Our Rating Methodology

What the cost figure means

The EUR/USD figure on each broker is an all-in cost: the spread plus any commission, on the account we name. A broker quoting “from 0.0 pips” with a commission can cost more per trade than one quoting a wider spread with none, so a single comparable number is the only honest way to put them side by side.

These figures come from independent tick-data measurement over a stated period. They are not the “from” figures brokers advertise, which describe the best conditions a broker has recorded rather than what a trade typically costs.

A small number of brokers publish their own historical averages rather than a minimum. Where we use that data we say so on the broker’s page, because it is the broker’s own measurement and not an independent one. It is still worth more than an advertised figure: an average over a stated period, at a stated trade size, is something a reader can check.

Where we have no measurement, the field is blank. We do not fill it with an advertised number — which means most brokers in our database show no cost figure at all, and the ones that do are the ones we can stand behind.

Costs are measured on whichever account the source sampled, and that account is named alongside each figure. A broker’s other tiers may price differently: a raw-spread account with commission and a marked-up spread-only account are different products, and which works out cheaper depends on how often you trade.

Coverage is uneven in a way worth naming. Measured data is easier to obtain for brokers running MetaTrader and similar platforms than for those running their own. A blank cost field reflects what we have been able to measure, not a judgement about the broker.

Two scores, averaged

Most comparison sites publish one number that quietly measures brand size. A broker with millions of accounts and a stock listing scores well whether or not its trading conditions are any good. That tells you which firms are big, not which one suits you.

So we publish the average of two scores, and both are visible on the broker’s page.

Brand standing covers scale, operating heritage, listing status and market position. It is assigned by us, and it is the part a reader can already sense: IG has traded since 1974 and sits in the FTSE 250.

Trading conditions is computed from the data we hold, on published weights:

• Regulation — 28%. How many tier-1 licences the broker holds, and whether the entity that would onboard you is one of them. A licence you cannot access counts for nothing.
• Execution — 22%. Whether a genuine raw or ECN account is available, or direct market access. A broker offering only a marked-up spread scores lowest here, because that is a real limit on what an active trader can do.
• Trading cost — 22%. The measured all-in EUR/USD figure, where we have one. Brokers we have not measured abstain rather than being scored as expensive.
• Platform range — 16%. How many platforms are genuinely supported.
• Operating history — 12%. Years trading under the same brand. Thirty years earns full marks; twenty is respectable rather than exceptional.

Because the second score is computed, the same broker scores the same whoever writes the review. That was not true before, and it is why the two are separated: brand standing is a judgement, trading conditions are an arithmetic.

When the two disagree

The gap between them is often more useful than the average. A broker with strong conditions and a modest profile is one most rankings overlook. A household name with ordinary conditions is one most rankings flatter. We show both numbers so you can see which you are looking at.

Conduct is handled separately and more bluntly. A broker with adverse regulatory findings is not given a lower score — it is removed from our rankings and matching altogether, and the reason is published.

Editorial Independence

PipsPal holds basic partnership arrangements with a small number of brokers. We have received no payment from any broker to date, and no partnership carries any weight in our ratings, rankings or matching results.

• Algorithm integrity: Partnership status has zero impact on compatibility scores or rankings
• Full disclosure: We clearly identify brokers we have relationships with
• No paid placements: You cannot pay to appear higher in our matching results
• Negative reviews included: We publish honest assessments, including weaknesses
• Regular audits: We periodically review our methodology for bias