Why almost every spread comparison is wrong

Almost every broker comparison on the internet compares spreads, and almost all of them compare wrongly.
Not because the figures are invented, but because the method does not control for the things it has to.

Six reasons a spread comparison reaches the wrong conclusion

1. Comparing advertised spreads instead of real ones

“Spreads from 0.0 pips” is the best figure under ideal conditions, on one specific instrument, on one
specific account type. The word “from” carries all the weight. Comparing “from” figures with each other
compares different sets of ideal conditions.

2. Not at the same moment

Spreads move constantly with liquidity. The same pair can be twice as wide in the early Asian session as it
is in London. Measuring broker A at 08:00 UTC and broker B at 21:00
tells you about the hours, not about the brokers.

3. Ignoring commission

A Raw account has a near-zero spread but charges $4.50–7 per round-turn lot. Comparing broker A’s Raw
spread with broker B’s Standard spread without adding commission compares two different things.

4. Not converting to a common unit

One pip of EUR/USD is $10 a lot. One pip of USD/JPY is about $6.70 and
moves with the rate. One cent of gold is $1. “A spread of 1.2” on those three instruments is three completely
different costs.

5. Averaging far too small a sample

One screenshot is not a measurement. Spread has to be measured repeatedly over several days, in the same
hours, and averaged. A single look reflects that minute, not the broker.

6. Ignoring the rebate

This is the most frequently omitted variable, and it can reverse the result. An account with a 1.2-pip
spread paying $8/lot has a net cost of $4; an account with a
0.2-pip spread plus $6 commission paying $3.50 has a net cost of $4.50. A spread table ranks them the other way
round.

The comparison that works, in four steps

1. Choose one instrument and one time window — the instrument and hours you actually
   trade.
2. Measure the spread daily for a week in that window, on each broker’s own platform, and
   average it.
3. Convert everything to dollars per round-turn lot — spread times
   pip value, plus round-turn commission.
4. Subtract the rebate for the account type you actually use, then multiply by your real
   monthly lots.

That final number is the only one worth comparing. It also explains why there is no general answer: the
cheapest broker for someone running 200 lots of scalping is not the
cheapest for someone running 5 lots of swing.

And swap

For anyone holding overnight, swap is usually larger than spread and commission combined. A short USD/JPY
position held 20 days can cost $286 in swap against $7 of spread. A comparison table without a swap column is
useless to a swing trader.

Why almost nobody does it properly

Because the correct comparison gives a different answer for every person, while a ranking needs a single
order. A ranking is easier to read, easier to share, and more wrong.

If you are reading a broker comparison, look for the methodology before the conclusion. Without a method,
the conclusion means nothing.

This article is for information only and is not investment advice. Conditions and fee levels are published by the brokers and can change at any time — check with the broker you actually use. Leveraged forex and CFD trading carries a high level of risk and can cost you your entire deposit.