Drawdown and monthly returns, decoded

What drawdown is

Drawdown is the percentage drop from a portfolio's all-time peak to a subsequent trough. Maximum drawdown is the worst such drop the portfolio has ever experienced — measured peak-to-trough, not calendar-year-to-calendar-year.

A portfolio that compounded 12%/year over a decade and had a 38% peak-to-trough fall along the way is a wildly different lived experience from one that compounded 9%/year with a max drawdown of 12%. Drawdown is the only number that distinguishes them honestly.

Peak, trough, and the underwater curve

The underwater curve plots, for every date, how far below the running peak your portfolio sat at that moment. Most of the time it's 0% (you're at or near a peak). The big dips are the moments that test your conviction — 2020 covid, 2022 inflation crash, 2008 GFC.

FolioInsights shows the underwater curve clipped to the period you selected in the benchmark hero, so the chart and the summary statistics describe the same window. Selecting a 5-year period and seeing a -22% max drawdown is a true statement; selecting all-time and seeing -42% is also true and probably more useful when judging if you can stomach the next one.

Why DCA portfolios distort raw monthly returns

A naive monthly return is end-of-month value divided by start-of-month value. The first time you deposit €5,000 into an empty account, that ratio is infinite. The first month after a 50% top-up shows a 50%+ 'return' before a single share has moved.

Without correcting for this, the analytics 'best month' and 'worst month' headlines end up reflecting deposits and withdrawals, not market performance. A user who DCA's monthly will see a +160% best month roughly the day they imported their first CSV — meaningless as a measure of how the portfolio behaves.

The fix: time-weighted return

Time-weighted return (TWR) nets out the cashflow on each transaction date — buys, sells, deposits, withdrawals — so what remains is the return of whatever you held that month. It's the same construction used by mutual funds for their official return numbers, which is why TWR is comparable across portfolios with very different cashflow patterns.

FolioInsights uses TWR for every monthly-return statistic in the analytics page: the histogram, the best/worst month headlines, the win-rate, and the inputs to alpha/beta. The drawdown chart itself uses raw net worth (peak-to-trough math is correct on the raw series), but the headline statistics next to it use TWR.

Reading the monthly-return histogram

The histogram is the distribution of monthly TWR over the selected period. A well-behaved equity portfolio looks roughly bell-shaped, centred slightly above zero, with a long left tail (a few brutal months) and a shorter right tail (rarely is a great month that great).

The win-rate — proportion of months in positive territory — is typically 58–65% for a diversified equity portfolio. Above 70% over multiple years is unusual and worth a sanity check; below 50% over multiple years means the market did most of the work against you, not for you.

See your drawdown →

What max drawdown is 'normal'?

For a diversified global equity portfolio, expect roughly -20% in a normal bear market and -40% to -50% in a once-a-generation crash — think the 2008 GFC, the 2000–02 dotcom unwind, or the 1973–74 oil-shock bear. Concentrated single-stock portfolios can have max drawdowns of -60% or more even when the broad market is doing fine.

Why is my best month so much smaller than my biggest deposit?

Because the headline is computed on time-weighted return, not raw value change. A €5,000 deposit doesn't count as a 'month' return — the TWR for that month reflects only how your existing holdings performed, not the new capital. This is intentional: it's the only way to compare a DCA portfolio to a benchmark fairly.

What's a good win-rate of months?

58–65% is typical for diversified equity over multi-year periods. Bond-heavy or income-focused portfolios run higher (more frequent small positives), high-volatility tech-heavy portfolios run lower. The number matters less than its consistency: a stable 60% over a decade is a healthier portfolio than one that swings between 50% and 80% year to year.

How is drawdown different from volatility or Sharpe ratio?

Volatility (standard deviation of returns) treats upside and downside symmetrically — a portfolio that doubles overnight is as 'volatile' as one that halves. Drawdown only cares about downside, which is closer to how investors actually experience risk. Sharpe combines return and volatility into one ratio; drawdown stays an honest worst-case figure.

Does the drawdown calculation include dividends?

It uses the same net-worth series as the rest of the analytics page, which is positions valued at current price plus historical dividend cash. Reinvested dividends compound, uncollected dividends sit in net worth as cash; either way they're in the curve.