Beyond Sharpe: 6 Metrics That Reveal a Strategy’s True Risk

Three strategies, backtested over 20 years of real market data, all landed within the same Sharpe ratio band of 0.64 to 0.79. One lost 55% in a single drawdown. Another lost 21% and was invested only 79% of the time. The third never lost more than 28% but was fully exposed every day. The Sharpe ratio alone would never have told you which was which.

What Sharpe Actually Measures (and What It Doesn't)

The Sharpe ratio divides a strategy's excess return over the risk-free rate by its standard deviation. It rewards return and punishes volatility. That makes it the single most widely used measure of risk-adjusted performance in finance, and also the most over-relied.

The problem isn't that the measure is wrong. It's that it's narrow. Standard deviation treats every wiggle, up or down, as risk. A strategy that jumps 8% in a month gets the same penalty as one that drops 8%. The ordering of returns, the depth and duration of drawdowns, whether you're in the market during a crash or sitting it out, how correlated your returns are to everything else. None of that shows up in a single number.

A good Sharpe ratio is typically 0.8 to 1.0 for diversified long-term portfolios, and above 1.0 is excellent, according to XBTO's risk-adjusted return guide. But as we'll show, two strategies with nearly identical Sharpe ratios can deliver fundamentally different experiences to the investor holding them.

The Backtest: Three Strategies, Similar Sharpe, Very Different Risk

We ran three strategies on real market data from January 2006 through July 2026, a 20.5-year window covering the 2008 financial crisis, the 2020 COVID crash, and the 2022 bear market. All returns are net of transaction costs where applicable. The data source is Tiingo split- and dividend-adjusted daily closes, accessed through the RiskHarvest backtest engine.

The three strategies:

• Buy & Hold SPY — hold the S&P 500 ETF, never sell.
• SMA 200 Crossover — hold SPY when its price is above its 200-day simple moving average; move to cash when it's below. Transaction cost: 5 basis points per trade.
• 60/40 SPY/TLT — 60% S&P 500, 40% long-term Treasury bonds, rebalanced annually. Transaction cost: 5 basis points per trade.

Here's how $1 grew under each strategy:

And here are the full metrics, not just Sharpe but the six additional metrics that reveal where each strategy's risk actually lives:

Read across the Sharpe row and the strategies look like siblings: 0.64, 0.75, 0.79. Read down any other column and they're strangers. Here's what each row tells you that Sharpe can't.

1. What Does the Sortino Ratio Measure?

The Sortino ratio uses the same numerator as Sharpe (excess return) but replaces total standard deviation with downside deviation: the volatility of returns that fall below a minimum acceptable return, typically zero. Upside surprises don't count against you.

This matters because the Sharpe ratio penalizes strategies for producing large positive returns. A trend-following system that occasionally captures a 15% month gets dinged for it. The Sortino ratio recognizes that most investors don't lose sleep over upside surprises. As Optimized Portfolio's comparison notes, the Sortino ratio's key advantage is that it penalizes only downside volatility, making it better suited for trend-following and asymmetric strategies.

In our backtest, the gap tells the story. Buy & Hold SPY has a Sharpe-to-Sortino gap of 0.64 to 0.78, decent because a lot of its volatility is upside. The 60/40 portfolio's gap is wider: 0.79 to 1.06. The bonds smooth the downside, so the Sortino ratio rewards the 60/40 more than the raw Sharpe suggests. The 60/40 doesn't look like the best strategy on Sharpe alone, but it is on Sortino. That's the metric doing its job.

2. Why Does Maximum Drawdown Matter More Than Sharpe?

Maximum drawdown is the largest peak-to-trough decline a portfolio experienced. It is the single most important number for any investor who has a behavioral breaking point, and most do. A drawdown of 55% doesn't just mean your portfolio is worth less than half. It means you have to double your remaining capital just to get back to where you started.

The difference between -55% and -21% at the same Sharpe level is the difference between an investor who holds through the pain and one who sells at the bottom. Maximum drawdown captures what Foxholm Financial calls the "worst-case actual loss experience" in a way that standard deviation cannot, because standard deviation averages across all periods while drawdown focuses on the specific period that matters most: the one that tests your fortitude.

Buy & Hold SPY hit -55.2% during the 2008–2009 financial crisis. The SMA 200 strategy, which moved to cash when the trend broke, never lost more than 21.0%. The 60/40 portfolio, cushioned by Treasuries, peaked at -28.5%. Same Sharpe neighborhood. Completely different drawdown profiles.

3. What Is the Calmar Ratio?

The Calmar ratio divides annualized return by the absolute value of maximum drawdown. It tells you how much return you earned for every percentage point of your worst decline. A Calmar of 2.0 means you earned twice your worst drawdown in annual returns. Below 1.0 means your worst drawdown exceeded your annual. It is particularly useful for evaluating strategies that use leverage, concentrated positions, or illiquid assets, where the maximum loss can be far more severe than day-to-day volatility suggests.

In our results, Buy & Hold SPY has a Calmar of 0.20: it earned 11% per year but lost 55% at the worst point. The SMA 200 strategy's Calmar of 0.40 is double that, because it cut the drawdown by more than half while giving up only about 2.6 percentage points of annual return. The 60/40 portfolio sits at 0.27; the bonds helped with drawdown but also dragged on returns.

The Calmar ratio has a known weakness worth stating: it depends on a single data point (the worst drawdown in the sample). If your backtest period doesn't include a major crash, Calmar will look artificially high. That's why it should complement a long enough sample, not replace other metrics in a short one.

4. What Is the Ulcer Index?

The Ulcer Index, devised by Peter Martin and Byron McCann in 1987 and introduced in their 1989 book The Investor's Guide to Fidelity Funds, measures the depth and duration of drawdowns together. It calculates the root-mean-square of all drawdown values across the entire period, not just the worst one. A strategy that drops 30% and recovers in two weeks gets a lower score than one that drops 30% and takes three years to come back. The name is deliberately literal: it's meant to approximate the emotional cost of holding a position underwater.

In our backtest, the Ulcer Index tells a story that max drawdown alone misses. Buy & Hold SPY scores 0.126. The 60/40 portfolio scores 0.074, 41% lower. But the SMA 200 strategy scores 0.084, higher than the 60/40 despite having a smaller max drawdown. Why? Because the SMA strategy spent more time in drawdown territory: smaller dips but more frequent ones, and when whipsawed during choppy markets it can sit in mild drawdowns for long stretches. The Ulcer Index is designed to measure "real pain of drawdowns, not just magnitude," which is exactly what shows up here. Sharpe and max drawdown don't capture this.

5. How Much Time in the Market Is Your Strategy Actually Taking?

A strategy that's invested 100% of the time and one that's invested 79% of the time are taking fundamentally different amounts of risk, even if their Sharpe ratios are similar. Time in market measures the percentage of trading days a strategy held a position rather than sitting in cash.

The SMA 200 crossover strategy was invested roughly 79% of the time over 20 years. It spent 21% of days in cash, out of the market entirely, avoiding risk. That's a significant structural difference from a buy-and-hold approach that bears market risk every single day. If two strategies have the same Sharpe ratio but one takes risk 100% of the time and the other 79%, the one that's less exposed has more room to deploy capital elsewhere or simply carry less tail risk during the periods it's in cash.

Time in market also interacts with behavioral risk. A strategy that goes to cash during a downtrend can give an investor a psychological break, not just a financial one. The investor who sees "the strategy went to cash in October 2008 and didn't come back until April 2009" has a narrative that helps them stay with the plan. The buy-and-hold investor in the same period had to endure the full decline with no signal telling them when they'd return.

6. Does the Strategy Actually Diversify? (Correlation to Benchmark)

Correlation measures how closely a strategy's returns move with a benchmark. A correlation of 1.0 means they move in lockstep. A correlation of 0 means they move independently. For the purpose of portfolio construction, correlation is what determines whether adding a strategy to your portfolio actually reduces risk or just adds more of the same exposure under a different label.

In our results, Buy & Hold SPY has a correlation of 1.00 to the S&P 500, obviously, since it is the S&P 500. The SMA 200 strategy, despite trading the same ETF, has a correlation of 0.60: because its cash periods break the link with the index. The 60/40 portfolio sits at 0.72; the Treasury allocation reduces but doesn't eliminate the equity dependence.

Beta, the related measure, quantifies how much a strategy moves when the benchmark moves. SMA 200's beta of 0.36 tells you that when the S&P 500 drops 10%, you'd expect the SMA strategy to drop about 3.6% on average. Not because it's hedged, but because it's often not in the market when that drop is happening. The 60/40 portfolio's beta of 0.38 achieves a similar reduction by a completely different mechanism: diluting equity exposure with an asset that historically moves differently. Same end result in beta terms. Correlation reveals that they got there by different routes: one through timing, one through diversification. Sharpe doesn't capture that distinction either.

What Sharpe Is Still Good For

None of this means you should throw out the Sharpe ratio. It remains the most useful single-number summary of risk-adjusted return, useful for a first-pass screen across many strategies, and useful for its universality (every fund reports it, every platform computes it). If a strategy has a Sharpe below 0.3 over a long period, that's a real signal that the return doesn't justify the volatility.

The mistake is using Sharpe alone. It gets you into the room. The other six metrics decide whether you stay.

Where This Meets Our Work

At RiskHarvest, we never evaluate a strategy on a single metric. Every backtest we run produces the full set: Sharpe, Sortino, max drawdown, Calmar, and turnover. And we inspect the equity curve for drawdown depth and duration before we draw any conclusion. The reason is practical: a strategy that looks good on Sharpe but hides a 50%+ drawdown is useless to an investor who can't hold through it, and we'd rather find that out in a backtest than in real time.

The deeper principle is that risk is not one-dimensional. Volatility, drawdown exposure, time in market, correlation: these are different kinds of risk, and a complete evaluation requires looking at all of them. The Sharpe ratio compresses risk into one number, which is convenient but lossy. The cost of that compression is real. It hides the difference between a strategy you'll hold for 20 years and one you'll abandon in year 3.

The One Thing to Remember

If you take away one point, let it be this: a Sharpe ratio of 0.75 can belong to a strategy that lost 55% of your money or one that lost 21%. The number alone won't tell you which. Six other metrics will. Use them.

Educational content, not financial advice. All backtest results are based on historical data and do not guarantee future performance. Past performance, transaction costs, and the specific period analyzed all affect these metrics. Different periods and assumptions will produce different results.