Ask a trader what their worst month was and they will tell you straight away. Ask how many months they spent below their last equity high, and most will have to go and check. The pain index in trading is built for that second question. It takes every drawdown in a return series, the small ones and the long grinding ones, and squeezes them into a single figure that says how much time your money spent underwater, and how far under.
We have already written about maximum drawdown, which records only the single worst peak-to-trough fall. That number is useful, but it is blunt. Two accounts can share the same maximum drawdown and give their owners very different years. The pain index tries to capture the part maximum drawdown leaves out.
In this article
What the pain index in trading measures
The measure comes out of Zephyr Associates and is attributed to Thomas Becker. Zephyr’s own StatFACTS guide describes it as a proprietary risk metric that quantifies the capital preservation tendencies of a manager or index, measuring the depth, duration and frequency of periods of losses.
Those three words are the whole idea:
- Depth: how far below the previous peak the account fell.
- Duration: how long it stayed below that peak before recovering.
- Frequency: how often it slipped under water in the first place.
A strategy that drops 5% and bounces back next month barely moves the needle. One that sits 5% under its high for a year adds a lot more, even though its worst reading never got any deeper. That is exactly the kind of slow bleed that wears traders down and pushes them into abandoning a system at the wrong time.
The formula, without the jargon
Start with a drawdown series. For each period (a day, a week or a month), compare the account value with the highest value it has reached so far. If the account is at a new high, the drawdown for that period is zero. If it is below the high, the drawdown is the percentage gap.
The pain index is then simply the average of those drawdowns, taken as positive numbers, across every period in the window:
Pain Index = (|D1| + |D2| + … + |Dn|) ÷ n
Here Di is the drawdown from the previous peak in period i, and n is the total number of periods. That is how the open-source PerformanceAnalytics package for R implements it, citing both Zephyr and Carl Bacon’s book Practical Portfolio Performance Measurement and Attribution.
Two details in that definition matter more than they look.
First, the divisor is the total number of observations, not the number of drawdown periods. The PerformanceAnalytics documentation calls this out directly: it is what separates the pain index from a plain “average drawdown”. Months at a new high count as zero and pull the figure down, so an account that is rarely underwater gets rewarded.
Second, the drawdowns are not squared. The same documentation notes that this is the difference from the Ulcer Index, which squares each drawdown before averaging and then takes the square root. Squaring punishes deep holes much harder than shallow ones. The pain index treats a percentage point of drawdown the same whether it comes from a shallow dip or a deep one.
Maximum drawdown tells you how bad the worst day of the storm was. The pain index tells you how much of the year you spent getting rained on.
There is also a neat visual reading. If you plot the drawdown curve below a zero line, the pain index is the shaded area between that curve and the line, divided by the length of the period. Zephyr’s guide describes it in exactly those terms: it is the shaded area on the drawdown graph, and the investor hopes that area is as small as possible.
A worked example with two strategies
Let us make this concrete. Below are two made-up strategies, both starting at ₹100 and tracked at month-end for twelve months. The figures are illustrative only. They are not a real stock, fund or index.
- Strategy A runs up to 108, takes one sharp hit to 94, then recovers and ends the year at 126.
- Strategy B peaks at 103 in the first month, slides slowly to 95 over six months, then climbs back to finish at 110.
| Month | A value | A drawdown | B value | B drawdown |
|---|---|---|---|---|
| 1 | 104 | 0.00% | 103 | 0.00% |
| 2 | 108 | 0.00% | 101 | −1.94% |
| 3 | 94 | −12.96% | 99 | −3.88% |
| 4 | 100 | −7.41% | 97 | −5.83% |
| 5 | 106 | −1.85% | 96 | −6.80% |
| 6 | 110 | 0.00% | 95 | −7.77% |
| 7 | 113 | 0.00% | 97 | −5.83% |
| 8 | 116 | 0.00% | 99 | −3.88% |
| 9 | 119 | 0.00% | 101 | −1.94% |
| 10 | 122 | 0.00% | 104 | 0.00% |
| 11 | 124 | 0.00% | 107 | 0.00% |
| 12 | 126 | 0.00% | 110 | 0.00% |
Take Strategy A’s month 3. The peak so far is 108 and the value is 94, so the drawdown is 94 ÷ 108 − 1, or about −12.96%. Every other figure in the table comes from the same step.
Now add them up.
- Strategy A: 12.96 + 7.41 + 1.85 = 22.22. Divide by 12 months and the pain index is about 1.85%. Maximum drawdown is 12.96%, and the account spent 3 of the 12 months below a previous high.
- Strategy B: 1.94 + 3.88 + 5.83 + 6.80 + 7.77 + 5.83 + 3.88 + 1.94 = 37.87. Divide by 12 and the pain index is about 3.16%. Maximum drawdown is only 7.77%, but the account spent 8 of the 12 months underwater.
Here is the interesting part. If you ranked these two on maximum drawdown alone, B looks safer, since its worst fall is roughly 5 percentage points smaller. The pain index flips that ranking. B’s losses were milder, but they dragged on for most of the year, and that shows up as a pain index about 70% higher than A’s.
For comparison, the Ulcer Index on the same numbers is about 4.34% for A and 4.20% for B. Because it squares each drawdown, A’s single deep month weighs heavily and the two strategies end up almost level. Same data, three risk measures, three slightly different stories. That is why we would never lean on just one of them.
From pain index to pain ratio
On its own, the pain index only describes risk. Zephyr’s guide is blunt about the limitation: the pain index completely ignores the upside. An account sitting in cash would have a pain index of zero, and nobody trades to earn a savings-account return.
So Zephyr paired it with a return measure. As set out in a Swan Global Investments note by Marc Odo, a former Zephyr research director, the pain ratio was developed by Thomas Becker and Aaron Moore of Zephyr. It takes the annualised return of the strategy, subtracts the annualised return of a cash-equivalent series, and divides the result by the pain index:
Pain Ratio = (Annualised return − Annualised cash return) ÷ Pain Index
If that looks like the Sharpe ratio, it should. Zephyr’s own write-up on drawdown risk makes the same comparison: the only difference is the denominator. The Sharpe ratio uses standard deviation, while the pain ratio uses the pain index. Higher is better in both cases.
Back to our example. Assume, purely for illustration, a cash return of 6.5% over the same year. Strategy A made 26%, so its excess return is 19.5 points, and 19.5 ÷ 1.85 gives a pain ratio of about 10.5. Strategy B made 10%, an excess of 3.5 points, and 3.5 ÷ 3.16 gives roughly 1.1. On this measure A is far more efficient: it earned much more for every unit of underwater time.
A lower pain index is always nicer to live with, but it only earns its keep when the return on top of it is worth the wait.
Where the pain index fits among drawdown measures
The pain index belongs to a small family of drawdown-based tools, and it helps to know how they differ:
- Maximum drawdown captures only the deepest fall. Zephyr’s guide notes it says nothing about how long recovery took, or about the smaller losses along the way.
- Ulcer Index, devised by Peter Martin in 1987 according to Wikipedia’s entry, squares drawdowns, so deep falls dominate. Its return-adjusted cousin is the Martin ratio, or ulcer performance index.
- Sterling and Burke ratios, along with Calmar, keep the Sharpe-style excess return on top but swap standard deviation for a drawdown-based measure, as the Wikipedia entry on drawdowns describes.
- Pain index averages every period’s drawdown in a straight line, with zeros included, so it rewards strategies that spend little time below their highs.
None of these is the “right” one. Each answers a slightly different question about the same equity curve.
How to use it on your own trades
You can compute the pain index in a spreadsheet in a few minutes. Put your account value or cumulative equity for each period in column A. In column B, keep a running maximum of column A. In column C, calculate A ÷ B − 1, which gives zero at new highs and a negative number otherwise. The pain index is the average of the absolute values in column C.
A few practical points from using it:
- Compare like with like. Zephyr’s guide says there is no hard-and-fast breakpoint for a good pain index. It should be judged against an appropriate benchmark or peer group, over the same window.
- Keep the frequency fixed. Daily data will catch intra-month dips that month-end data misses, so a daily pain index and a monthly one are not directly comparable.
- Watch the window. A strategy measured only over a rising market will show a flattering pain index. Include at least one rough patch before trusting the number.
- Use it for sizing decisions. If your pain index is creeping up while returns stay flat, that is a sign to check position sizes before the next drawdown, not after it.
It slots neatly into a wider risk management plan for trading. Stops and position sizing decide how deep a drawdown can go. The pain index tells you, looking back, how much of your trading life those drawdowns actually took up. For a lot of traders, that second number is the one that decides whether they stick with a system.
Disclaimer: This article is for educational purposes only and is not investment advice. The account values, returns and cash rate used in the example are illustrative and do not represent any real security, fund or strategy. Past performance does not guarantee future results. Please consult a SEBI-registered investment adviser before making investment decisions.





Leave a Reply
Join the conversation
Commenting is open to registered readers. Sign up with just your email — no password needed, and it takes under a minute.
Sign up to comment Log in