Methodology

This page explains how My Coin Partner produces a comparison, what we weigh, and what we deliberately leave out. If something on this site contradicts this page, this page is the one we are wrong against — tell us and we will correct it.

Does commission affect our rankings?

No. Ordering is produced by the rubric below, and commission is not one of its inputs. We hold this position for a specific structural reason: the moment commission enters an ordering, every other claim on the site becomes unverifiable, because a reader has no way to tell which parts were editorial and which were bought.

If that ever changes for a particular page, that page will say so in those words, in its Partner Declaration, above the body — not here, and not in a footer. A general policy statement is not a disclosure; a disclosure names the specific page and the specific relationship.

The rubric

Platforms and programs are scored on six dimensions. Weights are fixed before any option is scored, so a weighting cannot be chosen after the fact to produce a preferred winner.

• Cost — 30%. The all-in cost of using the platform: trading fees at realistic volumes, deposit and withdrawal fees, spread where it is material, and any conversion charge. For partner programs, the equivalent is the commission model and rate, read against the payout terms that determine whether the headline rate is achievable.
• Terms clarity — 20%. Whether the operator publishes its own terms in a form a reader can check. A platform that publishes a complete fee schedule scores above one that requires an account to discover its costs. This dimension rewards disclosure, which is the one thing we can assess without privileged access.
• Custody and control — 20%. Who holds the assets, what happens in an insolvency, whether withdrawal is unconditional, and what the operator has actually committed to in writing rather than in marketing copy.
• Jurisdiction and eligibility — 15%. Which markets are served, what verification is required, and how clearly restrictions are stated before a reader invests time in signing up.
• Track record — 10%. Documented incidents, outages, enforcement actions and how the operator responded. We weight the response as heavily as the incident.
• Support and recourse — 5%. Whether a human can be reached and whether there is a stated complaints route.

How figures get onto a page

Every fee, commission rate, cookie window and payout term we publish is read from the operator’s own published terms and carries the date it was read. Two rules follow from that, and they are absolute:

• An unknown renders “Not disclosed”, never an estimate. If an operator does not publish a figure, the cell says so. We do not infer it from a competitor, from an older version of the page, or from what would be typical.
• A figure without a source and a date does not render at all. This is enforced in the code that draws the tables, not left to editorial discipline.

Terms change without notice. A dated figure tells you how much to trust it; an undated one tells you nothing. Always check the operator’s current terms before acting.

What we do not test

This section matters more than the rubric, because it is the part most comparison sites omit.

• We do not conduct security audits. Nothing here is an assessment of whether a platform is safe to hold assets with. We report documented history; we cannot report what has not surfaced.
• We do not verify reserves. A published proof-of-reserves attestation is evidence about a moment in time and says nothing about liabilities unless it is paired with them.
• We do not test execution quality at size. Our cost figures come from published schedules, not from trading through the book.
• We do not assess whether a platform suits you. Eligibility, tax treatment and risk tolerance are yours to establish.
• We do not verify identity claims about operators beyond what a licence register or the operator’s own filings state.

Corrections

When we get something wrong we fix it, date the fix, and note it on the page and in our public corrections log. We do not quietly edit a figure and leave the page looking as though it was always right.

Where our market data comes from

Prices, market capitalisations, percentage changes, supply figures and rankings are read from CoinLore. Price charts and sparklines are drawn from Binance daily and four-hourly closing candles. The Fear & Greed Index comes from Alternative.me. Every figure on the site carries the provider it came from and the time it was read.

We are naming the providers we actually use rather than a longer list we do not. Two commonly-cited sources are deliberately absent: CoinGecko currently rate-limits our server, and our CoinPaprika quota is exhausted. Neither supplies any figure on this site, and we would rather say so than imply broader sourcing than we have.

Refresh intervals

• Market table and prices — cached 5 minutes.
• Price charts — cached 1 hour.
• Fear & Greed — cached 30 minutes.
• Verified trading pairs — cached 24 hours.

A figure older than ten minutes is labelled as stale on the page rather than presented as current. Caches are refreshed by a scheduled job rather than by visitor traffic, so a quiet period does not leave the site showing old numbers.

What a blank means

This is the most important thing to understand about our market pages. When we cannot source a value, we render nothing at all — not a zero, not a dash, not an empty row. A blank price means we could not retrieve it. A missing chart means the asset has no verified trading pair on our chart provider, not that its price was flat. A module whose data is entirely unavailable does not appear on the page rather than appearing empty.

This is enforced in the code that renders the figures, not left to editorial care: every formatter returns an empty string when handed a missing value, so there is no code path that can print an invented number.

Known limits of these figures

• Charts reflect one venue. Candles come from a single exchange pair, so they will not match a blended global price, and they show daily closes rather than intraday highs and lows.
• Trading pairs are verified, never guessed. We resolve each asset against the exchange’s live list of trading pairs. We never assume a pair exists by appending a quote currency to a ticker — a wrong pair would print a confident, wrong price.
• Market capitalisation is price multiplied by circulating supply. It is a comparison tool, not a measure of money invested, and circulating supply is itself a provider estimate.
• Providers occasionally report zero where they mean “unknown”. We pass through what the provider reports rather than second-guessing it with a heuristic, because a rule that blanks every zero would also erase a genuinely flat reading. If a figure looks implausible, treat the provider as the authority and tell us.