The division of labour
Macro can tell you which of two currencies has the better case this week: which central bank is leaning which way, which economy is surprising, which one the market runs to when it gets nervous. What it cannot tell you is where price is, what it has just done, or where you would be wrong.
A chart answers exactly those three and nothing else. It has no idea that the Reserve Bank meets on Tuesday. Put them in the wrong order, and you either trade a good level against the macro or hold a good macro idea through a level that should have stopped you out.
What that looks like in a week
- Monday morning: the side. Read the ranking and the reasons, and write down which currencies you are willing to be long and which short. The pairs you care about are the ones where a strong meets a weak.
- Then: the level. Only on those pairs, look for the entry you would have taken anyway. If there is no level, there is no trade, however good the macro reads.
- Before the entry: the calendar. Check what is due for both currencies. A setup that is right and badly timed against a release is still a loss.
- After the entry: the news, not the chart. The thing that ends a macro trade early is a change in the macro. A bias that has flipped is a reason to leave, whatever the chart says.
When they disagree
They will, often. A currency the macro likes can be in a downtrend, and a clean level can appear on a pair where both currencies look the same. The answer is not to pick a winner but to trade less: no level means no trade, and no macro means a smaller one at best.
Sitting out is the cheapest position available and the one most traders use least. On a quiet week, where every currency score sits in the middle of the range, that is usually the correct read of the market rather than a failure to find something.
What our own numbers say about this
Two things we can put a figure on, from our own record rather than from a theory.
The first: a macro pick is not a one day event. Since January 2024 a setup stayed in our top 3 for a single morning about two thirds of the time, and the runs that lasted four or five mornings were the ones that carried the most while they lasted. The move gave back in the days after a setup dropped out, which says the exit signal is the list itself rather than a fixed number of days.
The second: the day by day hit rate of the mechanical part of our model sits close to a coin flip, and holding those same picks for ten or twenty days made it worse rather than better. That is why we describe the macro as an edge in selection, not a system, and why the entry is left to you. Every day of it is published on the track record, misses included.
Where risk sits in all this
With you, entirely. The stop belongs to the chart, the size belongs to your account, and if that account is a funded one, the daily loss limit belongs to your firm. Nothing in MacroSetup has an entry, a stop or a target in it, and the day a macro service starts telling you where to put your stop is the day it has become something else.
What a bias is, and what goes into one, is on the fundamental bias page.