Monthly returns

The average is the least useful number

“Averaging 1.6% a month” sounds like a promise that every month will be roughly 1.6%. It is not. An average tells you where the months balance out, not what a normal month looks like, and certainly not what the worst one looks like.

Here is one invented year of monthly returns, in percent:

Jan	Feb	Mar	Apr	May	Jun	Jul	Aug	Sep	Oct	Nov	Dec
3.0	2.5	4.0	−6.0	2.0	3.5	1.5	2.5	−2.0	3.0	2.0	3.0

The average is 1.6%. The median, the middle month once they are sorted, is 2.5%. Compounded, the year finishes at about +20%.

All three numbers are true. None of them tells you that in April the account lost 6%, and that is the month in which people disconnect.

Average versus median

When one or two months are far from the rest, the average gets pulled towards them and the median does not. In the example, the two losing months drag the average almost a full point below the typical month.

The gap between the two is a quick signal. If a track record’s average and median months are close, the months are fairly even. If they are far apart, a few extreme months are doing a lot of the work, in either direction.

Look for the worst month, then the worst run

Before anything else, find two things on the monthly grid:

• The worst single month. Ask yourself honestly whether you would have stayed connected through it, measured in euros on your own deposit, not in percent.
• The longest losing streak. Two or three red months in a row test patience far more than one sharp loss followed by recovery.

These two numbers describe the ride. The average only describes the destination.

Why a perfectly smooth curve is suspicious

Markets move unevenly, so honest trading results are uneven too. A multi-year record with no losing months at all deserves a second look, not a first deposit.

Smoothness can be manufactured. Strategies that hold losing trades open indefinitely, or add to them as price moves against them, can show a steadily rising balance while the account’s equity carries a large open loss underneath. The monthly grid then looks perfect until the month it does not. The third and sixth lessons in this series show how to spot this with drawdown and the trade history.

This is also why regulated firms in the EU must show past performance over complete 12-month periods, covering five years or the whole life of the product if it is younger. The rule exists because a well-chosen window can make almost anything look smooth.

Reading our monthly grid

Each strategy page on this site shows a grid of monthly results, one cell per month, taken from the live Myfxbook account. Read it in this order:

1. Find the lowest cell.
2. Find the longest run of negative cells.
3. Only then look at the average and the total.

Volatility, the size of the swings from month to month, matters as much as the return itself, because it decides how the strategy feels to live with. A slightly lower return with calmer months is often the one people actually stay with long enough to collect.