How brokers are checked and ranked

The first filter: Kenya access

The first thing checked is whether a resident of Kenya can open an account, fund it and withdraw from it. A broker that fails any of those three steps is not ranked, however good its platform looks on paper. Funding is checked against the routes a Kenyan reader actually uses, with M-Pesa as the familiar reference point.

Withdrawal is treated as seriously as funding. A broker that accepts deposits easily but makes withdrawals unclear does not pass this filter. The regulator position is recorded at the same time: FxPro is licensed by the FCA (UK), CySEC and FSCA — not by the CMA of Kenya, and that fact is stated plainly rather than buried.

The known limits of the method

The biggest limit is that the reference price is static. A margin figure built from 77000.0 will not match a broker's required margin once BTC/USD has moved, because notional value moves with price while the leverage cap stays fixed. The method is sound for showing how the cap applies, but it cannot tell a reader what their broker will block today. Anyone sizing a real position must use the live quote in their platform, not our illustrative number.

A second limit is that we publish no spread, commission, swap or minimum deposit figure, so no cost estimate on this site is complete. We can describe what a cost consists of and what it depends on, such as the instrument, the account type and market conditions, but we cannot total it. A reader who needs a full cost picture has to take those numbers from the broker directly, and the absence of them here is a known gap rather than an oversight.

A third limit is regulatory scope. The leverage cap we reference is the one that applies to the FxPro entity serving Kenya, and it is a cap rather than a setting to aim at. We do not state any other ratio, and we do not imply that a Kenyan reader is covered by the CMA, because FxPro is licensed by the FCA (UK), CySEC and FSCA, not by the CMA of Kenya. Any assumption beyond that is outside what the method supports.

How a broker fact is dated and re-checked

Every broker fact carries the date it was last confirmed against the broker's own disclosure for the entity serving Kenya. The date sits with the fact, not in a general footer, so a reader can see how old the platform list, the M-Pesa funding note or the regulator statement is before relying on it. A dated fact is a claim about what the broker said on that date, not a promise about what it says now.

Re-checking is triggered by three things: the date growing old enough that an editor schedules a review, a reader flagging a discrepancy, or the broker publishing a change to its own terms. The check is a direct comparison against the broker's material, and the outcome is one of three: the fact is re-dated unchanged, it is corrected and re-dated, or it is withdrawn if it can no longer be confirmed. Withdrawal is preferred to leaving an unverified claim in place.

The regulator caveat is treated as a fixed statement and is re-checked word for word. It reads that FxPro is licensed by the FCA (UK), CySEC and FSCA, not by the CMA of Kenya. If the broker's disclosure for the Kenyan-serving entity ever contradicted that sentence, the sentence would be the thing that changed, and the change would be dated. We do not soften it, extend it to imply Kenyan licensing, or attach a second regulator to it without a dated source.

Where each class of figure on this site comes from

The reference price of 77000.0 for BTC/USD is a static snapshot used only for illustration and formula worked examples, not a live quote. It is the sole price figure we are given, so any margin, pip value or position-size example on the site is built from it and labelled as illustrative. Readers should treat it as a fixed teaching number, never as a claim about where bitcoin is trading at the moment they read the page.

Contract specifications are fixed and supplied: one standard lot equals 1 BTC and one pip equals 1.0. Those two constants drive every calculator on the site and do not change between sessions. The margin example of about $385.00 for a 0.10-lot bitcoin position is derived from the applicable leverage cap for the FxPro entity serving Kenya, and it is presented as a worked figure for that cap, not as a recommendation to trade at that size or at that leverage.

Broker facts such as the platforms offered (MT4, MT5, cTrader and the FxPro app), the funding route M-Pesa, and the regulatory position are taken from the broker's own disclosures for the entity that serves Kenya. The regulator statement is reproduced exactly: FxPro is licensed by the FCA (UK), CySEC and FSCA, not by the CMA of Kenya. We do not add a Kenyan licence claim, and we do not assign a spread, commission, swap or minimum deposit figure because none was supplied to us.

The formula each calculator uses, in words

The margin calculator divides the notional value of the position by the applicable leverage cap. Notional value is the number of lots multiplied by 1 BTC per lot multiplied by the BTC/USD reference price. For a 0.10-lot position at 77000.0, notional is 0.10 times 1 times 77000.0, which is $7,700.00, and dividing that by the cap gives the roughly $385.00 margin figure shown. The calculator never presents the cap as a target setting; it presents it as the maximum that applies to the entity serving Kenya.

The pip-value calculator multiplies the number of lots by the pip size and by the price, because for bitcoin one pip is 1.0 and the quote currency is USD. With 1 lot equal to 1 BTC, a one-point move in BTC/USD is worth $1.00 per standard lot before any conversion, so a 0.10-lot position moves about $0.10 per point. The calculator states the result in USD and then converts to KSh at a rate the reader supplies, because we are not given a fixed exchange rate to publish.

The position-size calculator works backwards from a risk amount the reader enters. It takes the KSh or USD amount the reader is willing to risk, divides it by the distance in pips between entry and stop, and then divides again by the per-pip value of one lot to return a lot size. Every output is rounded to two decimals and shown alongside the inputs used, so the reader can see exactly which reference price and which pip size produced the figure.

Where the numbers come from

Numbers come from the broker's own documents, read on a stated date. That includes the contract specifications, the leverage and margin terms that apply to the entity serving Kenya, and the platform list. We record the date each document was read so a reader can tell how fresh the figure is.

Where a figure is not published by the broker, the page describes what the cost consists of and what it depends on instead of inventing a number. For margin, the worked example we use is a 0.10-lot bitcoin position needing about $385.00 at the applicable cap. That is an illustration of the margin formula, not a promise about any account.

What happens when two sources disagree

When two sources disagree, the broker's own disclosure for the entity serving Kenya wins, and the conflicting figure is not published. A third-party comparison site, a forum post or an older copy of the broker's page does not override the current primary document. If the primary document itself is ambiguous, we narrow the claim to what it clearly supports, or we drop the claim and say the figure depends on the account and the entity rather than guess.

For calculated figures, disagreement is resolved by showing the arithmetic. If a reader's own result differs from ours, the inputs are compared first: the reference price, the lot size, the pip size and the leverage cap. Because one pip is 1.0 and one lot is 1 BTC, most differences trace back to a different price or a different cap, not to the formula. We publish the inputs next to the output precisely so a mismatch can be located rather than argued about.

Where the disagreement is about cost, we do not resolve it with a number at all. We are not given a spread, commission, swap or minimum deposit figure, so when sources conflict on cost we say the cost depends on the instrument, the account type and market conditions and point the reader to the broker for the actual charge. Publishing a contested cost figure would be worse than publishing none, because it would look precise while being unverifiable.

The limits of our testing

The desk tests with small accounts and reads documents; it does not run live money at scale, and it cannot see how a broker behaves in every market condition. A platform that works smoothly in a quiet week may behave differently in a fast one. We say so rather than implying the testing is exhaustive.

Leverage is described as a cap, not a setting to aim at. The maximum available to a Kenyan reader is the cap that applies to the entity serving this country, and using all of it is a choice with real risk attached. Where our testing simply could not reach a conclusion, the page says that too.

What is refreshed automatically and what is reviewed by hand

Nothing on this site is refreshed automatically from a live market feed. The BTC/USD reference price of 77000.0 is a stored value, and every calculator output is computed in the reader's browser from that stored value plus the inputs the reader types. That means a figure can be internally consistent and still be out of date relative to the market, and we say so wherever a worked example appears. There is no streaming quote, no tick-by-tick update and no auto-refresh interval to rely on.

Broker facts are reviewed by hand. The platform list, the M-Pesa funding route, the regulator caveat and the leverage cap for the entity serving Kenya are checked against the broker's own published material by an editor, not scraped or synced. A hand review is slower than an automated feed, and it is the reason we date each broker fact rather than implying it is current. When a fact cannot be confirmed against the broker's own disclosure, it is removed rather than left standing on an old value.

Calculated outputs are not stored at all, so there is nothing to refresh. The margin, pip-value and position-size results exist only for the moment the reader presses calculate, using the inputs on screen. If the reader changes the reference price, the lot size or the risk amount, the output changes with it. This is deliberate: it keeps the arithmetic transparent and prevents a stale cached figure from being mistaken for a live one.