Every market you follow, read and ranked before the US open.
By the time you sit down with a coffee, the work is already finished. Four models have read every name we cover and said where each one stands. You get their reasoning, how broadly they agree, and what changed overnight.
Every Monday it becomes a portfolio of ten names or fewer, with exact weights and cash for the rest. There is nothing left for you to interpret.
You read the same page I do, at the same time. I am Yimin Xu. I built this in London, drawing on extensive experience pricing two-way risk on a bank rates desk.
Everything published is listed there, including the free pieces, so you can see the shape of the work before deciding anything.
The hard part is rarely a shortage of information. It is having a process that still holds up in a bad month.
You may have arrived here after one of these three things has just happened.
At some point, a position quietly turns into a judgement on you.
This happens to everyone who has ever put a view into words. Once it has been said out loud, to a client, a committee, or only to yourself, part of you starts defending it. Supporting evidence gets read closely. The rest gets read for flaws. A losing position is held too long, because closing it concedes something. A winning one gets trimmed early, because banking it settles the question.
None of that is a character flaw. It is how most of us are built, which is why careful, experienced investors still do it. It rarely shows up in a review, and over a career it costs more than any single bad decision.
None of this is solved by reading more, and none of it is solved by resolving to try harder. What helps is a process that gives the same answer on a difficult afternoon as on an easy one, holds the view so nobody has to, and leaves a dated record either way.
Written for people who run capital to a professional standard, whether that capital is their own, a client’s, or a family’s.
I have been on the wrong side of that question, and it is not a pleasant place to answer from.
The hard part was never the loss. It was being asked why, and hearing myself reach for something that sounded like reasoning and was actually a feeling. I priced two-way risk on a G10 rates desk, where that answer does not survive. The lesson stuck. The edge is almost never the insight. It is whether the same work still happens on the difficult mornings.
I built the thing I could not buy. Four model families, a decade of testing on data they had never seen, a data pipeline, and the discipline to run it every day. At any serious scale that is a hire and a budget line. You are not being asked to trust my judgement. You get the machinery itself, the record it has already built, and the rules it has to follow even on the mornings when I would rather it did not.
The research operation a fund would build in-house, run to a desk standard every trading day and published in full where anyone can check it.
My professional background and my future vision →Three steps, and the first costs nothing.
There is no call to book, no discovery form and no enterprise quote. The published record and the construction notes are open precisely so this can be judged on the work rather than on the claims.
One service at two depths, and a companion portfolio for the parts of the cycle equities find hard.
Each level is the same research function at a different depth. The models produce the portfolio, and the research explains what it is holding through. The first question is how deep to go on Macro & Megacaps. The second is whether Commodities earns a place beside it, and for most subscribers it does.
Macro & Megacaps: Signals
The models, the portfolio, every morning.
Four models read all 73 names before the US open, from macro and rates to the Mag-7 and the crypto majors. Those signals build the portfolio, rebuilt weekly, with a daily check that sells anything that has turned.
- Daily multi-model signals on 73 names, before the US open
- The systematic portfolio: tickers and target weights, published weekly
- Backtested on a decade of data the models never saw
- Every coverage addition included, for as long as you stay
Macro & Megacaps: Signals & Research
A bad month is therefore never a surprise.
The portfolio will hold something through a bad stretch. It always does. Whether you stay with it comes down to whether you saw the risk coming, and a signals page cannot give you that.
- Everything in Signals, in full
- The macro note, twice a week: what kind of market we are in, where the money is flowing, the Fed path and the curve, then the indices, the Mag-7, metals and crypto
- Every read carries a structure, a bias, a measured target and the level that would prove it wrong
- Deep dives on the portfolio’s own holdings: the business, the numbers, the valuation, the risks
- Each one opens with why the models hold it, and what would drop it
Commodities: Signals
A companion portfolio for the parts of the cycle equities find hard.
24 names across metals, energy, agriculture and uranium, plus the producers behind them. It runs the same method on its own universe, with its own portfolio. That separation is the point of holding both.
- Daily multi-model signals on 24 names, before the US open
- The commodities systematic portfolio, published weekly
- Backtested on a decade of data the models never saw
- A third off when added to either Macro & Megacaps tier, at $2,000 rather than $3,000
Every subscription includes its portfolio at no extra cost. There has never been a separate fee for it. Your price is locked for as long as you stay subscribed. Prices exclude VAT and any other applicable taxes.
How my work helps, in the order you will notice it
Changing nothing is itself a position.
Very little in the left-hand column is carelessness. It is what happens when one person is quietly doing the work of a research team, and it is the normal outcome, not a failing.
- Another year in which the process is whatever there was time for that morning.
- Positions exited in the third week of a drawdown, for reasons that will not appear in any note afterwards.
- A mandate narrowed to what one person can realistically cover, with the gap left unpriced.
- A view that has to be defended, and a portfolio quietly shaped around defending it.
- Nothing to learn from, because nothing was written down before the outcome was known.
- One structured read a day, in the same shape, so one week is comparable to the last.
- A portfolio that reduces its risk on schedule and by rule, not by conviction on a bad afternoon.
- A drawdown that reads as the process working, because the risk was documented before it arrived.
- A view that belongs to the models, so changing it costs nothing but a line in the record.
- A dated, locked record that answers the question before it is asked.
The difference between the two columns is not intelligence, effort or access. It is whether the same work happens on the mornings when nobody would notice if it did not.
Independent means something specific here.
A lot of market research is produced next to trading, banking or advertising revenue, and those incentives quietly shape what gets said. I have kept this deliberately clean: no trading desk, no banking clients, no sponsors, no affiliate links. The only thing I sell is the subscription, so the work answers to one group, the people who read it.
Everything published is locked the moment it goes out. A day’s signals are frozen when they are sent and cannot be edited afterwards, and the portfolio can only hold a name the record already shows as long, a rule enforced in code rather than by good intentions. Downgrades are published the week they happen, including the ones that did not work, which is why the full record is free to read before you pay anything.
This will not be the right service for everyone, and it is better to say so here.
Being clear costs a few subscriptions and saves a lot of disappointment. If any of these sounds like you, something else will serve you better.
Someone who runs capital to a professional standard and wants the same work done on the difficult mornings as on the easy ones. The published record and the construction notes are open to everyone. Judge it from there.
This is what I run. You’re invited to read over my shoulder.
Nothing here tells anyone what to do with their money. That part stays yours, and so does the credit for it. This is the research operation I run for my own investing, published as I use it: the four models I read before every US open, the portfolio they build, the macro note I write twice a week, and the deep dives on the names that portfolio holds. A subscription is a seat beside that work while it happens, including the weeks it goes badly, because those are on the record too.
The work itself is closer to observation than to prediction. Each morning the models report where the market stands and how strongly they agree, the record shows afterwards what came of it, and the process keeps being refined as that evidence accumulates: which combinations hold up, which kinds of market each model reads badly, and what needs changing next. None of that asks anyone to be attached to a particular trade, mine included. The market is the thing being read, and both of us are watching it from the same side of the glass.