Methodology: how MacroSetup scores currencies

What a score means

Every currency gets a score from 0 to 100. A higher score means the macro backdrop supports that currency more than the others this week. Scores are relative: they are meant to be compared with each other, not read as a forecast of how far a currency will move.

The four pillars

A score is the sum of four pillars. Each pillar has a maximum number of points.

Risk sentimentup to 40 points

How markets are treating risk right now, read from equities, volatility, credit and commodities. Some currencies tend to gain when investors turn cautious, such as JPY, CHF and USD. Others tend to gain when investors are confident, such as AUD, NZD and CAD. This pillar scores each currency against the current mood.

Fundamentalsup to 35 points

The central bank’s stance and the direction it is moving in, the level of interest rates, the economic data the central bank is focused on (inflation, the labour market or growth), and the fiscal position.

Geopoliticsup to 20 points

Trade policy, conflicts, elections and political risk, and energy shocks that affect a currency.

Price strengthup to 5 points

How the currency has traded so far this week against the other majors. It is kept small on purpose: it can confirm or question the macro view, but never overrule it.

Reading a pair

A pair’s score is the score of the first currency minus the score of the second. On 17 September 2026 EUR scored 46 and USD scored 72, so EUR/USD scored 46 minus 72, which is -26: strongly bearish. The larger the gap, the stronger the label, from neutral through slightly bullish or bearish to strongly bullish or bearish.

The breakdown per pillar shows what drives the gap. Positive numbers favour the first currency, negative numbers favour the second.

Setups and their status

The top 3 setups are the pairs with the widest gap between a strong and a weak currency, adjusted for whether this week’s price action and conviction agree. The list avoids repeating the same currency where possible, so it doesn’t show three versions of the same trade. Every setup carries a live status:

Confirmed by price

Price action this week is moving in the direction of the macro view.

News ahead

High-impact news for one of the two currencies is due soon, so the picture can change quickly.

Stretched

The pair has already moved unusually far in a short time, so chasing the move carries extra risk.

Price against

Price is moving against the macro view.

Neutral

Price has not picked a clear direction yet.

Setups are research, not trade instructions. They come without entry, stop or target levels.

Conviction

Each currency carries a conviction label: high, mid or low. It shows whether recent price action agrees with the currency’s rank. Low conviction means the market is not confirming the macro view yet, so that currency is treated with more caution when setups are selected.

Event risk

High-impact releases, such as central bank decisions, inflation, jobs and GDP data, are flagged ahead of time. The flag appears on the currency and on every setup that includes it, so you can see when a position would be exposed to a scheduled release.

Time horizon and updates

The view covers the current trading week, Monday to Friday. A full outlook is published every Sunday, before the market opens. After that the research is checked every hour on trading days and updated as soon as new data or news changes the picture, and every change is logged with the reason behind it.

Sources

The outcome of a rate decision or a data release is taken from the bank or the statistics office that published it. These are the sources the scores are built on.

Central banks

Rate decisions, statements, minutes and votes, from the bank itself: Federal Reserve, European Central Bank, Bank of England, Bank of Japan, Swiss National Bank, Reserve Bank of Australia, Reserve Bank of New Zealand and Bank of Canada.

Statistics offices

Inflation, jobs and growth figures, from the office that publishes them: U.S. Bureau of Labor Statistics, Eurostat, Office for National Statistics, Statistics Bureau of Japan, Swiss Federal Statistical Office, Australian Bureau of Statistics, Stats NZ and Statistics Canada. Every release and its publisher is listed on the economic calendar.

Exchange rates

The ECB's euro foreign exchange reference rates, for how each currency has moved against the other seven over days and weeks. Source: ECB statistics.

Market prices

Hourly prices of the seven dollar pairs from a market data feed, for this week's price strength and the London opening prices the track record is measured on, and daily prices of stocks, volatility, credit and commodities, read for risk sentiment.

News

News coverage gives context, never the outcome itself. When a bank's own site cannot be reached, a figure is only used once two independent live reports agree on it, and it is marked as being checked until the source confirms it.

Score changes in the update log link to where the information came from.

Isn’t this data free?

Most of it is. The central banks publish their decisions, the statistics offices publish their figures, economic calendars are free on many sites, and the CFTC publishes futures positioning once a week. What none of them does is put the eight currencies side by side and say which way the balance leans this week, and why. That is the part MacroSetup adds:

• One score from four pillars. Risk sentiment, fundamentals, geopolitics and price strength, weighted 40, 35, 20 and 5, add up to one number per currency, so the eight can be ranked and the 28 pairs compared.
• Releases read against expectations. A figure on its own says little. It counts by how it compares with the forecast, and by whether it is the data the central bank says it is watching.
• Every trading day. The research is checked every hour on trading days and the scores move when the news does, not once a week.
• The reason next to every score. Each currency carries a line on what drives it, so you can disagree with the reasoning and not just with a number.
• A record you can check. Every day's top 3 setups since 15 September 2026 are measured against price on the track record, misses included.

We leave the CFTC's positioning report out. It describes futures positions from the Tuesday before, and we have not been able to show that it adds anything to a daily reading. How a release turns into points stays in-house; see what stays private.

How results are measured

Every trading day, the top 3 setups of that day are measured at the opening of the London session, 08:00 in London: from the first opening after they were published to the next one, so a list from Friday runs to Monday and gets its result on Monday morning. A setup counts as right when the pair moved in its direction, and the size of that move is shown with it. When a newer list starts from the same rate, the newer one counts, so every move counts once. The current model has been live since 15 September 2026, and every day is shown, including the misses.

How we write and correct

Who writes what

Mike, founder of MacroSetup, built the model, decides what goes into it and checks what is published.

The daily research follows a fixed process on a fixed schedule. It reads the central banks, the statistics offices and the news, and our own scoring code turns that reading into the scores, the same way every time.

The weekly and monthly reviews are drafted by the research system in the same way. A review goes live by itself once every check below passes, and Mike reads it afterwards. Each review says so at the top.

How figures are checked

Before a review goes live, a script recounts every figure of our own in it against the track record, to the hundredth of a percent. It holds the piece back if a number differs, if a link to one of our pages leads nowhere, if no primary source is linked, or if the text slips into the model's internal terms. Results on the track record are worked out by code from the London opening prices, never typed in by hand. Outcomes of decisions and releases come from the sources above.

Reporting an error

Spotted a mistake? Email info@macrosetup.com with the page and what you think is wrong. When we fix an error in a fact or a figure, the page gets a dated note that says what changed. In the daily research, a correction goes into the update log with its date and the reason.

How MacroSetup makes money

From subscriptions. Partners, such as trading educators, get a share from us of what each customer they bring pays in their first year, and they have to say so next to their link (see the partner terms). A member who invites a friend gets a free month once that friend has paid for two months. No broker pays us and no page is sponsored. Some prop firms pay a commission when a reader buys a challenge through our link on that firm's page or our page of discount codes, never anywhere in the research; see the affiliate disclosure.

What stays private

The exact scoring rules, thresholds and formulas, and the instructions the research routines follow. That protects the research from being copied, while the reasons behind every score stay visible to members.

Limitations

A macro score describes the backdrop, not the timing. Currencies can move against the macro view for days or weeks, and unexpected news can change the picture at any moment. Use the research as one input next to your own analysis and risk management. MacroSetup is not investment advice.