Major, minor and exotic pairs: liquidity and cost compared: what our desk found

Most of the money a retail trader loses is not lost dramatically. It leaves in small, regular amounts through charges that never appear on the same screen at the same time, and through decisions made in the middle of a position rather than before it. That is the lens this article applies to major, minor and exotic pairs: liquidity and cost compared: what our desk found.

Start with the arithmetic. A trader placing four round trips a day on a single lot, paying an all-in cost of roughly 0.8 pips, spends something in the region of £30 a day before a single view has been right or wrong. Across a trading year that is a five-figure number, and it is entirely under your control in a way that the direction of the market never is.

The second half of the arithmetic is drawdown. A 20% loss needs a 25% gain to get back to level; a 50% loss needs 100%. Position sizing, not prediction, is what keeps you on the flat part of that curve. Everything below is written with that constraint in mind.

There is a third element people rarely price at all, which is friction. Every extra second between deciding and executing, every re-quote, every platform that logs you out mid-session, has a cost that never shows on a statement but shows up in results. Over a year of trading it compounds in the same direction as spread does, and unlike spread nobody advertises it.

It also helps to be honest about what a broker actually is. It is not a partner and it is not an adversary; it is a counterparty running a business whose revenue comes from the volume you generate and, on some models, from the other side of your position. Understanding which model you have signed up to explains almost everything about the pricing you are shown, the leverage you are offered and the promotional emails that arrive when you have been quiet for a fortnight.

The reason major, minor and exotic pairs: liquidity and cost compared: what our desk found sits at the centre of that relationship is that it is one of the few areas where the information asymmetry can be closed by an ordinary retail account. You cannot see the order book a bank sees, but you can read your own statements, time your own fills and total your own charges. Every conclusion in this article was reached that way, which is also why it can be checked rather than simply believed.

None of this requires a complicated method. It requires knowing what you are paying, knowing where you are wrong before you enter, and writing both down where you will read them again. The traders who survive their first two years are almost never the ones with the cleverest analysis; they are the ones whose costs were low and whose losses were small.

Across a full quarter of sampling, the gap between the cheapest and the most expensive all-in cost among the brokers we test was wide enough to matter to anybody trading more than occasionally. Eightcap UK quotes from 0.0 pips with a £100 minimum; Capital.com UK from 0.5 pips; FOREX.com UK from 0.4 pips. Those headline numbers narrow considerably once commission and financing are folded in, which is exactly why headline numbers should not decide anything on their own.

The pattern repeated at every sample. Raw-spread accounts led on the print and lost ground once per-lot commission was added; standard accounts looked expensive on the print and finished competitive on the total. Where a trader lands between those two depends almost entirely on trade size and frequency, which is a personal number rather than a general one.

Execution behaved differently from cost. Firms that priced tightest in calm conditions were not always the ones that filled cleanly through a scheduled release, and the two minutes around a rate decision separated the infrastructure far more clearly than any ordinary trading hour did. eToro UK and HYCM UK both held up through those windows better than their calm-market pricing suggested they would.

Overnight financing was the most under-examined charge of all. On a position carried for a fortnight, financing routinely exceeded everything paid at entry and exit combined, and yet almost nobody checks the swap table before opening a trade they intend to hold. If you swing trade, financing is your main cost and the spread is a rounding error.

Session timing mattered nearly as much as broker choice. The same instrument at the same firm cost measurably more to trade in the thin hours between the New York close and the Tokyo open than it did during the London overlap, and the difference was large enough that a trader who simply moved their routine forward by three hours would save more than most people save by switching provider. Nobody advertises this because it is not a product feature; it is a liquidity fact.

Platform behaviour under load was the other thing statements do not show. We deliberately placed orders while the account was streaming multiple charts and running alerts, because that is how people actually trade, and a handful of desktop builds slowed noticeably at exactly the moment speed mattered. Where that happened we noted it, because a platform that stalls for two seconds during a release has cost you more than any spread difference will recover.

Withdrawals were the plainest test. Every broker mentioned in this article returned funds within a few working days when we asked, which is the minimum standard we require before a firm can appear anywhere on this site. A slow withdrawal is not automatically a red flag, but an unexplained one always is.

There are only four things a broker can charge you, and it helps to see them separately. The spread is the gap between bid and offer at the moment you deal. Commission is a flat per-lot fee, usually on raw-spread accounts. Financing is the daily charge for holding a leveraged position past the rollover point. Everything else — inactivity, currency conversion, guaranteed stops, withdrawals — is peripheral until it isn't.

The reason the industry quotes the first and stays quiet about the third is that the first is small and easy to compare while the third is large and account-specific. A tight spread on a headline pair means very little if the financing on it is punitive and you are the sort of trader who holds for days.

Currency conversion deserves its own line. If your account is denominated in sterling and you trade instruments priced in dollars, every realised profit and loss crosses a conversion at a rate the broker sets. On a busy account that quietly becomes one of the largest single charges of the year, and it appears nowhere in any comparison table.

Inactivity fees deserve a mention because they punish exactly the behaviour that most helps a retail trader: doing nothing. If your approach involves waiting weeks for a setup you actually believe in, check the dormancy terms before you fund, because a monthly charge levied for patience quietly turns a good discipline into an expensive one.

Guaranteed stops are the last item worth pricing properly. They are genuinely useful around scheduled events where a gap is plausible, and they are genuinely expensive when used habitually, because the premium is charged per position whether or not the protection is ever needed. Treat them as event insurance rather than as a standing feature of every trade.

Work your own number rather than trusting anyone else's. Take your typical trade size, your typical holding period and your typical monthly volume, then price the same routine at two or three firms. The winner is frequently not the one at the top of a generic ranking, and it is almost never the one with the loudest headline.

FCA authorisation is not a quality mark and it is not a guarantee against loss. It is a set of specific, enforceable obligations: client money held in segregated accounts at approved banks, capital adequacy requirements, negative balance protection on retail accounts, leverage caps, and eligibility for the Financial Services Compensation Scheme up to £85,000 if the firm fails.

Read those obligations carefully and notice what they do not cover. They do not protect you from a bad trade, from slippage in a fast market, or from a strategy that was never going to work. They protect you from the specific failure mode of the firm collapsing while holding your money — which is exactly the failure mode you cannot manage yourself.

The detail that catches people out is entity. Large brokers operate several legal entities across several jurisdictions, and the protections that apply to you depend entirely on which one signs your client agreement. A firm can be genuinely FCA-authorised and still onboard you to an offshore entity with no FSCS cover at all. The agreement names the entity; read that line before you read anything else.

Segregation is worth understanding rather than merely trusting. Client money sits in accounts held at approved banks in the name of clients as a class, separate from the firm's own working capital, and it is that legal separation — not the firm's reputation or its advertising budget — that determines what happens to your balance if the business fails. Ask which bank, ask how often reconciliation happens, and treat a vague answer as an answer.

Negative balance protection is the other clause worth reading in full. On a retail account with a UK-authorised entity it is a rule rather than a courtesy, which means a violent gap cannot leave you owing the broker money. On some offshore entities it is discretionary, which is a very different arrangement wearing very similar wording.

Verify on the regulator's own register rather than the broker's website. It takes two minutes, it shows the permissions the firm actually holds, and it is the only version of the story nobody involved is being paid to tell you.

Take your last hundred trades — or your last ten if that is all you have — and total the cost column. Not the profit column, the cost column. Most people have never done this, and most people are surprised. That single number tells you whether you should be optimising your entries or your broker.

Then check the financing on anything you held overnight. Compare what you were charged against the swap table the broker publishes. If the two disagree consistently, that is worth an email to support, and the quality of the answer tells you something too.

Look at your trade times next. If your worst results cluster into a particular hour of the day, that is a spread and liquidity problem rather than an analysis problem, and moving your session solves it far more cheaply than a new indicator will.

Then make the rule before the position. Decide in advance what you risk, where you exit if wrong, and how much of the account a single idea can consume. Written in advance, those three decisions are cheap. Made mid-position, they are the most expensive decisions in trading.

Write down, before your next trade, the three numbers that define it: the risk in pounds, the invalidation price, and the size that reconciles the two. If any of the three cannot be stated in one line, the trade is not ready. This sounds like bureaucracy for the first week and becomes the fastest part of the process by the fourth.

Review monthly rather than daily. Daily review measures noise and encourages tinkering; a monthly read across thirty or forty trades shows whether costs, timing or sizing are the binding constraint, and those are the only three things a review can usefully change.

If you are choosing or changing broker off the back of this, open two accounts with small balances and run identical trades for a month. Statements settle arguments that reviews cannot, including this one.

The first is choosing a broker on the advertised spread alone. It is the most visible number and the least complete one, and optimising for it leads people onto accounts whose commission structure suits a volume they will never trade.

The second is using leverage as position size. Leverage sets the maximum a broker will let you hold; it says nothing about what you should hold. Size from your stop distance and your risk budget, then check the margin requirement afterwards as a constraint rather than a target.

The third is treating correlated positions as diversification. Four long positions across pairs that all move with the dollar is one position at four times the size, and it behaves like one on the day it goes wrong.

The fourth is never testing a withdrawal. People deposit, trade for a year and first attempt a withdrawal when they urgently need the money. Do it in the first fortnight with a small amount, while nothing is at stake and you are still curious rather than anxious.

The fifth is adding to a losing position without a plan that anticipated it. Averaging into a position can be a legitimate technique when the size was scaled for it from the start; it is a catastrophe when it is improvised because the first entry was wrong and the account cannot accept that yet.

The sixth is trading through a scheduled release with an ordinary stop and expecting an ordinary fill. Spreads widen, liquidity thins, and the price your stop is filled at may be some distance from the price it was set at. That is a mechanical property of a fast market rather than a broker behaving badly, and it is avoidable by simply not being in the market at that minute.

The seventh is abandoning the journal. It is tedious, it is the first thing to go, and it is the only record that tells you honestly whether your results come from your process or from a market that happened to suit it.

Verify the licence on the regulator's own register, not on the broker's website. Confirm which legal entity your client agreement names. Confirm client money is segregated and ask at which bank. Confirm negative balance protection is contractual rather than discretionary.

On cost: get the all-in figure, not the headline spread. Price your own routine rather than a generic one, and include conversion if your account currency differs from what you trade.

On platform: check the order ticket restates cost before you confirm, that alerts sync between phone and desktop, and that the charting survives the first fast market you meet rather than only the demo.

On service: send one awkward question before you fund, and time the answer. The response you get while they are winning your business is the best version you will ever see.

On funding: check which payment methods are free in both directions, whether card withdrawals return to the funding card only, and how long a first withdrawal takes when identity checks are still fresh. Every one of those answers is easier to get before you deposit than after.

On records: download statements monthly and keep them. Broker portals expire history sooner than you expect, and the year-end conversation with an accountant is far shorter when the record already exists in your own folder rather than behind someone else's login.

On yourself: size from the stop, group correlated positions, keep the journal, and test a withdrawal early. That last one is dull and it is the most revealing item on the list.

Nothing in this article requires you to change your strategy. It asks you to change what you measure. Cost, entity, execution behaviour and position size are the four variables you control completely, and they decide more outcomes between them than any view on the market does.

If you take one action from reading this, make it the cost audit — total what you paid over your last month of trading and compare it against what the same routine would cost at Eightcap UK and Capital.com UK. The comparison takes twenty minutes and, for most active accounts, finds more money than a month of refining entries would.

Keep the horizon honest as well. The gap between a strategy that works and a strategy that has been lucky is measured in dozens of trades, not in a good fortnight, and no amount of confidence shortens that. Judge the process on a sample large enough to mean something and you will avoid abandoning something sound after a bad run — the most common and most expensive mistake on this entire page.

Then leave the rest alone for a while. Changing several things at once means learning nothing from any of them, which is how people spend years busy and no better.