A dashboard rarely lies with a single false number. It lies in the gaps — between numbers that were each calculated honestly, but over different sets of trades, different windows, or different definitions. Put two of them side by side and they imply a strategy that doesn't exist. After auditing a lot of MT4/MT5 dashboards, the same handful of mismatches come up again and again. Here they are, each with a worked illustration and the check that catches it.
The habit that catches almost all of these: before you trust any two numbers together, confirm they were measured over the same trades, the same dates, and net of the same costs. If they weren't, they don't reconcile — and a page that doesn't reconcile isn't a track record, it's a collage.
1. Profit factor below 1.0, next to a positive net profit
This is the classic. A tile reads "Profit factor 0.94" — a losing ratio — and right beside it, "Net profit £3,100." Both can be true at once, because they're computed over different populations. The profit factor is lifetime, across every trade the account ever placed; the net profit is this quarter. The strategy lost money overall and made money recently, and the page quietly staples the good window to the bad-looking ratio so you can't tell which is the real one.
How to catch it: read the date range and trade count on every tile. If the profit factor covers 2,000 lifetime trades and the profit covers 180 from the last three months, they're describing two different strategies. More on profit factor →
2. A headline built on a tiny window
"Profit factor 3.2" sounds elite. Then you find it's over 18 trades. Eighteen trades is a coin-flip sample — two lucky wins produce a 3.2 as easily as a middling strategy in a good fortnight. The number isn't false; it's noise dressed as signal.
How to catch it: find the trade count behind every ratio. Under roughly 30 trades, treat profit factor, win rate and expectancy as meaningless. Ratios need a sample before they mean anything. More on expectancy →
3. Gross figures, quoted before costs
An EA's trade list sums to £6,400 of profit. What the list doesn't show is the £1,900 of commission and £280 of swap that came out of the balance separately. The trade-by-trade profit column on MT4 is frequently gross — costs are booked as their own ledger entries — so a dashboard that just totals the column overstates what actually landed. Real net: £4,220.
How to catch it: reconcile the quoted profit to the equity curve's start and end. If the curve went from £10,000 to £14,220 but the tile says £6,400, the difference is the costs nobody subtracted. Ask outright whether figures are net of commission and swap.
4. One huge winner carrying the whole record
Profit factor 1.5 — £12,000 of gross profit against £8,000 of gross loss. Healthy, until you sort the trades by size. The single biggest was £5,000: one news spike the strategy happened to be sitting on the right side of. Strip that trade and it's £7,000 against £8,000 — profit factor 0.875, a loser. The record isn't built on an edge; it's built on a trade that won't come again on cue.
How to catch it: remove the largest winning trade and recompute. If the profit factor falls through 1.0, the strategy is one outlier away from being a losing system. More on profit factor →
5. Drawdown measured on balance, not equity
A tile proudly shows "Max drawdown 8%." It felt far worse to trade, and it was — because that 8% is balance drawdown, which counts a trade only once it's closed. The strategy is a grid that sat on three open positions £2,200 underwater on a £10,000 account before they came good. Equity drawdown — floating losses included — hit 22%. Balance drawdown is the number that lets a martingale look calm while the account is one adverse candle from a stop-out.
How to catch it: ask whether drawdown includes open floating losses. If it doesn't, it's describing a smoother account than the one you'd actually have sat through. The equity figure is the honest one. More on maximum drawdown →
6. "This year" that's really the last N trades
A tile labelled "2026" shows a tidy return. Count the trades behind it and there are 40 — but the EA places roughly ten a week, so the year to date should hold several hundred. What you're looking at is the last 40 trades relabelled as a calendar period, cropping out the stretch that would drag the figure down. The window is chosen first, then given a date's name.
How to catch it: sanity-check the trade count against the strategy's frequency and the period claimed. Ten a week over eight months is not 40 trades. When the arithmetic doesn't fit the label, the label is the lie.
7. An uptime badge that ignores the offline hours
A green "99% uptime" badge sits at the top of the page. Reassuring — until you ask uptime of what. Often it's the monitoring service's own uptime, or the fraction of time the account was reachable, not the fraction of trading hours the terminal was actually connected and running the EA. A VPS that dropped for six hours over a weekend, missing every signal in the gap, can still wear a 99% badge if the badge measures the wrong thing.
How to catch it: ask what the denominator is. Uptime means something only if it measures the terminal's connection during market hours — the time the EA could actually have traded. Anything else is a green light that says nothing about whether your strategy was running.
The habit that catches all seven
None of these needs forensic skill. They need one reflex: don't read a number, read the number and its population. Every figure has a hidden denominator — which trades, which dates, gross or net, balance or equity — and the lie almost always lives there, not in the digits on the front. Make two numbers prove they came from the same set of trades before you let them tell you a story together. A page that reconciles is worth trusting. A page that doesn't is several different strategies wearing one account's name. More on spotting fakes →
The Karnek note
Karnek is built around the reconciliation habit. Every figure — profit factor, expectancy, net profit, drawdown — is computed from the same live trade history, over a date range you set, net of commission and swap, with the trade count in plain sight. Equity drawdown, not balance. Uptime measured on the terminal's connection during market hours, not a badge. It reads read-only and can never place a trade; it just makes the numbers agree with each other.
Comparing tools? Forex account monitoring & verification →
Written and reviewed by the Karnek Research team. Last updated August 2026.
Educational content only - not financial advice. Past performance does not predict future results. Trading carries significant risk.