The S&P 500 just experienced a 20% peak to trough drawdown in rapid fashion. It’s the third fastest pullback from an all time high in history, lagging only behind 1929 and 1987.
What a minute. That’s wrong? It was much more than 20%. And many, many people have written about how this the the fastest bear market ever and the quickest, deepest pullback in history.
Maybe we’re both right?
I used monthly data for my statement above. The articles you have seen stating the coronavirus market is the worst ever use daily returns. Since you lived through the last couple months, you know fully well the picture I painted above doesn’t do our recent experience justice. The articles provide a far more accurate description of what happened.
The S&P 500 fell 34% peak to trough using daily data. Thirty four percent is a lot deeper, a lot more painful, than 20%. My statistics are “correct”, but misleading.
You Live Daily Returns
Most people experience daily returns. Maybe you’re one of the lucky few that only checks your investments every few months. Problem is, when the media is talking about stock losses (and they always do during a crash), you are likely to wonder how your investments are doing and check them during the worst moments. Meaning most will “experience” the full drawdown.
If you’re looking to understand how a strategy performs, how it “feels” in the moment, it makes sense to use daily returns.
That’s why I used daily data to report maximum drawdown and develop drawdown charts. I have the data, and reporting anything less would distort the true experience of the strategy.
Calculations are my own. Results are theoretical, do not represent returns that any investor actually attained, and are not an indicator of future results. Returns do not reflect trading fees. Dividends are re-invested.
In 2020, Geometric Balancing pulled back about 8.6% from its all time high. From a monthly perspective it only pulled back 1%. One percent is a tiny blip that doesn’t really describe the March experience.
What Do Most Backtests Use?
Many backtests use monthly returns to report their drawdowns. Often this is out of necessity as there isn’t always daily data to perform the backtest. Sometimes thought there probably is daily data available, and monthly drawdowns are still reported.
I hope you can see now why daily returns provide a much clearer perspective of a strategy’s performance than monthly returns. There is nothing wrong per se with monthly drawdown numbers, but if you compare two strategies, you have to check to see that the “resolution” of the drawdown is the same.
Most importantly, you need to understand that real life will likely feel worse than the experience a monthly drawdown implies.
As an example, my unlevered strategy reports a theoretical 14.3% max drawdown over the last 42 years, occurring in September 2001. But if you use monthly numbers instead of daily returns, that drawdown becomes only 9.6%.1 A one-third reduction in reported drawdown.
Daily Drawdowns are Superior
If you can, try to use and look for daily drawdown numbers. When comparing strategies, always look to confirm that the drawdown calculation periods are identical.
Pay attention to the resolution of the reported drawdown, because the picture they paint might distort the real experience.
1-The worst drawdown actually moves to February 2009, at 10% if you use monthly numbers.