MovePithWHY PRICES MOVE

Why bitcoin rises: which layers are usually at work

This page doesn't tell you why it went up today — that kind of page is dead in three days. It tells you what traces each of the four layers leaves on the way up, and how to work out from those traces which one is doing the pushing. Same method, any day.

L1–L4 OverviewLu ZhiyuanUpdated 1556 words / 7 min read

Fix the question first

“Why is bitcoin up” smuggles in an assumption: that rising is unusual and requires a reason. In a market that never closes and has no daily limits, a few percent is ordinary and needs no explanation.

The version of the question worth asking is: “Did this move exceed the normal range in size or speed? If so, which layer fits?” Separating everyday noise from events that need explaining saves you a great deal of wasted searching.

What each layer looks like on the way up

Layer 1 — supply: slow, but directionally stable

Changes on the supply side (less new issuance, long-term holders stepping back from selling, exchange balances falling) don't produce a one-day surge. They produce weeks or months of steady climbing: a little each day, shallow pullbacks, no dramatic expansion in volume. If that's the shape in front of you, go look at exchange balance trends and long-term holder behaviour, not at today's headlines.

Layer 2 — macro: rising alongside other risk assets

One action identifies it: look at the equity indices. If technology stocks are up and the dollar is softening over the same window, this is very likely not a crypto story but an improvement in conditions for risk assets generally. The tell is that it feels like “everything is up” rather than “bitcoin is up”.

Layer 3 — leverage: the sharpest and the briefest

The upside version of the leverage amplifier is a short squeeze: rising prices trigger forced closes on short positions, and closing a short requires buying, which pushes the price higher, which triggers the next tier of shorts.

Its shape is unmistakable: a vertical rise inside minutes, a burst of volume, and usually a partial giveback afterwards. Verify it against liquidation data for the same window — if short liquidations dwarf long liquidations, this layer is the protagonist. Mechanism in how liquidation cascades amplify moves; current readings in funding rates and leverage.

Layer 4 — structural events: it matches a timestamp

A regulatory document is published, a capital channel opens, an institution discloses a holding. The test is whether there's a primary source and whether the start of the move matches its timestamp to the minute. A gap of hours means the causal claim needs questioning.

Three misreadings specific to rallies

1. Treating “people are buying” as an explanation

Every trade has both a buyer and a seller. “Buying pressure was strong” is true of every rally ever, which is why it explains nothing. What actually needs explaining is why sellers demanded a higher price before agreeing to trade — because sell orders thinned out, or because buyers became more urgent. Those two leave different traces in the data.

2. Reading market cap growth as money arriving

Market cap is circulating supply times the last trade price. A 10% rise in price is a 10% rise in market cap, and that does not mean a corresponding sum entered the market. “$50 billion was added to the market today” is arithmetically misleading.

3. Treating a confirmed positive as the cause

A story discussed for weeks is usually not new information on the day it's confirmed. If most of the advance happened before publication, the publication isn't the cause — the formation of the expectation was. See why the price already moved before you saw the news.

A quick self-check

Ask: if the price had gone the other way today, would my explanation still work? “Improving flows” and “better sentiment” read just as well in reverse, which is how you know they aren't explanations. Anything that passes this test contains evidence independent of price.

The four checks for a rally

  1. Is the whole market up?

    Majors rising together by similar amounts is systemic. Bitcoin alone rising usually points to its own structural event. A single small-cap rising has nothing to do with the market at all. The market heatmap answers this in one glance.

  2. Vertical or a climb?

    A minute-scale vertical move sends you to liquidation data (layer 3). Days of steady climbing send you to supply and macro (layers 1 and 2).

  3. What's happening outside crypto?

    Equity indices, the dollar, gold over the same period. Moving together means the macro layer; moving independently means you have to look inside crypto.

  4. Is there a primary source?

    An official notice, a regulatory filing, an on-chain record you can open yourself. “Sources say” and “analysts believe” don't count.

To look up indicators layer by layer, the layer explorer is quicker.

Three shapes a rally takes, and where each one gives itself away

Sorting by shape beats sorting by “reason”: the shape is an objective record that existed while the move happened, whereas reasons get attached afterwards. These three cover almost everything.

Shape one: the staircase

Days of small advances, shallow pullbacks, no meaningful expansion in volume. This is what the supply layer looks like — sellable supply thinning out, or steady bids quietly working through the resting offers.

Its tell is that it has no story. You'll notice the headlines assign a different reason to each day of the same staircase, and those reasons contradict each other. What's worth checking is exchange balance trends and unlock schedules, not today's news.

Shape two: the minute-scale vertical

Days of normal range covered in minutes, a burst of volume, usually a partial giveback. This is a short squeeze: rising prices force shorts to close, closing a short requires buying, and that buying pushes higher still.

Its tell is that the buying is mechanical and says nothing about demand. Check the long/short split in liquidation data over the same window — short liquidations far exceeding longs essentially confirms it. Whether such a move persists depends on real buyers turning up afterwards, which is a separate question.

Shape three: the rally that finished before the news

A story discussed for weeks finally lands, you open the chart, and the advance was already complete before publication — with the day of the announcement giving some of it back.

That isn't irrationality, it's ordinary expectation pricing: markets price the probability of a thing happening, so when it happens only the unpriced remainder is left to adjust. Mechanism in why the price already moved before you saw the news.

Diagnostic cues for the three shapes (observation directions only, no price judgement)
ShapeTime signatureData to checkUsually misread as
StaircaseDays to weeksExchange balance trends, unlock schedulesAssigned a different “reason of the day” each day
VerticalMinutesLong/short liquidation split, change in open interest“Demand surge” — actually forced buying
Expectation realisedWeeks before the announcementPrimary timestamp vs start of the move“Good news drove it” — the order is reversed

Volume: the one extra reading that separates the shapes

On price alone the three shapes can look similar on a daily chart. Add volume and they separate immediately:

  • Price up, volume flat — usually the staircase. No influx of new participants, just resting offers being absorbed.
  • Price up with volume spiking several-fold and then collapsing — the liquidation signature. The spike is involuntary execution, not an increase in willing buyers.
  • Volume spikes at the moment of an announcement but price barely moves — the news was already priced; what you're seeing is positions changing hands.

One caveat: volume is measured differently across venues (whether derivatives are included, whether self-trades are stripped), so comparing absolute figures between platforms is meaningless. Only relative change within one venue tells you anything.

Three recurring questions with shorter answers than you'd expect

“It went up — will it keep going?” We don't make that judgement and don't believe anyone makes it reliably. This article is a method for attributing a rise that has already happened; it does not extend forward. That isn't caution: attribution works with complete information and a known outcome, prediction faces an unknown one, and every public input you have is available to everyone else.

“Why can I sometimes find no reason at all?” Because many moves genuinely have no single cause — they're the aggregate of countless independent decisions. “No identifiable driver” is an honest conclusion, and more useful than an invented one, because an invented reason stays in your head as a faulty template for next time. In a market with no daily limits that never closes, a few percent needs no explanation.

“Does ‘institutions are buying’ count?” Only with a specific disclosure, flow figure or on-chain record attached. Otherwise it sits in the same category as “sentiment improved”: price up means institutions arriving, price down means institutions leaving, and no path could falsify it. The test never changes — is there evidence independent of the price itself?