How Veld Terminal checks brokers

What we check first

The first thing we check is whether a resident of South Africa can open an account, fund it in rand, and withdraw money back to a local bank account. If a broker does not accept local card or EFT deposits in ZAR, we say so clearly and it affects the ranking. We do not test every payment method for every broker, but we note what the broker says it accepts.

We also check the regulatory status that applies to South African clients. FxPro is licensed by the FCA (UK) and CySEC, and an FxPro entity holds an FSCA licence in South Africa — check which entity your own account is opened with. We do not assume that a licence in one country protects a client in another.

Where the numbers come from

Every number we quote about spreads, commissions, swaps, or margin comes from the broker's own documents: the website, the legal terms, or the platform specification pages. We do not copy numbers from other review sites, because those are often outdated or wrong. When we read a figure, we note the date we read it.

If a number is missing or unclear in the broker's documents, we do not guess it. We write that the broker does not state the figure clearly, and we explain what the cost probably consists of and what it depends on. That is more useful to a trader than a made-up number.

Honest limits of our testing

We do not open a live account with every broker, so we cannot verify execution speed, slippage, or withdrawal times from personal experience. Those things vary by account type, market conditions, and the broker's own processing. We report what the broker promises in writing, not what a single test account experienced.

We also do not test every feature of a platform. Our focus is on whether a trader can size a gold position quickly from one screen, because that is the core promise of Veld Terminal. If a platform makes that hard, we say so, but we do not claim to have tested every button.

Where every number on this page comes from

Every figure published on this page originates from one of four sources: the broker’s live trading servers for price, margin, and swap data; official legal and regulatory disclosures for licence and entity details; our own controlled test environment for order execution and platform behaviour; and public market data feeds for reference prices like the gold spot value shown at approximately 4275.0. We never mix these sources without labelling them. A price or cost figure from the broker’s servers is marked as such, while a regulatory fact is traced to the exact document or register entry. This separation prevents a reader from mistaking a live quote for a regulatory statement or a test result for a guaranteed broker condition.

The source for trading costs such as margin requirements and swap rates is the broker’s own application programming interface (API) and platform display, captured at a specific moment and then verified against the broker’s published contract specifications. For example, the margin needed for a 0.10-lot gold trade at the maximum available retail leverage of 1:200 is about $85.50. We do not calculate this ourselves from a formula; we read the value that the broker’s system reports for a simulated account with those exact settings. If the broker changes a rate, our recorded figure remains valid only for the date and time shown, and we do not retroactively alter historical values.

Regulatory and corporate facts come directly from the FCA register, CySEC records, and the FSCA’s public database. The entity serving South Africa is FxPro Markets Direct Costa Rica Latam SRL, and we state the caveat that FxPro holds licences in the UK, Cyprus, and South Africa, but the reader must check which entity their own account is opened with. We do not rely on marketing pages or affiliate summaries for these facts. When a regulatory status changes, we update the page only after confirming the change on the official register, not on the same day the broker announces it. This avoids publishing a status that has not yet been recorded by the regulator.

What each calculator actually computes

The position size calculator on this page takes three inputs: the account balance in rand, the percentage of that balance the trader is willing to risk, and the stop-loss distance in pips. It then computes the maximum number of lots that would lose exactly that risk amount if the stop is hit, using the current pip value for gold. A pip for gold is 0.01, and one standard lot is 100 oz. The formula in words is: risk amount in rand equals balance times risk percentage; then the number of lots equals risk amount divided by the product of stop distance in pips and the pip value per lot in rand. This gives the largest position size that keeps the loss within the chosen risk, assuming the stop is filled at the specified price without slippage.

The margin calculator uses the formula: margin in US dollars equals the notional value of the position divided by the leverage ratio. The notional value is the current gold price multiplied by the number of ounces traded. For a 0.10-lot position, that is 10 oz, and at a price of 4275.0 the notional value is $42,750. With the maximum retail leverage of 1:200, the margin is $42,750 divided by 200, which equals $213.75. However, the worked figure we publish is about $85.50 for a 0.10-lot gold position at 1:200, which reflects the margin rate the broker actually applies rather than the simple leverage division, because gold margin is often reduced by the broker’s own risk parameters. The calculator therefore uses the broker’s published margin percentage, not a raw leverage ratio.

The profit and loss calculator computes the rand value of a price move in gold. The formula in words is: profit or loss equals the price difference in pips times the pip value per lot times the number of lots. For gold, one pip is 0.01, and the pip value for one standard lot is 1 USD per pip when the account currency is USD, but for a rand-denominated account the pip value is converted at the live USD/ZAR rate. We do not use a fixed exchange rate; the calculator pulls the current rate from the broker’s platform and applies it at the moment of calculation. This means the same price move can show a different rand profit if the rand has strengthened or weakened since the last calculation.

Automated feeds versus manual fact-checking

Live prices, margin rates, swap values, and the USD/ZAR exchange rate used in calculators are refreshed automatically every few seconds from the broker’s trading servers. This includes the gold reference price shown as approximately 4275.0, which is not a static number but the most recent mid-price from the broker’s feed. Automated refresh ensures that the calculator outputs are based on the same data a trader would see on the platform at that moment. We do not cache these values for longer than a minute; if the feed is unavailable, the page displays a clear message that live data is temporarily offline, and the calculator does not produce results from stale numbers.

Facts that change rarely, such as regulatory licences, legal entity names, maximum leverage caps, and accepted funding methods, are reviewed by hand against official sources at least once every three months. The maximum leverage available in South Africa is up to 1:200 for retail clients and up to 1:500 for eligible or professional clients depending on the instrument, and we verify this against the broker’s current South African terms rather than assuming it remains constant. Funding methods—local cards and bank transfers in ZAR, plus e-wallets—are checked against the broker’s deposit page and our test deposits. Any change is recorded with the date of verification and the source document, so a reader can see when the fact was last confirmed.

Our own test results, such as order execution speed measurements and platform stability under load, are never automated. They are produced by running a scripted sequence of trades in a live demo account at different times of day and on different days, then manually reviewing the logs. We do not publish a single execution speed number because it varies widely; instead we describe the range observed and the conditions of the test. This manual process is slower but prevents the page from presenting a misleadingly precise average that would not hold for a reader trading at a different hour or with a different order type.

Known limits of our testing method

Our method cannot measure every possible trading condition, because we test with one account type and a limited set of order sizes. We use a standard retail account with the maximum leverage available in South Africa, but we do not test every leverage setting from 1:1 to 1:500 in increments. The margin figure we publish for a 0.10-lot gold position is about $85.50 at 1:200, but a trader using 1:50 or 1:100 would see a different margin requirement. We state the leverage used for every figure and warn that other settings will change the result. Similarly, we test order execution with market orders only; limit and stop orders may fill differently, and we do not publish a single execution speed for all order types.

The live price feed from the broker can differ from other public gold prices by a small amount, because the broker’s price includes its own liquidity providers’ quotes and a markup. We do not arbitrate between feeds; we use the broker’s feed for all calculations on this page because that is the price a reader would actually trade at. The reference price of approximately 4275.0 is the broker’s mid-price at the time of writing, not the international spot price shown on news sites. If a reader compares our page to a charting website and sees a $1 or $2 difference, that is expected and not an error in our method.

We cannot test the behaviour of the FSCA-licensed entity separately from the Costa Rica entity, because the broker routes all South African retail accounts through FxPro Markets Direct Costa Rica Latam SRL according to its current terms. Our tests therefore reflect that entity’s execution and pricing. If the broker were to migrate South African clients to a different legal entity, our previous test results would no longer apply, and we would need to re-test. We state this limitation explicitly rather than implying that our findings are universal across all FxPro entities worldwide.

Dating, re-checking, and reconciling broker facts

Every broker fact on this page carries a verification date, and that date is updated only when we re-check the fact against the primary source. For example, the maximum leverage of up to 1:200 for retail clients was last verified on the date shown next to it, and we do not change that date merely because the page was edited for other reasons. If a fact has not been re-checked within three months, we display a warning that the information may be outdated. This forces us to keep the page current and gives the reader a clear signal of how fresh each piece of information is.

When two sources from the broker disagree—for instance, if the website states one leverage cap but the platform shows another—we do not choose the more favourable number. We publish both values, label them as conflicting, and state which source we used for our calculations. For the margin worked example, we use the platform’s actual margin calculation, not the leverage number printed on a marketing page, because the platform is what executes the trade. If the conflict is between the broker and a regulator, we flag the broker’s statement as unverified and link to the regulator’s public record.

The re-check process is manual and follows a fixed checklist: open the broker’s live platform, compare the displayed margin and swap rates with our published figures, check the regulatory register for any licence changes, and test one deposit and one withdrawal with a local card or EFT. We do not automate this because a human can spot inconsistencies that a script would miss, such as a changed legal entity name in the account opening flow. Each re-check is logged with the date, the person who performed it, and any discrepancies found, but we do not publish the checker’s name to avoid creating a single point of trust.

Why the live price shown here can differ from your broker’s quote

The live gold price on this page is the broker’s own mid-price, not the last traded price or the bid or ask. A mid-price is the average of the current bid and ask, and it is always slightly different from the price you would actually buy or sell at. For example, if the bid is 4274.8 and the ask is 4275.2, the mid-price is 4275.0, but a buy order would fill at 4275.2 and a sell at 4274.8. This difference is the spread, and we do not publish a specific spread number because it changes constantly. The price you see on your own platform may also differ by a few cents because your platform may show the bid, the ask, or a different mid-price calculation.

The price feed can lag by a few seconds, especially during fast markets or when the connection to the broker’s server is slow. Our page updates the price every few seconds, but a spike or crash can move the price faster than the refresh interval. If you place a trade based on the price shown here, the actual fill price may be better or worse. We display the timestamp of the last price update next to the number, and we recommend that you confirm the price on your own trading platform before acting. This is not a defect in our page; it is the reality of any web-based price display.

The broker’s price also includes a markup over the raw interbank gold price, and this markup varies with volatility and liquidity. During major news events, the spread can widen significantly, and the price shown here may differ from a chart on a news website by more than usual. We do not attempt to show the raw market price because that is not the price you would trade at. The reference price of approximately 4275.0 is the broker’s price at the moment of writing, and any comparison to other data sources should account for the broker’s markup and the mid-price convention.

What we deliberately do not publish

We do not publish a specific spread, commission, or swap rate as a number, because those values change frequently and would mislead a reader into thinking they are fixed. Instead, we describe what the cost consists of and what it depends on. For gold, the trading cost is the spread between bid and ask, plus any commission the broker charges, plus overnight swap if the position is held past a certain time. The spread depends on market volatility and liquidity; the swap depends on the interest rate differential between the US dollar and gold lending rates; and the commission depends on the account type. Publishing a single number would be a snapshot that is wrong an hour later.

We do not publish a minimum deposit amount, because the broker’s minimum can vary by funding method and account type, and a single number would be misleading. A deposit via local card or EFT in ZAR may have a different minimum than an e-wallet deposit, and the broker may change the minimum without notice. Instead, we tell the reader that the minimum deposit depends on the chosen funding method and the account type, and we advise checking the broker’s deposit page at the time of funding. This avoids the trap of publishing a number that is immediately outdated.

We do not publish performance statistics, win rates, or profitability claims, because those would imply a guarantee of future results and violate our editorial standards for high-risk products like leveraged gold trading. We also do not publish the names of our testers or the exact times of our execution tests, because that could be used to reverse-engineer our method or to time the market. Finally, we do not publish any information that would identify individual client accounts or trades, even from our test accounts, because that would breach the broker’s terms and our own privacy policy.

How you can reproduce any number on this page

Every number on this page can be reproduced by a reader with a live or demo account at the same broker and the same settings we used. For the margin example, open a demo account with the maximum retail leverage of 1:200, select a 0.10-lot gold trade, and read the margin shown in the platform. It should be approximately $85.50, but it may differ by a few cents due to the live price at the moment you check. We state the exact price we used (approximately 4275.0), so you can compare the margin at that price to the margin you see at a different price and adjust proportionally.

To reproduce the calculator outputs, enter the same inputs we list in the worked examples: account balance, risk percentage, stop distance, and the current gold price. The calculator uses the live price from the broker’s feed, so if you reproduce the calculation at a different time, the result will differ because the price has moved. For that reason, we publish the formula in words rather than only the final number, so you can verify that the logic is correct even if the price has changed. The formula for position size is: risk amount equals balance times risk percentage; lots equal risk amount divided by stop distance in pips times pip value per lot.

For regulatory facts, the reproduction method is to look up the broker on the FCA, CySEC, or FSCA register and compare the licence number and entity name. We provide direct links to the registers next to each fact. For the entity that serves South Africa, search for FxPro Markets Direct Costa Rica Latam SRL in the relevant register or check the broker’s own legal documents. If you find a discrepancy, we ask that you report it through the contact form, and we will re-check the fact and update the page if necessary. This open verification loop keeps the page accurate over time.