Maximum drawdown is the largest percentage decline from any peak to a subsequent trough in the price series over the lookback period. Formula: MaxDD = min(Pt / max(P0..t) − 1) over all t. For UK Tax Drag, we use 5 years of daily total-return prices (more granular than monthly — captures intraday peak-to-trough). The metric is always negative or zero. A max drawdown of −35% means at some point the ETF was down 35% from its prior peak. Recovery time is the days until the price exceeded that prior peak again.
The formula
Drawdownt = Pt / max(P0, P1, ..., Pt) − 1 MaxDrawdown = min(Drawdown0, Drawdown1, ..., Drawdownn) where: Pt = total return price at time t max(P0..t) = highest price observed from time 0 through t (running peak) Drawdownt = current decline from the running peak Always expressed as a negative number or zero.
Conceptual walkthrough: at each day, ask "what would have happened to someone who bought at the previous peak?" The worst answer across the lookback period is the max drawdown.
Inputs we use
| Input | Source | Notes |
|---|---|---|
| Daily total-return prices | Issuer + LSE end-of-day close | 5-year lookback by default |
| Currency | GBP, converted at daily spot rate | Total return basis (reinvested distributions) |
| Running peak | Calculated forward-only | No look-ahead — peak is "highest so far" |
Why daily data, not monthly
Most ETF metrics use monthly returns because monthly is the academic standard. For max drawdown, daily is better because:
- Real-life experience: investors see their portfolio daily, not at month-end. A 25% intra-month drawdown that recovers by month-end shows as 0% in monthly data — but the daily experience was traumatic.
- Captures crashes: COVID March 2020 was a fast drawdown that recovered by April. Monthly data shows only a modest decline; daily captures the full −34% drop.
- Better recovery tracking: daily data tells you exactly when the price returned to the prior peak. Monthly data gives ~30-day uncertainty.
Worked example — VWRL max drawdown
VWRL daily prices, 5 years (April 2021 to April 2026).
| Peak price: 30 December 2021 | £98.50 |
| Trough price during subsequent decline: 12 October 2022 | £77.20 |
| Drawdown: (£77.20 / £98.50) − 1 | −21.62% |
| Days from peak to trough | 286 days |
| Date price exceeded prior peak again: 17 January 2024 | ~£98.55 |
| Days from trough to recovery | 462 days |
| Total time underwater (peak to recovery) | 748 days (~25 months) |
| Max drawdown over 5-year window | −21.62% |
VWRL's max drawdown of −21.62% over the 5-year window includes the 2022 bear market. An investor who bought VWRL at the December 2021 peak would have seen a 22% loss before recovering — and would have been underwater for roughly 25 months total.
Different lookback windows give different max drawdowns: a 10-year window including 2020 COVID would show closer to −34% max drawdown.
What max drawdown tells you
- What "bad" actually looked like. Standard deviation describes the average month; max drawdown describes the worst experience.
- Recovery time matters. A 30% drawdown that recovers in 6 months is very different from a 30% drawdown that takes 5 years to recover. Always look at recovery time alongside the percentage.
- Asset class context. Equity ETFs: 30-50% max drawdowns over 10-year windows are normal. Bond ETFs: 10-20% is typical. Cash equivalents: 0-2%.
- Concentration risk. Single-country / single-sector ETFs typically have larger max drawdowns than diversified globals.
What max drawdown does NOT tell you
- Probability of future drawdowns. Past max drawdown is not a forecast — markets can produce drawdowns larger than anything observed historically.
- Frequency of drawdowns. An ETF could have one big 30% drawdown and no others, OR five 15% drawdowns. Same max DD, very different experience.
- The shape of the decline. A 30% drawdown over 18 months (slow grind) feels very different from 30% in 2 weeks (crash). Max DD doesn't capture the path.
- How likely you would have stayed invested. Many investors sell during drawdowns — the realised drawdown for an investor who sold at −20% is different from the ETF's max drawdown.
Drawdown by asset class — typical historical ranges
| Asset class | Typical 10-year max drawdown | Typical recovery time |
|---|---|---|
| UK gilts (intermediate) | −10% to −15% | 6-18 months |
| Investment-grade corporate bonds | −10% to −20% | 12-24 months |
| Global aggregate bond (hedged) | −10% to −18% | 12-24 months |
| Global equity (VWRL-like) | −30% to −40% | 1.5 to 5 years |
| US equity (S&P 500) | −30% to −45% | 1.5 to 5 years |
| UK equity (FTSE 100) | −30% to −40% | 2 to 7 years |
| Emerging market equity | −45% to −60% | 3 to 8 years |
| Smaller-cap equity | −45% to −60% | 3 to 8 years |
| Single-sector (tech, biotech) | −50% to −75% | 3 to 10+ years |
| Leveraged 2× ETFs | −70% to −90% | 5+ years (often never) |
How to reproduce this yourself
- Download 5 years of daily total-return prices for your ETF (issuer page or Yahoo Finance — Yahoo's "Adj Close" is total return).
- In a spreadsheet, calculate the running maximum: =MAX(A2:A$2) (where A is the price column, starting at row 2).
- Calculate drawdown: =(A2 / MAX(A$2:A2)) − 1.
- Find the minimum drawdown: =MIN(drawdown_column).
- For recovery analysis, find the row where the running max started, the row of minimum drawdown, and the row where price exceeded the running max again.
Sources and methodology
Max drawdown methodology follows standard academic and industry practice (Brooks, Introductory Econometrics for Finance). 5-year daily lookback aligns with Morningstar's reporting standard. The ETF Data Methodology documents all data sources. The site methodology documents the broader review process.
Related metric pages
How UK Tax Drag holds itself to account
Every page is reviewed against the editorial standards, written from primary sources, sourced openly, and corrected publicly. No affiliate revenue. No sponsored content. No paid placements.