Win Rate Without Expected Value Is a Vanity Metric — Two Broker Fee Structures Prove It

An MT5 account history tab, forwarded through a Mumbai trading group last month: 247 closed positions over 90 days. Win rate displayed at the top — 68.4%. The equity curve below it slopes downward, account down roughly a third from its opening deposit. The screenshot circulated with the caption 'what am I doing wrong.' MT5 does not display expected value per trade. That absent number is the answer. We used published spread schedules — Exness Pro at 0.1 pip on EUR/USD, FXTM Standard at 1.5 pips — and calculated what EV per trade looks like under each fee structure at that same win rate. The divergence is where this story begins.

Eight dimensions separate these two account types in ways that reshape expected value per trade. Most of the differences look trivial in a feature-comparison table. Two of them are not trivial at all — and those two explain why a 68.4% win rate can still drain an account.

Spread Cost and the Break-Even Win Rate It Quietly Forces

Stare at the table's third row. A trader who averages 5-pip wins and 15-pip losses — the pattern endemic to anyone who cuts winners early and holds losers hoping for a reversal — needs a 75.5% win rate just to reach zero on Exness Pro. On FXTM Standard, that threshold climbs to 82.5%. The 68.4% win rate from the screenshot falls short of both.

The formula behind those numbers is not complicated. Break-even win rate equals average loss plus spread, divided by average win plus average loss. The denominator stays fixed at 20 pips in the skewed scenario. Only the numerator moves — and it moves because spread is added directly to the loss side of every trade, regardless of outcome. Win a trade by 5 pips on a 1.5-pip spread and you keep 3.5. Lose by 15 pips on that same spread and you lose 16.5. The asymmetry is structural. It does not depend on skill, timing, or market conviction.

Now run the expected value calculation at the screenshot's 68.4% win rate under the skewed profile. Exness Pro: 0.684 multiplied by 4.9 (win minus spread) minus 0.316 multiplied by 15.1 (loss plus spread). Result: approximately -1.42 pips per trade. FXTM Standard: 0.684 multiplied by 3.5 minus 0.316 multiplied by 16.5. Result: -2.82 pips per trade. Nearly double the bleed. Over 247 trades in 90 days, the aggregate gap is roughly 346 pips — a number that exists entirely because of which account type the trader selected, not because of any difference in market analysis.

Both spread schedules are published. The arithmetic takes ninety seconds. Yet the screenshot that circulated through that Mumbai group did not mention spread once.

Leverage Multiplies Expected Value in Both Directions

Both accounts cap leverage at 1:2000. The row looks like a draw. It is not.

The LBMA PM fix — the daily benchmark for global gold pricing set each London afternoon — clarifies why identical leverage produces divergent outcomes. Institutional desks reference that fix for physical settlement. Retail traders on MT5 reference whatever their broker quotes, plus the spread layered on top. That spread is the starting distance from profitability on every position. Leverage does not shrink the distance. It amplifies whatever outcome that distance produces.

Here is the mechanism most leverage guides omit. Leverage does not generate expected value. It multiplies the EV that already exists per trade by the position size a trader opens relative to their deposited capital. A strategy running negative EV at -1.42 pips per trade on 1:100 leverage bleeds gradually — the kind of slow decline that takes months to register psychologically. The same strategy at 1:2000 bleeds at a pace that turns months into weeks, because position sizes are twenty times larger relative to equity. Both Exness Pro and FXTM Standard offer the identical maximum. But the trader paying 1.5 pips starts every leveraged position further behind the line than the trader paying 0.1.

Indian retail accounts funded through UPI — often with deposits of ₹5,000 or ₹10,000 — frequently default to high leverage precisely because the deposit size demands it. At those account balances, the ratio of spread cost to available margin determines whether leverage is a tool or an accelerant. A 0.1-pip spread on a ₹8,300 account leaves room for the trade to breathe. A 1.5-pip spread on the same balance means the position opens already consuming a meaningful fraction of available margin. Same leverage. Different mathematics.

Account Tier — Same Broker, Different Expected Value

This dimension cannot fit inside a single table row. Exness publishes a standard-account EUR/USD spread of 1.0 pip and a Pro-account spread of 0.1 pip. Same broker. Same FCA authorisation. Same MT5 interface. FXTM mirrors the pattern: its standard tier carries a published EUR/USD spread of 1.5 pips while its lower-spread tier drops to 0.1.

The break-even win rate calculation across all four tiers, assuming symmetric 1:1 risk-reward at a 10-pip target, tells the story concisely:

• Exness Pro (0.1 pip): 50.5%
• Exness Standard (1.0 pip): 55.0%
• FXTM lower-spread tier (0.1 pip): 50.5%
• FXTM Standard (1.5 pips): 57.5%

Seven percentage points separate the cheapest tier from the most expensive. A trader running a strategy that genuinely produces a 55% hit rate is profitable on either pro-tier account and underwater on FXTM Standard — executing identical trades, on the same instrument, during the same session hours. The strategy did not change. The dropdown menu during account registration did.

This is the dimension we spent the most time tracing through the data, because its implications are uncomfortable. Account selection happens once, typically within the first few minutes of the onboarding flow, and it determines the expected value of every subsequent trade. No indicator, no chart pattern, no moving-average crossover strategy compensates for a structural cost advantage that the trader on the other account tier receives by default. The trader from the screenshot may have been running a perfectly viable approach — undone not by the market, but by a default setting they never revisited.

FCA Oversight and What It Means for Execution Quality

The FCA's Conduct of Business Sourcebook — specifically its best execution obligations under COBS 11.2 — requires authorised firms to take sufficient steps to obtain the best possible result for clients when executing orders. Both Exness and FXTM hold FCA authorisation. That shared row in the table conceals a dimension most comparison articles ignore entirely: the gap between published spread and realised spread, and what regulatory oversight actually does about it.

Published spread is a schedule. Realised spread is what the trader pays when the order fills. Slippage, requotes, and asymmetric fill quality — where losses fill at worse prices than wins — introduce a hidden drag that no published schedule captures. FCA authorisation does not eliminate slippage. What it requires is that the broker monitor execution quality, report on it, and demonstrate that its practices serve the client's interest rather than the desk's. The difference between a broker publishing 0.1 pip and consistently filling at 0.1 versus one publishing 0.1 but filling at 0.3 during London-open volatility is the difference between theoretical EV and the realised number in the account history tab.

For the Indian retail trader funding an account through UPI, the RBI does not regulate offshore forex brokers. Period. There is no domestic regulator auditing the execution quality of trades placed through Exness or FXTM from a Mumbai IP address. FCA oversight, imperfect as any cross-border enforcement is, remains the closest independent check on whether the spread numbers in the table above translate into actual fills — or whether they serve primarily as marketing numbers that widen under volatile conditions the retail trader cannot monitor in real time.

Both brokers carry the same regulatory credential on this dimension. The table row is a tie. But the existence of that row — versus its absence on an unregulated offshore alternative — is the structural floor beneath every EV calculation in this article.

Which Dimension Actually Matters Most

Spread cost, alone, explains the majority of the expected value gap between these two account types. Not leverage — identical on both sides. Not regulatory standing — shared under the same authority. Not withdrawal speed, though capital locked for three business days does carry an implicit drag. The spread is the variable that shifts the break-even win rate upward, reshapes the EV per trade downward, and compounds silently across every position over 90 days of active trading. The trader from the screenshot was not missing a strategy insight. The trader was missing a cost insight.

We would reverse this conclusion under one specific condition: if MetaQuotes added expected value per trade as a default metric in the MT5 account history tab — displayed with the same prominence as win rate — and if both brokers published realised execution quality data showing actual fill prices across account tiers during peak and off-peak sessions. Until that information exists in a format a retail trader can audit without exporting to a spreadsheet and running formulas manually, the published spread remains the single most decision-relevant number in the table above. The 68.4% win rate was never the problem. The absent calculation was.

Frequently Asked Questions

Is a 68.4% win rate good or bad?

Neither, in isolation. Win rate without the corresponding average win size and average loss size is an incomplete statistic. A 68.4% hit rate with symmetric risk-reward — equal pip gain per win and pip loss per loss — produces positive expected value at tight spreads. The same 68.4% with a skewed profile where average losses run three times larger than average wins is negative EV under every fee structure we examined. The number displayed on the MT5 history tab communicates frequency. It says nothing about magnitude.

Does switching to a lower-spread account fix a losing strategy?

It narrows the break-even win rate threshold, which means a borderline strategy might cross from negative to positive expected value. But if the underlying risk-reward ratio is severely skewed — average losses consistently dwarfing average wins — no spread reduction makes 68.4% sufficient. Reducing spread from 1.5 pips to 0.1 pips shifts the skewed break-even threshold from 82.5% to 75.5%. That is meaningful. It is also not enough when the strategy itself produces a lopsided loss distribution. Lower spreads buy margin for error. They do not buy a functioning strategy.

How do I calculate expected value from my own MT5 trade history?

Export your closed-trade history to a spreadsheet. Compute three numbers: average win in pips, average loss in pips, and win rate as a decimal. Expected value per trade equals win rate multiplied by average win, minus the quantity one minus win rate multiplied by average loss. If your account type charges commission separately from spread, subtract commission per round turn from the average win figure before running the calculation. A negative result means each trade costs you money on average — regardless of what the win rate percentage suggests.