Drawdown and recovery calculator

Depth is the number people quote. Duration is the number that ends strategies. This gives both for every episode in your series, with dates, and marks anything still underwater as open rather than closing it with a date it has not reached.

Worked example — Maximum drawdown, illustrative series: -51.3%. From Sep 2013 to Sep 2014, needing 105.2% to get back; recovery took 60 more months

Paste a price, balance, or return series

Add a date to each line to get durations in calendar time. The episode table, the underwater chart, and the Ulcer and Calmar ratios update as you type.

Try:

Illustrative decline and recovery: The page default: a real-shaped multi-year decline that later recovers, with dates on both halves of the episode.

Series, one period per line. Add a date to get durations in real time

2012-01-31, 99.94
2012-02-29, 101.08
2012-03-31, 101.19
2012-04-30, 103.31
2012-05-31, 104.49
2012-06-30, 104.56
2012-07-31, 105.87
2012-08-31, 106.52
2012-09-30, 106.20
2012-10-31, 109.09
2012-11-30, 107.37
2012-12-31, 107.62
2013-01-31, 109.13
2013-02-28, 110.46
2013-03-31, 110.37
2013-04-30, 110.17
2013-05-31, 111.89
2013-06-30, 110.87
2013-07-31, 110.96
2013-08-31, 111.42
2013-09-30, 113.03
2013-10-31, 110.29
2013-11-30, 104.10
2013-12-31, 95.64
2014-01-31, 89.45
2014-02-28, 84.70
2014-03-31, 78.82
2014-04-30, 74.36
2014-05-31, 69.41
2014-06-30, 65.38
2014-07-31, 61.64
2014-08-31, 58.35
2014-09-30, 55.08
2014-10-31, 56.15
2014-11-30, 56.76
2014-12-31, 58.41
2015-01-31, 60.46
2015-02-28, 61.15
2015-03-31, 61.82
2015-04-30, 62.87
2015-05-31, 63.75
2015-06-30, 66.70
2015-07-31, 65.79
2015-08-31, 67.49
2015-09-30, 67.02
2015-10-31, 68.83
2015-11-30, 69.93
2015-12-31, 70.63
2016-01-31, 71.61
2016-02-29, 72.13
2016-03-31, 73.43
2016-04-30, 73.77
2016-05-31, 76.56
2016-06-30, 76.77
2016-07-31, 77.89
2016-08-31, 75.73
2016-09-30, 79.07
2016-10-31, 79.51
2016-11-30, 78.67
2016-12-31, 79.28
2017-01-31, 83.11
2017-02-28, 84.24
2017-03-31, 84.22
2017-04-30, 83.69
2017-05-31, 83.81
2017-06-30, 83.11
2017-07-31, 86.38
2017-08-31, 85.99
2017-09-30, 88.29
2017-10-31, 89.45
2017-11-30, 87.69
2017-12-31, 89.31
2018-01-31, 90.28
2018-02-28, 89.15
2018-03-31, 93.73
2018-04-30, 92.96
2018-05-31, 95.37
2018-06-30, 93.48
2018-07-31, 96.77
2018-08-31, 98.37
2018-09-30, 98.68
2018-10-31, 99.58
2018-11-30, 98.74
2018-12-31, 98.49
2019-01-31, 102.93
2019-02-28, 103.06
2019-03-31, 102.31
2019-04-30, 105.03
2019-05-31, 106.22
2019-06-30, 107.40
2019-07-31, 110.33
2019-08-31, 111.27
2019-09-30, 114.46
2019-10-31, 116.13
2019-11-30, 116.49
2019-12-31, 120.47
2020-01-31, 115.80
2020-02-29, 114.82
2020-03-31, 112.76
2020-04-30, 108.66
2020-05-31, 105.11
2020-06-30, 103.39
2020-07-31, 108.11
2020-08-31, 111.46
2020-09-30, 110.67
2020-10-31, 114.07
2020-11-30, 116.38
2020-12-31, 120.58
2021-01-31, 121.86
2021-02-28, 125.02
2021-03-31, 126.59
2021-04-30, 130.55
2021-05-31, 129.42
2021-06-30, 134.54
2021-07-31, 131.89
2021-08-31, 134.31
2021-09-30, 134.11
2021-10-31, 136.03
2021-11-30, 137.33
2021-12-31, 141.80
2022-01-31, 141.19
2022-02-28, 141.16
2022-03-31, 142.79
2022-04-30, 145.13
2022-05-31, 146.78
2022-06-30, 146.43
2022-07-31, 146.49
2022-08-31, 148.75
2022-09-30, 150.25
2022-10-31, 148.40
2022-11-30, 151.58
2022-12-31, 151.11

132 values read with dates. The series on load is an illustrative path, not a real fund, index, or strategy. Replace it with your own.

The column holds

Frequency

Advanced options 2

Risk-free rate, annual (for Calmar/Martin)%Episodes to list

From Sep 2013 to Sep 2014 it fell -51.3% over 12 months. Getting back took 60 more months, to Sep 2019. Total time underwater: 72 months.

Current drawdown-0.31%

Gain needed from here0.31%

Time underwater75.8%

Ulcer index22.43

Calmar ratio0.075

Martin ratio0.083

Show gain-needed-to-recover overlay

Left axis: drawdown from the running peak. Right axis: the gain needed to recover from that point, computed at every period, not only at each episode's trough. The dots mark each episode's trough.

The worst episodes, with both halves of the underwater period, on the illustrative series still loaded
Peak	Trough	Depth	Gain to recover	Peak to trough	Trough to recovery	Total underwater
Sep 2013	Sep 2014	-51.27%	105.2%	12 months (365 days)	60 months (1826 days)	72 months
Dec 2019	Jun 2020	-14.18%	16.5%	6 months (182 days)	6 months (184 days)	12 months
Jun 2021	Jul 2021	-1.97%	2.0%	1 months (31 days)	3 months (92 days)	4 months
Oct 2012	Nov 2012	-1.58%	1.6%	1 months (30 days)	2 months (62 days)	3 months
Sep 2022	Oct 2022	-1.23%	1.2%	1 months (31 days)	1 months (30 days)	2 months

An episode runs from the last high, through the trough, to the month the old high is regained. Anything still below its peak at the end of your series is marked open and given no recovery duration, because it does not have one yet. Calmar divides the 3.86% compound growth over 10.9 years by the maximum drawdown; Martin divides the excess of that growth over your 2% cash rate by the Ulcer index.

At the compound rate this series itself produced, climbing out of the current -0.3% would take about 1.0 more months.

How much do I need to gain back a 30% loss?

A 30% loss needs a 42.9% gain to get back, not 30%. The required gain is 1 / (1 + d) minus 1, and the asymmetry widens fast past 30% because the denominator keeps shrinking while the numerator does not.

Gain required to recover, by loss size
Loss	Gain needed to recover
5%	5.26%
10%	11.11%
15%	17.65%
20%	25.00%
25%	33.33%
30%	42.86%
40%	66.67%
50%	100.00%
60%	150.00%
70%	233.33%
80%	400.00%
90%	900.00%

What is the Ulcer index?

The Ulcer index is the root mean square of the whole drawdown path, not just its lowest point, so it charges a strategy for depth and for time spent underwater, not depth alone. On this page's illustrative series the Ulcer index is 22.43 against a maximum drawdown of -51.3%: the closer those two numbers are, the longer the series spent near its worst point rather than passing through it quickly. A shallow, permanent drawdown can score a worse Ulcer index than a violent, quick one, which is the opposite of what the maximum drawdown alone would suggest.

How long does it take to recover from a drawdown?

On this page's illustrative series, the worst episode fell -51.3% from Sep 2013 to Sep 2014, then took 60 more months to regain the old high, on Sep 2019. That is the actual path; the calculator also converts any open drawdown into a time estimate at the compound rate the series itself produced, which is the honest version of "it always comes back": it comes back on a clock, and the clock is the part nobody budgets for.

What JMM thinks: time underwater is what ends strategies

Investors do not usually capitulate at the bottom. They capitulate eighteen months later, when the position is still down 20%, the thesis still sounds fine, and the money has been doing nothing while something else went up. That is why the Ulcer index is worth more attention than the maximum: a strategy with a shallow permanent drawdown scores worse than one with a violent quick one. If you are choosing between two return streams with similar Calmar ratios, take the one with the lower Ulcer index and the shorter longest underwater period. Drawdowns are measured on the closing values you paste, so an intraperiod low deeper than any close will not appear, and a maximum drawdown is a record from one sample, not a bound on the next one. To pair this with a forward-looking risk estimate, use the portfolio volatility calculator, and to test whether the return justified the drawdown, the Sharpe ratio calculator. In a retirement plan this is the history half of the stress test; the FIRE calculators hub pairs it with the forward-looking sequence risk step.

The next question after this one

Sharpe ratio calculatorCalculatorDepth is one dimension. This adds the return earned per unit of risk, with an interval around it.Open next

What Pro adds here

Conditional drawdown at risk, drawdown duration distributions, and side-by-side comparison of two pasted series are Pro features. The launch list sends one email at launch.