Skip to content

What moves the market

Level 10, the last. Assumes paper to real money.

The idea

Every level so far has treated price as a series to be measured. This one asks why the series does what it does — because a rule is an assumption about behaviour, and knowing what produces that behaviour tells you when the assumption has stopped holding.

This level bends the rule of this section

Every lesson here ends in a study, and most of what follows Arvo cannot test: it keeps bars, dividends, quotes and chains, not interest rate expectations or fund flows. It earns its place anyway, because the testable question is not "does macro predict prices" — it is "did my rule's behaviour change when its conditions did", and Arvo measures exactly that. The limit is named rather than hidden, which is the house rule that outranks the other one.

Four things that actually move price

1. The discount rate

Any asset is worth its future cash flows, valued in today's money. The rate used to do that valuing is built on what interest rates are expected to be. Change the expected rate and every valuation changes at once, in the same direction.

This is the most important mechanism on the page, and it explains something you have already met. In relative strength you read that positions selected by the same criterion fail together. Rate-driven moves are why diversification across names fails exactly when you need it: the names are different, the discount rate is shared.

Long-duration assets — those whose value is mostly in distant cash flows — move most. That is why an unprofitable growth company and a government bond can fall on the same day for the same reason.

2. Expectations, not news

Price responds to the difference between what happened and what was expected, not to whether it was good.

A company reporting record profits can fall 8% on the day, because the market had priced in better records. This is the single most common source of confusion for new traders: "the news was good and it went down". The news was good and the expectation was better. The expectation was already in the price.

The practical consequence: by the time a fact is known to you, it is known to everybody, and the price you can trade at already contains it.

3. Flows — people who must trade

The mechanism behind several of the families in level 6, and the most underrated item on this page. Some participants transact regardless of price:

Who Why they must
Index funds a name enters or leaves the index; they buy or sell it, at whatever it costs
Funds meeting redemptions investors want cash; positions are sold to raise it
Option dealers short options must be hedged as price moves, mechanically
Companies buying back stock an authorised programme buys on a schedule
Anyone stopped out a stop is an order that fires without regard to value

None of these are opinions about value. They are constraints, and they push price away from where two-sided trading would have left it. This is what you are paid for in mean reversion and in selling options — supplying what a constrained participant needs, at a moment that suits them and not you.

4. Positioning

When a trade becomes crowded, the people in it are the people who would have to sell. An unwind is then self-reinforcing: selling triggers stops, stops trigger selling.

This is the mechanism behind the momentum crashes in relative strength and the correlated losses in selling options. Both are the same event seen from different positions — a crowded trade unwinding faster than it accumulated.

Underneath: microstructure

Below the level of why price moves is the question of how it moves — the mechanics that decide what your fill actually is.

Market makers quote both sides and profit from the spread, and they widen that spread when uncertainty rises. That is why costs are worst exactly when your rule most wants to trade, which you met in markets and paid for in opening range.

You do not need to model any of this. You need to know it exists, because it is where your assumed fill and your actual fill part company — and Arvo measures that gap as divergence rather than leaving you to wonder.

Why regimes change

The honest answer to the question level 3 left open. A regime is not a weather pattern that arrives for no reason — it is what the four mechanisms above look like from the outside:

  • A trend is usually a repricing that takes time: information diffusing, a flow completing, a rate expectation shifting over months.
  • A range is usually the absence of any of those: no new information, no constrained participant, nothing to reprice.
  • A regime change is one of them starting or finishing.

Which is the useful takeaway of the whole ladder: your rule does not have an edge in the abstract. It has an edge while a particular mechanism is operating, and the mechanism has causes you can at least name.

What Arvo can and cannot see here

Cannot: rate expectations, earnings surprises, fund flows, positioning data, index rebalancing schedules. There is no data layer for any of it, and a rule needing it has nowhere to read it from. Do not try to encode a macro view in a ruleset — you will be encoding your guess about a series Arvo cannot verify.

Can: the consequences, and this is the part to use.

What you want to know What Arvo gives you
Did the conditions my rule needs go away? the regime divergence reason: most entries, at least three, in a regime the finding never traded in
Did my rule stop firing? the frequency reason: fewer than a quarter, or more than four times, as many entries as were due
Did my costs change? the execution reason: slippage past twice what the cost model assumed
Was my result an average of two different markets? the regime split on a finding

A rule that fires a quarter as often as it did is a regime change before it is a loss. That sentence is the whole practical value of this level: the mechanisms are untestable here, and their footprints are measured, named, and raised as an alert while there is still time to act.

What people get wrong

Trading the news. By the time you read it, it is in the price. What is not in the price is the part nobody has worked out yet, and you are unlikely to be first.

Assuming a reason means a direction. "Rates are rising, so stocks fall" is a mechanism, not a trade. The market has the same information and has already acted on it.

Explaining a loss with a narrative. After any drawdown you will find a plausible macro story. It is indistinguishable from coincidence and it will make you change a rule for the wrong reason. The divergence reasons exist so you have a measured answer instead of a story.

Watch

  • (to be chosen) — how interest rates flow into valuations.
  • (to be chosen) — index rebalancing and forced flows, with a worked example.
  • (to be chosen) — market makers and why spreads widen.

You have finished the ladder

Ten levels, and the thread running through all of them: a claim is only worth as much as the test behind it.

Where to go now: