Most people buy crypto hoping it goes up.
That is one way to be in this market, and it has made people a great deal of money and cost them a great deal too. It has one feature that never changes: you are entirely dependent on the direction being kind to you.
We do something quieter. Every strategy we run is built so that the market's direction largely cancels out — and what is left over is a specific, identifiable payment that somebody in the market has a reason to make. A fee that a certain kind of trader must pay. A rate that somebody wants locked in. Interest that a borrower owes. These are not predictions, and they do not need anyone to be right about the future.
Nothing here guesses a price
Same discipline as the rest of this site. We are not in the business of telling you where anything is going.
Several sources at once
The return is drawn from several unrelated places at once, so no single one of them has to work for the whole to hold up.
Slow is the point
These are designed to be dull. The aim is a return that turns up steadily rather than one that arrives all at once.
Getting paid for something other than being right.
The return comes from a few broad places, described here in outline. Enough to judge whether the logic holds — how we do it, and with what, is ours.
We are paid by people in more of a hurry than us
Crowded positions are expensive to hold. That expense is somebody else's income.
Much of this market runs on borrowed money, and impatience in it carries a running cost — paid continuously, for as long as the position is held. That money does not disappear; it transfers to whoever is willing to sit calmly on the other side of the crowd. We arrange to be that party, with our own exposure to the underlying price deliberately cancelled out, so what we keep is the payment rather than the movement. This is not a mispricing that closes once people notice it — it is simply how these markets are built, and it has been paying for as long as they have existed.
Income that is contracted rather than hoped for
A return you know at the moment you commit, not one you find out later.
A substantial part of what we run earns in ways that are settled before the fact: obligations that pay a known amount on a known date, and dollar balances placed where they are lent out against more collateral than has been borrowed. Nothing in this group asks anyone to be right about anything. It asks for patience, and for knowing where the terms are better than average — which is a question of doing the work, not of having a view. It is the least exciting thing we do and, deliberately, the part we lean on hardest.
Where we do take a position, we take the direction out of it
Held as one thing against another, so the market's mood cancels itself.
Some of what we run is a genuine position rather than a payment collected. Even then it is expressed in matched, offsetting amounts, so that a general rise or fall affects both sides equally and nets to nothing — leaving only the relationship between them, which is the part we actually studied. Money moves through this market in currents rather than all at once, and those currents persist long enough to be worth positioning for without ever needing a call on the market as a whole.
Alongside these, anything not currently committed is held in short-term dollar form and earns while it waits. None of these sources can absorb an unlimited amount, so there is always a balance between them — and letting it sit idle would quietly cost more than most of the decisions above.
The shape of it, honestly.
This is not the part of crypto that doubles. Returns here are the sort you would describe in single digits a year, arriving in small pieces rather than in one dramatic move, with an account that is meant to stay fairly level along the way rather than swing with the market.
The trade-off is worth being clear about. When crypto is running hot, this will look slow next to simply owning it — and we would never claim otherwise. What it offers instead is a return that does not depend on that run happening, and which carries on doing its job in the long, flat, unglamorous stretches where most of the calendar actually lives.
Largely irrelevant
The point of every structure above is that up and down cancel, leaving the payment behind.
Steady, small, repeated
Many small contributions from unrelated sources, rather than one position that has to come good.
Capacity, not conviction
Each of these can only hold so much before it stops working. We size to that, not to enthusiasm.
The same standard we hold the rest of this site to.
Everything on the front page of Predicter is published with a live scorecard attached, because we think anyone making claims about markets should be checkable. We run money the same way. An idea has to state in advance what would count as working and what would count as failing; it is measured against the return on plain cash rather than against zero; every cost it incurs is charged to it before we call it a success; and it has to earn its place against ideas we have already discarded.
What is described above is what survived that process. Far more ideas have not survived it than have — and we would rather tell you that than present a list that looks effortless.
If this is the kind of thing you'd want to hear more about.
This page is deliberately a sketch. The full picture — what each one earns, what it risks, how long it has been tested and where it falls short — is a conversation, not a web page. Leave your address and we'll come back to you. We won't send anything automated, we won't share it, and you can ask us to remove it at any time.