beta and relative performance are only meaningful against something with a similar mandate.
• Treating low volatility as low risk. Illiquid or thinly traded assets show flattering volatility precisely because they reprice rarely.
• Optimising for the ratio instead of the outcome. Strategies tuned to maximise a backward-looking Sharpe ratio tend to be fragile.
• Ignoring your own tolerance. The best portfolio on paper is worthless if its drawdown profile makes you sell in month seven.
• Counting holdings as diversification. Thirty correlated positions are one position with extra paperwork.

Conclusion

Portfolio risk metrics are not academic decoration. They are the difference between a strategy you understand and a strategy you are merely hoping about. Volatility tells you how bumpy the ride has been, beta tells you how much of it was simply the market, the Sharpe ratio lets you compare two options fairly, and maximum drawdown tells you the thing that matters most, which is how bad it has actually got and whether you would have stayed.

Start with drawdown, be sceptical of any metric that looks too good, and measure over a period that includes a genuine fall. If you want to see these numbers for the portfolio you actually hold rather than one account at a time, InvestInsight’s portfolio tracker brings every account into one performance and allocation view, so the risk you are running is visible before the market shows it to you.

Further reading

Risk metrics are descriptive, not predictive. They summarise what has happened, which is useful context but never a guarantee about what comes next.

• Investopedia: the Sharpe ratio
• Investopedia: maximum drawdown
• Investopedia: beta

This article is for general information only and does not constitute financial advice. All figures are illustrative. Past performance and past risk measures are not reliable indicators of future results. Consider your own circumstances or speak to a qualified adviser before making investment decisions.