CAGR, the compound annual growth rate, is the single yearly rate that would grow your starting balance into your ending balance if it earned that rate every year and compounded. It smooths a bumpy record into one number. An account that went from £10,000 to £17,000 over three years did not grow in a straight line, but its CAGR tells you the equivalent steady pace: about 19.3% a year.
What it is
A simple total return tells you how much you made overall: £10,000 to £17,000 is +70%. What it does not tell you is the pace, and it gives you no way to compare periods of different lengths. CAGR fixes both. It converts the whole run into the constant annual rate that, compounded year on year, produces the same ending balance.
The real account may have jumped 40% one year and lost 5% the next. CAGR reports the smooth equivalent, here 19.3% a year, as if growth had been even. That is what makes it the standard way to compare returns across accounts, strategies and time spans that do not line up.
How it works
CAGR = (ending balance ÷ starting balance) ^ (1 ÷ years) − 1
Three inputs: what you started with, what you ended with, and how long it took in years. Divide the ending balance by the starting balance to get the total growth factor, raise it to the power of one over the number of years, subtract one, and read it as a percentage. The exponent is what turns a multi-year total into a per-year rate. The compounding is baked in, because each year's growth is assumed to build on the last, which is how a trading account actually grows when profits are left to run.
A worked example
An EA takes an account from £10,000 to £21,000 over four years. The total return looks huge at +110%. Spread across four years, though, the compound rate is:
(21,000 ÷ 10,000) ^ (1 ÷ 4) − 1 = 2.1 ^ 0.25 − 1 = 20.4% a year
Now compare it with a second EA that made +90% in two years: (1.9) ^ (1 ÷ 2) − 1 = 37.8% a year. The first EA has the bigger total return, but the second grew almost twice as fast. Total return made the slower system look better; CAGR shows which one actually compounds harder. Leave money in each for the same length of time and the second pulls clearly ahead.
Why it matters
Total return flatters long records and punishes short ones, so it is nearly useless for comparison. A 200% gain sounds better than 50% until you learn the first took a decade and the second took a year. CAGR puts every record on a per-year footing, which is why funds, indices and honest track records quote it.
For automated trading it also sets a realistic yardstick. An EA advertising 300% total return means little without the timeframe. Converted to CAGR it might be a respectable 26% a year over five years, or an implausible figure over five months that no live account will hold.
Common misconceptions
- CAGR is what you actually earned each year. No. It is a smoothed average. The real years were lumpier, and a high CAGR can still hide a brutal drawdown along the way, so always read it beside maximum drawdown.
- A higher CAGR is simply better. Only for the same risk. A 40% CAGR that survived a 60% drawdown is a worse business than a 20% CAGR that never fell past 12%.
- CAGR handles deposits and withdrawals. It does not. If you added or took out money, start-to-end growth is polluted by the cash flow, and you need a time-weighted return instead.
- A long backtest CAGR predicts live returns. Backtested compounding is the number most inflated by curve-fitting and optimistic fills. Live CAGR is almost always lower.
The Karnek note
Karnek computes growth figures from your live terminal on the same balance history and window shown beside them, so a headline rate reconciles with the equity curve it came from rather than a rounder number from somewhere else. It reads the account read-only and can never place, close or modify a trade.
Written and reviewed by the Karnek Research team. Last updated August 2026.
Educational content only - not financial advice. Past performance does not predict future results. Trading carries significant risk.