Coinpric

Blog

  • “Government adoption” of crypto means six different things

    “Government adoption” of crypto means six different things

    The short version

    “Country adopts crypto” can describe six unrelated actions: holding it as a reserve asset, accepting it for payments, granting it legal tender status, licensing an industry, using the technology for a government function, or issuing a state digital currency. Only two of those are anything like endorsement, and a central bank digital currency is arguably the opposite. The distinctions decide whether a headline matters.

    Few phrases in crypto coverage do less work than “government adoption”. It is applied to a treasury purchase, a licensing regime, a land-registry pilot and a central bank project, as though these were points on one scale. They are not even in the same category.

    Here are the six things the phrase actually describes, in rough order of how strong a signal each represents.

    1. Holding it as a reserve or treasury asset

    A state, or a state-controlled fund, holds a crypto asset on its balance sheet. This is the strongest version of adoption, because it involves the government taking price risk with public money.

    What to check before treating it as significant: is the holding purchased or seized? Governments accumulate substantial crypto through law enforcement, and a seized holding is not an investment decision — it is an asset awaiting disposal, and the disposal is a supply event. Also check the size relative to total reserves. A headline-friendly absolute number can be a rounding error in a sovereign portfolio.

    2. Accepting it for payments

    A government agency will accept crypto for taxes, fees or fines. Modest but real, because it creates a genuine use.

    The detail that decides how meaningful it is: does the state hold what it receives, or immediately convert it? Most such schemes convert at the point of receipt, often through a payment processor who bears the price risk. In that arrangement, the government has accepted a payment method, not the asset — and the economic exposure sits with a private intermediary. Useful for users, much less significant than it sounds.

    3. Granting legal tender or equivalent status

    The rarest and most legally consequential. Legal tender status means the asset must generally be accepted in settlement of debts. It has knock-on effects across contract law, accounting and taxation, and it usually requires primary legislation.

    Distinguish it carefully from three weaker things that get reported identically: permitting use, which most jurisdictions already do; recognising it as property or an asset class, which is classification rather than endorsement; and declaring it acceptable for specific purposes only. Only the full version is legal tender, and coverage very often uses the term for the weaker cases.

    4. Licensing and regulating an industry

    The most common form, and the most consistently misread. A jurisdiction creates a licensing regime for exchanges, custodians or issuers, with capital requirements, custody rules and reporting duties.

    Is this adoption? It is legitimisation — a clear signal that the activity is expected to continue and be supervised. But it is not endorsement of any asset, and it frequently makes life harder for firms in the short run, since compliance costs are real and some participants exit. Reported as “country embraces crypto”, it is better read as “country decides who may operate”.

    5. Using the technology for a government function

    A land registry, a supply-chain record, a credentials system, a procurement audit trail. These use distributed-ledger technology without necessarily involving any crypto asset at all.

    The question that settles the significance: is it a public permissionless chain, or a private ledger run by the agency? A private permissioned ledger is a database with extra steps and unusual governance. It may be a perfectly sensible piece of public administration. It says essentially nothing about crypto as an asset class, and conflating the two is the most common error in this whole category.

    6. Issuing a central bank digital currency

    A digital form of the national currency, issued by the central bank. Frequently filed under crypto adoption, and arguably the opposite.

    A retail central bank digital currency is a liability of the central bank, centrally issued, and typically designed with identity requirements and the technical capacity for transaction-level oversight. That is a different object from a permissionless asset with no issuer — in several respects a competing one. A state can be building one while restricting crypto assets, and several are.

    The comparison worth keeping

    Action Is it endorsement? The question that decides
    Holding as reserve Yes, strongest Purchased or seized? What share of reserves?
    Accepting for payment Partly Held, or converted instantly by a processor?
    Legal tender status Yes, and legally deep Full status, or merely permitted?
    Licensing an industry Legitimisation, not endorsement Does it raise or lower barriers in practice?
    Technology for public services Essentially unrelated Public permissionless chain, or private ledger?
    Central bank digital currency Often the opposite Does it compete with, or use, crypto assets?

    Two more filters

    Announcement versus implementation. Pilots, memoranda, working groups and strategies are announcements. Many never ship, and the announcement is reported far more loudly than the quiet cancellation. Ask whether anything is actually running.

    Reversibility. An executive decision can be undone by the next administration. Primary legislation is harder to unwind. Where a change sits on that spectrum matters more for anything long-term than the size of the initial move.

    None of this is a view on whether any of it is good, or a prediction about prices. It is a way to tell which of six very different things a headline is actually describing — which is usually the whole question.

    Key takeaways

    • Six unrelated actions share the “government adoption” headline, and only two resemble endorsement.
    • Seized holdings are not investment decisions — they are pending supply.
    • Most payment-acceptance schemes convert instantly, so the state accepts a method rather than the asset.
    • Legal tender status is rare and legally deep; “permitted” and “recognised” are much weaker and reported identically.
    • Licensing regimes legitimise an activity and often raise costs. That is not endorsement of any asset.
    • A private permissioned government ledger says almost nothing about crypto as an asset class.
    • A central bank digital currency is a central bank liability — frequently a competitor to permissionless assets, not an adoption of them.
  • Crypto tax: the four questions every jurisdiction asks

    Crypto tax: the four questions every jurisdiction asks

    The short version

    Almost every tax system asks the same four questions of a crypto transaction: what kind of thing is it, has a taxable event happened, what was the gain in your own currency, and what type of income is it. The answers vary enormously by jurisdiction. The questions barely vary at all — and the most-missed answer is that swapping one crypto asset for another is usually a taxable disposal, even though no conventional money was involved.

    Crypto tax writing tends to be either jurisdiction-specific and quickly stale, or so general it says nothing. There is a more durable middle: the four questions that nearly every system works through. Learn them and you can read your own rules — which you will have to do anyway, because nobody can do it for you from here.

    This is general information about how tax systems approach crypto, not tax advice, and it does not describe your obligations. Rules differ by jurisdiction and change frequently. Consult a qualified tax professional where you are resident.

    Question 1: what kind of thing is it?

    Everything follows from classification, and the label is rarely “currency”. Common treatments include property or a capital asset, an intangible asset, a financial instrument, or — narrowly, in a few places — a form of money.

    This matters more than it sounds. If a crypto asset is property, then spending it is disposing of property, and disposing of property is generally a taxable event. That is why using crypto to buy something can trigger tax while paying with a bank card does not: the card moves money, and money is not property being disposed of.

    Question 2: has a taxable event happened?

    Systems distinguish acquiring and holding from disposing. Holding, in most places, is not itself an event. Disposal usually is — and “disposal” is broader than selling.

    Action Commonly treated as
    Buying with conventional money Not an event; establishes your cost
    Holding, however long Not an event
    Selling for conventional money Disposal
    Swapping one crypto asset for another Disposal — usually taxable
    Spending it on goods or services Disposal
    Moving between your own wallets Not an event; ownership unchanged
    Receiving it as payment or reward Income, at the value when received

    The fourth row is the one that catches people. Swapping asset A for asset B feels like rearranging a portfolio, and no conventional money appears. But in a system where crypto is property, you disposed of A — and the gain on A is measured and taxed even though you never saw a bank transfer. Someone who trades actively between assets can accumulate substantial liabilities without ever withdrawing anything.

    The last row has a second sting. Receiving crypto as income is generally taxed at its value on receipt, and that value becomes your cost for the asset. If it then falls before you sell, you may owe income tax on a figure well above what the holding is now worth.

    Question 3: what was the gain, in your own currency?

    Gain is proceeds minus cost, both expressed in your national currency at the time of each transaction. Two hard parts.

    Which units did you sell? If you bought the same asset at several prices, something has to decide which ones you disposed of. Systems prescribe different methods — first in first out, average cost, or specific identification. This is not a detail: on the same trades, different methods produce materially different taxable gains.

    What was it worth at that moment? A crypto-to-crypto swap has no national-currency figure attached, so one must be established for both sides at the time of the trade. This is where record-keeping becomes the whole game, and why reconstructing years of history afterwards is so painful. Our profit calculator can help you work through an individual trade’s arithmetic, but it is not a tax engine and does not know your jurisdiction’s rules.

    Question 4: what type of income or gain is it?

    Most systems tax capital gains and ordinary income differently — often at very different rates, with different allowances and different loss rules. So the character of the receipt matters as much as the amount.

    Recurring points of difficulty: staking rewards, which may be income on receipt or only taxed on disposal; mining, which may be a business activity with deductible costs or passive income; airdrops, which may be income at receipt even though you did nothing to earn them; and lending or liquidity provision, where the return may be interest, a gain, or something the rules never anticipated.

    Frequency and intent can also change the character. Someone trading constantly may be treated as carrying on a business rather than making investments, which changes the rate, the deductions and the reporting.

    Losses: the part people forget to claim

    Losses are usually recognised on disposal, and can often offset gains. But the rules are specific: what a loss can be set against, whether it can be carried forward, whether it must be claimed within a window. Many people who report gains diligently never claim allowable losses, which means paying tax they did not owe.

    A separate trap: an asset that has become worthless or unreachable — a failed project, a lost key, a collapsed venue — has usually not been “disposed of” in the technical sense. There is often a specific procedure for claiming relief, and it is not automatic.

    What to do, practically

    Keep records as you go, not afterwards. For every transaction you want the date and time, what was disposed of and received, the value of both in your own currency at that moment, the fee, and the venue. Reconstructing this later from exchange exports across venues that have since changed or closed is the single largest avoidable cost in crypto tax.

    Then find your own jurisdiction’s primary guidance — most tax authorities publish crypto material directly — and answer the four questions against it. That is a genuinely tractable afternoon, and it is far better than assuming.

    Key takeaways

    • Four questions drive nearly every system: what kind of thing, was there an event, what was the gain, what type of income.
    • If crypto is property where you live, spending it is a disposal — which is why paying with crypto can trigger tax.
    • Swapping one crypto asset for another is usually a taxable disposal even though no conventional money moved.
    • Crypto received as income is generally taxed at its value on receipt, which can exceed what it is later worth.
    • The cost-basis method your jurisdiction prescribes materially changes the taxable gain on identical trades.
    • Losses often offset gains but must usually be claimed properly; worthless or unreachable assets need a specific procedure.
    • Record every transaction as it happens. Reconstruction after the fact is the biggest avoidable cost in crypto tax.
  • How crypto rules actually get made, and why the result looks inconsistent

    How crypto rules actually get made, and why the result looks inconsistent

    The short version

    There is rarely a single “crypto regulator”. Instead, several existing bodies apply existing law to a new object, each through the lens of its own mandate — securities, commodities, payments, tax, sanctions, consumer protection. The same token can be a security to one, a commodity to another and property to a third without anybody being wrong, because they are answering different questions. The inconsistency is structural, and knowing the structure lets you read the news properly.

    Crypto regulatory coverage tends to read as a series of shocks: a decision here, a contradictory one there, a consultation somewhere else. It gives the impression of institutions flailing. The reality is more mundane and much more predictable once you see the machinery.

    This article deliberately names no specific rule, case or jurisdiction. Specific rules change, and an explainer that lists them is a liability within months. The structure changes far more slowly, and it is what actually lets you interpret a headline.

    The default is that existing law applies

    The most common misconception is that crypto is unregulated until a crypto law is passed. Almost nowhere is that true. Laws are generally written in terms of activities and economic substance rather than technologies — issuing an investment, taking custody of client assets, operating a trading venue, transmitting money, realising a gain.

    So the first question a regulator asks is not “what rules should exist for this” but “which existing category does this activity fall into”. This is why enforcement often arrives before legislation. The rules were already there; the question was whether a given arrangement fell inside them.

    Six mandates, one object

    Different bodies have different statutory questions, and none of them is trying to produce an overall verdict on crypto:

    Mandate The question it asks What it can reach
    Securities Is this an investment contract sold to the public? Issuance, promotion, intermediaries
    Commodities / derivatives Is this a traded commodity or a derivative on one? Futures, leverage, market conduct
    Payments / money transmission Is a business moving value for others? Exchanges, custodians, on-ramps
    Tax Has a taxable event occurred? Every holder, directly
    Sanctions / financial crime Who is transacting, and may they? Any regulated intermediary
    Consumer / conduct Was this fairly sold and honestly described? Marketing, disclosures, complaints

    Read that table and the apparent contradictions dissolve. A token can genuinely be a security when sold by a promoter making promises, a commodity when traded between two parties with no promises attached, and property for tax purposes throughout. These are not competing verdicts on its nature. They are answers to three different questions.

    Why the same asset gets classified differently over time

    Several mandates care about the circumstances of a sale rather than the object sold. Whether something is an investment offering can depend on what buyers were told, what they were relying on, and how central a promoter was to the expected return.

    The consequence is genuinely counterintuitive: the same token can be an investment offering at launch, when a team is raising funds against a roadmap, and not one years later, when it trades between strangers and no promoter is promising anything. Coverage describing this as a regulator “changing its mind” has usually misread the test as being about the asset when it is about the transaction.

    Four instruments, easily confused

    Headlines flatten very different actions into “regulators say”. They are not equivalent:

    • Legislation — a new statute. Slow, durable, and the only instrument that genuinely creates new categories.
    • Rulemaking — an agency writing detailed rules under powers it already has. Usually preceded by public consultation.
    • Guidance — an agency stating how it reads existing law. Fast, influential, and generally not binding in the way a rule is.
    • Enforcement — action against a specific party. Decides one case, and signals a reading of the law that others must weigh.

    The most common reporting error is treating enforcement as though it settled a general question. An enforcement action decides a dispute between named parties on particular facts. It tells you what one agency believes; it does not tell you what the law now is for everyone.

    Why jurisdictions diverge, permanently

    Different legal traditions produce different results from the same facts. Some systems work from broad principles applied case by case, others from detailed prescriptive rules written in advance. Some concentrate authority in one financial regulator, others distribute it across several. Some have constitutional constraints on how much an agency may decide without the legislature.

    Add differing priorities — attracting business, protecting retail investors, preserving monetary control, enforcing sanctions — and permanent divergence is the expected outcome. Convergence is the unusual event and normally requires an international standard-setting body to broker it, which takes years.

    How to read a regulatory headline

    Five questions, and most coverage answers none of them:

    Which body, and what is its mandate? That tells you the scope of what was actually decided. Which instrument? Enforcement, guidance, a rule and a statute have wildly different reach. Who is bound? One firm, a category of firms, or everyone. Is it final? Consultations, proposals and appealed decisions are routinely reported as settled. What is the effective date? Many rules are announced years before they bite.

    Applied to a typical headline, these usually reveal something narrower than the framing implied. That is not a reason for cynicism about the reporting — it is a reason to read the primary document, which is nearly always public.

    This is a structural explainer, not legal advice. Rules differ by jurisdiction and change; nothing here describes your obligations. Consult a qualified professional in your own jurisdiction.

    Key takeaways

    • Crypto is not unregulated pending a crypto law — existing law is written around activities, and applies by default.
    • Several bodies ask different statutory questions, so one token can be a security, a commodity and property at once.
    • Some tests examine the circumstances of a sale, so the same token’s status can legitimately change over time.
    • Legislation, rulemaking, guidance and enforcement have very different reach and are routinely conflated.
    • An enforcement action decides one case on particular facts. It is not a general rule.
    • Permanent divergence between jurisdictions is the expected outcome, not a temporary failure to coordinate.
    • Ask which body, which instrument, who is bound, whether it is final, and when it takes effect.
  • What on-chain data cannot tell you

    What on-chain data cannot tell you

    The short version

    A blockchain is a complete record of transfers between addresses. It is not a record of who owns what, why they moved it, or whether a transfer was economically meaningful. Addresses are not people, exchange wallets pool millions of users, an entity can hold thousands of addresses, and a large share of activity is internal shuffling. On-chain data is excellent evidence about movement and weak evidence about everything else.

    On-chain analysis has a strong claim on your attention: unlike almost any other market, the ledger is public. Every transfer, every balance, every timestamp, verifiable by anyone. Compared with equity markets, where ownership data arrives quarterly and partially, this looks like an enormous advantage.

    It is a real advantage. But the completeness of the record creates a specific trap. Because you can see everything that happened on the chain, it is easy to believe you can see everything that happened — and the most common on-chain claims quietly depend on inferences the data does not support.

    What the chain genuinely records

    Be clear about the solid ground first. For a public chain you can establish, with certainty and without trusting anyone:

    • That a transfer of a stated amount occurred between two addresses, in a specific block, at a specific time.
    • The balance of any address at any point in history.
    • The complete transfer graph — which addresses have ever interacted.
    • The fee paid, and the state of the contract code that executed.
    • Total supply, issuance, and anything else the protocol computes.

    That is a lot, and it is genuinely more than other markets offer. Everything below is about what does not follow from it.

    1. An address is not a person

    The relationship between addresses and people is many-to-many in both directions, and both directions break naive analysis.

    One entity, many addresses. Creating an address is free. A single participant may use thousands — routinely, for privacy, operational hygiene, or to appear as many participants. So “the number of holders grew” may describe one person opening accounts. Metrics based on counting addresses are, at best, an upper bound on participants.

    One address, many people. The inverse is more consequential. Exchange wallets hold customer assets in aggregate: a single address may represent millions of individual balances. When an analyst reports that “a whale holding 2% of supply moved funds”, the address is frequently an exchange doing internal treasury management on behalf of users who did nothing at all.

    2. A transfer is not a trade

    This is the error that produces the most confidently wrong headlines. Movement between addresses has many causes, and most of them are not economic decisions:

    What you see What it might actually be
    Large outflow from an exchange Cold-storage rotation, or a custody migration
    Large inflow to an exchange Collateral posting, or market-making inventory
    Dormant coins moving A wallet upgrade, or a key rotation
    Two addresses transacting repeatedly One entity’s internal accounting
    Enormous single transfer Consolidation of many inputs the owner already held

    None of these involve a buyer or a seller. They change which address holds the coins and nothing else. Yet each is regularly reported as a directional signal, because the alternative reading requires admitting the data is ambiguous.

    3. Exchange flow metrics rest on a guessed mapping

    “Exchange netflow” is among the most cited on-chain metrics, and it depends on a step that is never on-chain: deciding which addresses belong to which exchange. That mapping is built by heuristics and inference. It is often good. It is never authoritative, it is not published in full, and it goes stale whenever a venue changes its wallet structure.

    So an exchange-flow chart is a model output, not a measurement — and two providers can produce genuinely different numbers for the same day without either being dishonest. When a metric depends on a proprietary address labelling, the honest presentation says so.

    4. The chain has no idea what anything is worth

    Any on-chain metric denominated in dollars — realised value, profit and loss by cohort, capitalisation of coins last moved in some window — is a blend of chain data with off-chain price data. The chain records that 10 units moved. Attaching a dollar figure requires choosing a price source and a timestamp, and different reasonable choices produce different answers.

    Worse, “unrealised profit” style metrics assume the price when a coin last moved was the price its owner paid. For a coin that moved between two wallets of the same owner, that assumption is simply false, and there is no way to tell the two cases apart from the chain.

    5. Intent is never recorded

    The deepest limit. Two identical transfers can mean opposite things — a sale into strength or an accumulation moved to storage — and they are indistinguishable on-chain. When you read that a movement shows accumulation or distribution, that word is doing work the data cannot support. Direction of transfer is observable; motive is inferred.

    6. Multi-chain reality breaks single-chain metrics

    Assets exist on many chains at once, bridged and wrapped. Activity that leaves a base layer for a layer 2 has not left the ecosystem, but it does leave the base layer’s metrics. A chart showing base-layer activity declining may be measuring success at moving traffic elsewhere. Any metric computed from one chain in a multi-chain world is measuring a shrinking fraction of the whole.

    How to read on-chain claims well

    Four questions clear up most of it. Does this metric require an address-to-entity mapping, and is the mapping disclosed? Does it require a price, and from where? Does it assume an address is one person? Does the interpretation smuggle in intent — words like accumulation, capitulation, conviction?

    If a claim survives all four, it is probably about movement, magnitude and timing — which is where on-chain data is genuinely strong, and genuinely better than what other markets offer. If it does not survive them, you are reading a model with a story attached, and the story is not in the data.

    This is also why we do not publish on-chain “signals”. We are happy to describe what the ledger shows. We are not going to tell you what someone was thinking when they moved coins, because the chain does not say, and pretending otherwise would be exactly the kind of confident emptiness our editorial guidelines exist to prevent.

    Key takeaways

    • The chain proves transfers, balances and timestamps with certainty. Everything else is inference.
    • An address is not a person: one entity can hold thousands, and one exchange address can represent millions of users.
    • Most large transfers are custody rotation, consolidation or internal accounting — not buying or selling.
    • Exchange flow metrics depend on an undisclosed address-to-entity mapping, so they are model outputs, not measurements.
    • Every dollar-denominated on-chain metric blends chain data with a chosen price source and timestamp.
    • Intent is never recorded. Accumulation and distribution are interpretations, not observations.
    • Single-chain metrics measure a shrinking share of a multi-chain ecosystem, so declining base-layer activity can mean success.
  • Support and resistance are not lines on a chart

    Support and resistance are not lines on a chart

    The short version

    Support and resistance are not features of price; they are places where orders are resting. A level “holds” because there is enough size there to absorb what arrives, and it breaks when there is not. This reframing explains three things a line on a chart cannot: why obvious levels are the least reliable, why a break often accelerates rather than stalls, and why a level is a zone rather than a number.

    Draw a horizontal line under a few lows and you have identified support. That is how it is usually taught, and it makes the concept sound like a property of the asset — as though the price knows the line is there.

    It does not. Price is the output of matching orders. If a decline stops at a particular level, the reason is that enough buy orders were sitting at or near that level to absorb the sell orders arriving. That is the entire mechanism, and everything useful about support and resistance follows from taking it literally.

    A level is a place where size is resting

    The order book is a live list of resting orders at each price. Most of the time it is thin: a modest amount available at each level near the current price. Occasionally there is a shelf — a level with substantially more resting size than its neighbours.

    Sell into a thin level and the price moves through it. Sell into a shelf and the price stalls while that size is consumed. From a chart, those two outcomes look like “the level failed” and “the level held”, which invites the conclusion that the level had a property. It did not. It had inventory.

    Why does inventory cluster at particular prices? Several unremarkable reasons, all of them about human behaviour rather than market structure:

    • Prior transactions. People who bought at a level and are now underwater often place exit orders back at their entry, to get out flat. That creates real selling pressure at a specific price.
    • Round numbers. Orders cluster at round figures because people choose round figures. This is a fact about base ten, not about the asset.
    • Visible prior extremes. A previous high or low is something everyone can see, so orders accumulate there — including stop orders placed just beyond it.
    • Mechanical levels. Liquidation prices are determined by leverage and entry, so at certain prices a predictable quantity of forced selling exists.

    Why the obvious levels are the weak ones

    This is the counterintuitive consequence, and the most useful thing in this article.

    If a level is visible to everyone, then everyone’s stop orders are just beyond it. A stop order is not resting inventory that absorbs pressure — it is a market order waiting to be triggered. So the most widely watched level has the thinnest genuine support and the densest cluster of fuel immediately below it.

    That is why price so often trades slightly through a famous level and then reverses hard. The move through the level triggers the stops, the stops sell, that selling is absorbed by whoever wanted the asset lower, and once the fuel is spent the price recovers. From a chart it looks like a “false break” or a “stop hunt”, as if someone engineered it. Usually nobody did. The structure produced it, because the stops were where the chart said to put them.

    The practical version: a level’s reliability is inversely related to how many people are watching it.

    Why breaks accelerate

    When a shelf is genuinely consumed rather than briefly pierced, the price does not drift onward — it usually jumps, because the next resting size may be some distance away. The book below a large level is often thin precisely because that level was doing the work.

    Add liquidations and the effect compounds. Forced sellers do not choose their price, so a cascade sells into whatever is left, and each level reached can trigger the next tranche. This is the mechanism behind moves that look wildly disproportionate to any news, and we go into it further in what actually moves the bitcoin price.

    A level is a zone, and treating it as a number is an error

    Because clustering is approximate, the “level” is really a band. Orders sit near a memorable price, not exactly on it. Two consequences:

    First, precision is false comfort. A support level quoted to five significant figures implies knowledge nobody has. The zone is the honest object.

    Second, “the level broke” needs a definition before it means anything. A one-second wick through a zone and an hour of trading below it are different events, and if you have not decided in advance which one you mean, you will decide afterwards — in whichever direction is more comfortable.

    What this reframing buys you

    Chart language What is actually happening
    “Support held” Resting bids absorbed the arriving sell flow
    “Support broke” Arriving sell flow exceeded resting bids
    “False break” Stops beyond the level were triggered and then absorbed
    “Resistance became support” Sellers at that price are finished; new buyers now rest there
    “The level is strong” An untestable claim about inventory you cannot see

    The last row is the point. On most venues you can see some depth, but you cannot see hidden orders, orders that will appear only when price arrives, or size sitting on other venues. Anyone describing a level as “strong” is inferring inventory from a chart, which is a guess about the thing that actually matters.

    The limits, stated plainly

    None of this makes support and resistance predictive. It makes them descriptive — a vocabulary for where transactions concentrated, which is genuinely useful for understanding what happened and for thinking about where an idea would be wrong.

    What it will not do is tell you which way price goes next. A level is a place where two opposing views have historically met in size. That is information about the past distribution of orders. It is not information about the future, and the confident version of this analysis that you will read elsewhere is confident about something it cannot know.

    If you use levels at all, the honest use is defensive: as candidates for where a thesis is invalidated, which is exactly the input a position size calculation needs. That is a much smaller claim than most technical writing makes, and it is the one the mechanism actually supports. Nothing here is a recommendation — see our disclaimer.

    Key takeaways

    • Support and resistance describe resting order inventory, not a property of the price.
    • Levels hold when resting size absorbs arriving flow, and break when it does not. That is the whole mechanism.
    • The most widely watched level is the least reliable, because everyone’s stops sit just beyond it.
    • “Stop hunts” usually require no manipulation — the structure produces them because the chart told everyone where to put stops.
    • Breaks accelerate because the book beyond a large level is often thin, and liquidations compound the effect.
    • A level is a zone. Quoting one to five significant figures implies knowledge nobody has.
    • Define “broken” before the event, or you will define it afterwards in whichever direction is more comfortable.
  • How to read a trading session without a narrative

    How to read a trading session without a narrative

    The short version

    Read the session in this order: breadth before the headline move, volume before the percentage, dispersion before the leaders, and positioning before any explanation. Most published commentary reverses this — it starts from the biggest number and builds a story back toward it, which produces confident sentences and very little information.

    Every day, something is the biggest mover and something is the biggest loser, and every day there is an article explaining why. The explanations are produced at high volume under deadline, and the honest ones would mostly read “positioning, probably”. Since that does not fill a page, they read as something more definite instead.

    This is not a complaint about journalists. It is a structural feature of daily coverage: the move is known and the reason is not, so the reason gets constructed. The useful response is to have your own reading order — one that starts with the questions that have answers.

    1. Breadth, before you look at the headline number

    A single asset’s move tells you about that asset. What you usually want to know is whether something happened to the market, and for that the question is how many things moved, not how far one did.

    Count advancers and decliners across the top fifty by market capitalisation. Three situations look completely different once you do:

    • Broad and aligned — most assets moving the same way. This is a market event: a shift in the cost of capital, a change in overall risk appetite, or a liquidation cascade.
    • Narrow and large — one or two assets moving sharply while the rest are flat. This is an asset-specific event, and the market-wide explanation you are about to read is wrong.
    • Mixed with a large index move — the aggregate is being carried by the largest constituents. Worth knowing before you conclude anything about “crypto”.

    Our market table shows the constituents so you can count for yourself, which takes about thirty seconds and is the highest-value half-minute in this list.

    2. Volume, before you take the percentage seriously

    A percentage change with no volume behind it is a fact about a thin book, not about demand. Small assets can print dramatic percentages on trivial flow, which is why “biggest gainers” lists are so consistently populated by things nobody is trading.

    The check: is today’s volume unusual relative to that asset’s own recent average? Not relative to bitcoin’s volume — relative to its own. A 15% move on twice-normal volume and a 15% move on a fifth of normal volume are different events, and only one of them tells you anything about conviction.

    This is also where liquidity and slippage stop being abstractions. If a move happened on thin volume, the price you see is not a price you could have transacted at in any size.

    3. Dispersion, which tells you what kind of day it is

    Look at how tightly the top fifty are clustered. When almost everything is within a percentage point or two of everything else, individual asset stories are irrelevant — one factor is driving the whole set, and picking through project news is wasted effort.

    When dispersion is wide, the opposite holds: something asset-specific is happening, and it is worth identifying what. The mistake is applying the wrong mode. Most bad market commentary consists of asset-specific explanations offered on days when dispersion was near zero.

    4. Positioning, before you accept any explanation

    Now, and only now, consider why. And start from the least glamorous possibility: nothing informational happened, and the market moved because of where positions were.

    The signature of a positioning move is recognisable. It is fast — most of it inside a few minutes. It is disproportionate to any news you can find. It frequently reverses a meaningful part of itself within hours. And it tends to occur at levels where you would expect stops or margin calls to cluster.

    The signature of an informational move is different: it starts at an identifiable moment, it holds, and it is accompanied by sustained rather than spiked volume. If a move has the first signature and the article gives you the second kind of explanation, the article is wrong regardless of how plausible it sounds.

    Two things that are not signals

    Sentiment readings. A fear and greed reading describes mood, and mood is largely derived from price. Using it to predict price is close to circular. It is a useful summary of where the crowd currently is; it is not an entry signal, and anyone treating it as one is trading a lagging restatement of the chart.

    Round numbers. An asset “testing” a round number is a fact about the base-ten number system and about where people place orders, not a fact about the asset. Sometimes the order clustering makes it briefly self-fulfilling. That is not a reason to build a view around it.

    What a session cannot tell you

    One day is a very small sample of a very noisy process. Bitcoin daily moves of several percent are ordinary, which means most single sessions contain no signal at all — they are the noise floor of the asset class, and reading meaning into them is the most common analytical error in crypto.

    The practical consequence: if your conclusion from a session would change your behaviour, be suspicious of it. A reading process is for staying oriented, not for generating decisions. That distinction is also why we stamp every price-relevant article with the price it was written at — so you can see afterwards what conditions a piece was produced in, without anyone having to claim it was a call. Our methodology explains how that works.

    Key takeaways

    • Read breadth first: how many assets moved matters more than how far one of them did.
    • A narrow, large move means the market-wide explanation you are about to read is wrong.
    • Judge volume against the asset’s own recent average, never against bitcoin’s.
    • When dispersion is near zero, asset-specific explanations are noise — one factor is driving everything.
    • Positioning is the default explanation for a fast move that reverses. Informational moves start, hold and sustain volume.
    • Sentiment readings are largely derived from price, so using them to predict price is close to circular.
    • Most single sessions contain no signal. If a day’s reading would change your behaviour, be suspicious of it.
  • Stablecoins: four designs, four different promises

    Stablecoins: four designs, four different promises

    The short version

    “One dollar” is a claim, and the four stablecoin designs back that claim in incompatible ways: dollars at a bank, over-collateralised crypto, an algorithm with no collateral, or a commodity. Each has a characteristic failure. Fiat-backed depends on a custodian and its bank. Crypto-backed depends on liquidations clearing during a crash. Algorithmic depends on continued confidence, which is the thing that disappears first. Knowing your design tells you your risk.

    A stablecoin’s job is to be boring. It is the unit people use to hold value between trades, price things, and move money without touching the banking system twice. Because it is boring, it gets treated as a single homogeneous thing — “a dollar on-chain” — and the differences between designs get flattened away.

    They should not be. The word “stable” describes the intent, not the mechanism, and the mechanism is where the risk lives. There are four broad approaches.

    1. Fiat-collateralised: dollars somewhere else

    The most common design. An issuer takes dollars, holds them in reserve, and issues one token per dollar received. Redemption is the anchor: if a token can reliably be exchanged for a dollar, arbitrage keeps the market price near a dollar without any clever machinery.

    What it depends on. Everything rests on the reserve genuinely existing, being liquid enough to meet redemptions in a rush, and being reachable. Two questions matter more than any others: what are the reserves actually invested in, and who is allowed to redeem at par? If redemption is only available to large institutional clients, retail holders depend on those institutions arbitraging on their behalf — which works until it does not.

    How it fails. Not usually through fraud. The realistic failure is the reserve being held somewhere that becomes unavailable — a bank that fails, a jurisdiction that freezes, an asset that cannot be sold at par under stress. The token then trades below a dollar not because the dollars are gone but because nobody can get at them today. This design converts crypto risk into banking and custody risk. It does not remove risk.

    2. Crypto-collateralised: over-collateralised and liquidated

    Here the backing is other crypto assets, held in a smart contract rather than at a bank. Because the collateral is volatile, the system demands more than one dollar of it per dollar issued — often substantially more. If the collateral falls toward the debt it supports, the position is liquidated and the debt repaid.

    What it depends on. Three things, all of which are stressed at the same moment: an accurate price feed, enough liquidity for liquidations to actually clear, and participants willing to buy the collateral being sold. It is transparent in a way the fiat-backed design is not — you can verify the collateral on-chain, right now, yourself.

    How it fails. Through correlation. In a sharp decline the collateral falls, liquidations trigger, the liquidations sell collateral into a falling market, and that pushes collateral lower still. If prices fall faster than positions can be closed, the system ends up under-collateralised, holding bad debt. Every mechanism designed to protect it — the price feed, the liquidation queue, the block space needed to execute — is under maximum load exactly when it is needed. See liquidation for the mechanics.

    3. Algorithmic: confidence as collateral

    The most ambitious design and the one with the worst record. Here there is little or no collateral. Instead the protocol manages supply against a companion asset: when the stablecoin trades above target, supply expands; when it trades below, holders are given an incentive to swap into the companion asset, contracting supply.

    What it depends on. That the companion asset has value. And the companion asset’s value derives largely from the expectation that the system will keep working — which makes the argument circular.

    How it fails. Quickly, and in public. If the stablecoin trades below target and the swap mechanism requires issuing more of the companion asset, then a falling companion asset means issuing more of it to redeem the same value, which pushes it down further. Confidence and collateral are the same variable, so they fail together and there is no floor. This is not a hypothetical — the collapse of a major algorithmic stablecoin in May 2022 followed exactly this path and erased tens of billions of dollars of nominal value in days.

    Some designs are partially collateralised, sitting between this and the previous category. They are better, and the honest way to assess them is to ask what happens at the collateral portion alone, because that is the part that does not depend on belief.

    4. Commodity-collateralised: a claim on a physical thing

    The token represents a quantity of a physical commodity, most often gold, held in a vault. It is not a dollar stablecoin at all — it is stable against the commodity, and therefore moves against the dollar.

    What it depends on. Whether the metal exists, is audited by someone independent, is unencumbered, and can be redeemed at a size and cost that makes redemption real rather than theoretical. Storage costs money, so there is usually a fee that quietly erodes the holding.

    How it fails. Through the custody chain, and through redemption terms that turn out to be impractical for anyone holding a normal amount.

    The one comparison worth keeping

    Design Backed by Verify by Characteristic failure
    Fiat-collateralised Cash and short-term instruments Attestations, reserve reports Custodian or bank becomes unreachable
    Crypto-collateralised Over-collateralised crypto On-chain, directly Correlated crash outruns liquidations
    Algorithmic A companion asset and confidence Nothing external Reflexive collapse with no floor
    Commodity-collateralised Physical metal in a vault Independent vault audit Custody chain, unusable redemption

    What to actually do with this

    Find out which design you are holding — it is usually stated plainly in the issuer’s own documentation, and if it is not, that is informative. Then ask the one question that fits that design: for fiat-backed, where are the reserves and who can redeem; for crypto-backed, what is the collateral and how has the liquidation system behaved under stress; for algorithmic, what is the non-reflexive floor; for commodity, who audits the vault and what does redemption really cost.

    And keep the general point in view. A stablecoin is not cash. It is a claim on someone or something, and “stable” is a description of how it usually behaves rather than a guarantee about how it must. None of this is advice about which to hold — see our disclaimer.

    Key takeaways

    • “Stable” describes intent, not mechanism. Four designs back the same claim in incompatible ways.
    • Fiat-collateralised converts crypto risk into banking and custody risk. It does not remove risk.
    • Ask who is permitted to redeem at par — if only institutions can, retail depends on their arbitrage.
    • Crypto-collateralised is verifiable on-chain but fails through correlation, when liquidations cannot clear fast enough.
    • Algorithmic designs use confidence as collateral, so the two fail together and there is no floor beneath them.
    • A commodity-backed token is stable against the commodity, not the dollar, and storage fees erode it over time.
    • A stablecoin is a claim on someone, not cash. Identify your design, then ask the one question that fits it.
  • How to judge an altcoin before you look at its chart

    How to judge an altcoin before you look at its chart

    The short version

    Look at supply before price, because a low unit price means nothing without the unit count. Then find out who holds it, what unlocks and when, whether the thing has users distinguishable from incentives, where it trades and how deeply, and who can change the rules. Seven checks, no chart, mostly answerable in twenty minutes — and if you cannot answer them, that is itself the answer.

    A chart is a record of decisions other people have already made. It is genuinely useful information, and there is a case for reading it — we have written about how to do that without fooling yourself. But it cannot tell you what you would actually be buying, and if you look at it first it will colour everything that follows. Momentum is persuasive in a way that fundamentals are not.

    So here is the order we would use instead. None of it requires a subscription, and none of it is a valuation model — there is no reliable one for this asset class. It is a list of ways to find the thing that would embarrass you later.

    1. Supply, before you look at the price at all

    The single most common error in crypto is treating a low unit price as cheap. It is not a fact about value; it is a fact about how many units exist. What matters is market capitalisation — price multiplied by circulating supply.

    Number to find What it tells you
    Circulating supply Units actually in the market now
    Total supply Units that exist, including locked ones
    Maximum supply Whether issuance ever stops
    Circulating as % of maximum How much future dilution is already scheduled

    That last row is the one people skip. If a fifth of the eventual supply is circulating, then four times the current float is arriving at some point, and every existing holder is diluted by it. The market may have priced this in. It may not have. But you cannot form a view without the number, and it takes about a minute to find. Our coin comparison tool puts these side by side, and each asset’s own page carries the full set.

    2. The unlock schedule, which is when dilution actually lands

    Knowing that supply will grow is not enough; the timing is what moves markets. Most projects publish a vesting schedule covering team, investor and treasury allocations. Read it and note the dates where a large tranche becomes transferable.

    Two questions matter more than the total. First, are early investors sitting on a very large paper gain? A holder up several multiples behaves differently from one at cost. Second, is the tranche large relative to daily traded volume, not relative to market capitalisation? A release worth a few percent of market cap can be many days of volume, and that is what determines whether it can be absorbed.

    If there is no published schedule, treat that as a finding rather than an absence of one.

    3. Concentration: who actually holds this

    Public blockchains let you inspect holdings, and a block explorer will show you the largest addresses for most assets. You are not looking for a specific threshold. You are looking for whether a handful of addresses could exit through the available liquidity without destroying the price.

    Read the result carefully — this is where naive on-chain analysis goes wrong. A very large address is often an exchange holding customer assets in aggregate, which is not concentration in any meaningful sense. Conversely, one entity can hold many addresses and look like many holders. On-chain data has real limits here, which we cover in what on-chain data cannot tell you.

    4. Usage that is distinguishable from incentives

    This is the hardest check and the most valuable. Plenty of networks show impressive activity that consists almost entirely of people farming a reward. That is not demand for the product; it is demand for the subsidy, and it stops the day the subsidy does.

    Useful questions: is there activity from addresses that receive no rewards? Does usage survive a reduction in emissions? Is there revenue paid by users who want the service, as opposed to paid to users to attract them? Does the fee the network earns come from anywhere other than the token it issues?

    You will often not be able to answer these definitively. Getting a clear negative — usage collapses without incentives — is still an excellent outcome for twenty minutes of work.

    5. Liquidity and where it lives

    Liquidity is your ability to exit at a price near the one you see quoted, and it is not implied by market capitalisation. An asset can carry a large valuation while its order book is thin enough that a modest sale moves it several percent.

    Check how many venues list it, how concentrated volume is on the largest one, and how much of the “volume” is on a single venue with unverifiable reporting. Then look at the depth near the current price, not the headline 24-hour figure. If most trading happens on one venue, your exit risk includes that venue’s operational risk, whatever you think of the asset.

    6. Who can change the rules

    Ask what can be altered, by whom, and how fast. Can an upgrade change the token’s behaviour without holder approval? Is there an address that can pause transfers, mint new units, or freeze balances? Is the code deployed the code that was audited? A privileged key is not automatically damning — many legitimate projects retain one to fix bugs — but it is a risk you should be holding knowingly rather than discovering later.

    7. Whether the claims are specific

    Read whatever the project says about itself and sort each claim into “checkable” or “not”. “Partnered with a major bank” is unfalsifiable as written. “Processes settlement for this named counterparty, live since this date” is checkable. Vagueness that persists across a whole document is a pattern, and the pattern is the information.

    Then, and only then, look at the chart

    Having done this, the chart becomes useful rather than persuasive. You know what the supply overhang is, so you can see whether a decline coincided with an unlock. You know where liquidity is, so you can tell a real move from a thin one. You have your own view, and the chart is now evidence to test it against rather than the thing forming it.

    Two honest caveats. This process is designed to help you avoid obvious traps, not to identify winners — the checks are much better at producing a confident “no” than a confident “yes”. And nothing here is a recommendation to buy anything. We do not make those. Read our disclaimer for the full position.

    Key takeaways

    • A low unit price is a fact about unit count, not about value. Start with market capitalisation and supply.
    • Circulating supply as a percentage of maximum supply tells you how much dilution is already scheduled.
    • Judge an unlock against daily traded volume, not market capitalisation — volume is what has to absorb it.
    • Large addresses are often exchange omnibus wallets, and one entity can hold many addresses. Read concentration carefully.
    • Separate usage from incentives: activity that disappears when rewards stop was never demand for the product.
    • Liquidity is not implied by valuation. Check venue concentration and depth near the price, not the 24-hour headline.
    • These checks reliably produce a confident “no” and rarely a confident “yes”. That is the correct use of them.
  • Ethereum’s fee market: base fee, priority fee, and why gas spikes

    Ethereum’s fee market: base fee, priority fee, and why gas spikes

    The short version

    An Ethereum fee has two parts. The base fee is set by the protocol, adjusts automatically toward half-full blocks, moves at most 12.5% per block, and is destroyed rather than paid to anyone. The priority fee is what you add on top to be included sooner, and it goes to the block proposer. Fee spikes are the base fee compounding: 12.5% per block for ten consecutive blocks is roughly 3.2x in about two minutes.

    Most explanations of Ethereum fees stop at “it depends on network demand”, which is true and useless. The mechanism is actually quite precise, and once you can see it, gas stops feeling arbitrary. There are two numbers, one algorithm, and one hard constraint on how fast that algorithm is allowed to move.

    Gas is a unit of work, not a currency

    Gas measures computational work. Every operation the network can perform has a fixed gas cost, set by the protocol and identical for everyone. A simple transfer of ether costs 21,000 gas whether the network is quiet or overwhelmed. Interacting with a smart contract costs more, depending on what the contract does.

    So the amount of gas you need is a property of the transaction. What varies is the price you pay per unit — and that is where the two-part structure comes in.

    The base fee: an algorithm, not an auction

    Since the fee-market change introduced in 2021, each block carries a base fee per unit of gas that is set by the protocol rather than bid for. Every user in a given block pays the same base fee. It is calculated from the previous block, using one rule:

    • Blocks have a target size and a maximum of twice that target.
    • If the previous block was exactly at target, the base fee stays the same.
    • If it was above target, the base fee rises — by at most 12.5%.
    • If it was below target, the base fee falls — by at most 12.5%.

    That is the whole mechanism. It is a feedback loop steering block occupancy toward the target, and the 12.5% cap is the reason fees behave the way they do.

    The base fee is then burned — removed from supply entirely rather than paid to a validator. This matters for two reasons. It means no participant profits from congestion, so nobody has an incentive to manufacture it. And it means network usage permanently reduces the ether supply, which is why fee revenue is discussed alongside issuance when people argue about Ethereum’s supply dynamics.

    Why “gas spiked instantly” is never quite true

    The 12.5% cap means the base fee cannot jump. It can only compound. Ethereum produces a block every 12 seconds under normal conditions, so a sustained run of completely full blocks gives you:

    Consecutive full blocks Elapsed time Base fee multiple
    1 12 seconds 1.125x
    5 1 minute 1.80x
    10 2 minutes 3.25x
    20 4 minutes 10.5x
    40 8 minutes 111x

    Every figure above is just 1.125 raised to the number of blocks. Nothing else is needed to explain a tenfold fee increase over a few minutes, and nothing about it requires a conspiracy or a bug.

    The same arithmetic runs in reverse. Once demand falls away and blocks come in under target, the base fee decays at up to 12.5% per block, which halves it in about six blocks — a little over a minute. This is the single most practical consequence of the design: fee spikes are short-lived unless demand is genuinely sustained. If a transaction is not urgent, waiting ten minutes is frequently the entire optimisation.

    The priority fee: the part you actually choose

    The base fee gets you into the queue at the protocol’s price. The priority fee is a tip on top, paid to whoever proposes the block, and it determines your ordering relative to other transactions competing for the same space.

    When blocks are below target there is spare room, so a minimal tip is usually enough — you are not competing with anyone. When blocks are full, the tip becomes a real auction, and this is the part of your fee that can genuinely be bid up without limit. A wallet showing you “slow / average / fast” options is almost always varying the priority fee, not the base fee, because the base fee is not yours to vary.

    You also set a fee cap: the maximum total you are willing to pay per unit of gas. If the base fee rises above your cap while your transaction is pending, it simply waits rather than executing at a price you did not agree to. Anything you were willing to pay above the actual base fee plus your tip is refunded.

    What makes fees high in the first place

    Congestion is competition for a fixed amount of block space. In practice it comes from a few recognisable sources: a heavily oversubscribed token distribution where thousands of participants race for the same allocation; a sharp market move that triggers many simultaneous liquidations and rebalances; a newly popular application whose transactions are individually gas-heavy; and periods where automated arbitrage between venues intensifies.

    Notice what these have in common — they are all bursts of simultaneous, time-sensitive demand. Ordinary steady usage does not spike fees, because the base fee algorithm has time to find its level.

    The structural answer: move the work elsewhere

    Ethereum’s long-run approach to fees is not to make the base layer cheaper by making blocks bigger. It is to settle transactions on a layer 2 network and post compressed data back to the base layer, so many transactions share the cost of one settlement.

    A later protocol change added a dedicated, separately priced data channel for exactly this traffic, with its own independent fee market. The practical effect is that layer-2 costs are no longer tightly coupled to base-layer congestion — which is why the fee you pay on a rollup and the fee you pay on Ethereum itself can now diverge sharply during the same busy hour.

    Checking the current number

    Because the base fee is a live protocol value, the only honest way to know it is to read it. Our gas tracker reads the current base fee and priority fee from a node and states when it last did so. Everything in this article explains how that number got there; it deliberately does not quote one, because any figure written into a sentence here would be wrong within the minute.

    Key takeaways

    • Gas is a fixed unit of computational work; a plain ether transfer is always 21,000 gas. Only the price per unit moves.
    • The base fee is set by the protocol, is identical for everyone in a block, and is burned rather than paid to anyone.
    • It can change by at most 12.5% per block, so spikes compound rather than jump — about 3.2x in two minutes, 10.5x in four.
    • The same cap applies to the decay, which halves the base fee in roughly six blocks. Waiting is often the whole fix.
    • The priority fee is the part you choose and the part that can be bid up without limit when blocks are full.
    • Your fee cap protects you: if the base fee exceeds it, the transaction waits instead of executing at a price you did not accept.
    • Layer-2 networks with their own data fee market are the structural answer, which is why rollup and base-layer fees can now diverge.
  • What actually moves the bitcoin price

    What actually moves the bitcoin price

    The short version

    Bitcoin has no cash flow, so its price is set entirely by what someone else will pay — which makes it a question about flows and positioning rather than valuation. The forces that actually move it are net new buying, forced selling, leverage, the cost of holding dollars, supply that is already fully scheduled, and the plumbing that connects the two. Almost everything presented as an explanation is one of these six wearing a story.

    Every asset with a cash flow has an anchor. You can argue about the right multiple for a company’s earnings, but there is an earnings number to argue about. Bitcoin has no cash flow, no dividend and no coupon. Its price is therefore not the output of a valuation model; it is the clearing level between people who want to own it and people who want to hold something else instead.

    That sounds unsatisfying, and it is often used as a criticism. It is really just a statement about which questions are worth asking. If price is a clearing level, the useful question is not “what is it worth” but “what is changing about who wants to hold it, and at what size”. Here are the six things that genuinely change that, in rough order of how often they matter.

    1. Net new buying, which is much rarer than volume suggests

    Exchange volume is not demand. Most of it is the same coins changing hands repeatedly between traders who intend to be flat by the end of the week. That activity sets the price minute to minute and contributes almost nothing to the direction over months.

    What moves the price durably is capital arriving that was not previously in the asset and does not intend to leave soon. It is genuinely hard to observe in real time, which is why so much analysis substitutes something easier to measure and hopes it is a proxy. Treat any claim about “institutional demand” that rests on a volume chart with suspicion — volume tells you how much trading happened, not how much ownership changed hands permanently.

    2. Forced selling, which is the fastest mover of all

    Voluntary sellers are price-sensitive. They have a level in mind, and if the market is below it they wait. Forced sellers have no such option: a liquidation sells at whatever the book will pay, immediately, in whatever size the position was.

    This asymmetry is why declines in crypto are so much sharper than advances. An advance has to persuade holders to part with coins. A decline can simply take them. When you see a move of several percent in minutes with no news attached, the explanation is almost always positioning rather than information — someone’s stop cluster or margin call was where the price was going anyway, and it accelerated through it.

    3. Leverage, which decides how violent any move becomes

    Leverage does not change direction. It changes amplitude. The same amount of net selling produces a small dip in a market with little borrowed money and a cascade in a market saturated with it, because each price level reached triggers the liquidation of positions that then sell into the next level down.

    This is the single most useful thing to know about crypto market structure, and it explains a pattern that otherwise looks irrational: the largest single-day moves tend to arrive after a long quiet stretch, not after a volatile one. Quiet markets are where leverage accumulates, because nothing has happened recently to punish it.

    4. The cost of holding dollars instead

    Bitcoin competes with cash. When holding cash pays a meaningful real return, the opportunity cost of owning a non-yielding asset is high, and the marginal buyer needs more conviction. When cash pays nothing in real terms, that hurdle falls away.

    This is the mechanism behind the observation that bitcoin trades with a sensitivity to interest-rate expectations, and it is why it often moves on macroeconomic releases that have nothing to do with crypto. It is not that bond markets have an opinion about bitcoin. It is that the required return on everything shifts at once, and a long-duration asset with no cash flow is at the far end of that shift.

    5. Supply, which is fully known and therefore rarely the news

    Bitcoin’s issuance schedule is fixed in the protocol. New supply arrives with each block, and the amount per block is cut in half every 210,000 blocks — roughly every four years given a ten-minute target block interval. Total issuance is capped at 21 million.

    Property Value Can it change?
    Maximum supply 21,000,000 Only by consensus rule change
    Halving interval 210,000 blocks Fixed in protocol
    Target block interval 10 minutes Held by difficulty adjustment
    Difficulty retarget Every 2,016 blocks Fixed in protocol

    Because all of this is public and has been for over a decade, it is poor material for a surprise. A halving is not new information on the day it happens; it was knowable years in advance. What can still matter is the interaction between a known supply reduction and an unknown demand path — and the honest version of that argument is much weaker than the confident version you will usually hear. If you want to watch the schedule rather than read about it, our halving countdown tracks it directly.

    6. Plumbing: the difference between wanting to buy and being able to

    The last force is the least discussed and often the most important at turning points. Demand only reaches the price if there is a path for it. Custody arrangements, banking access, the presence or absence of a regulated venue in a given jurisdiction, and the depth of the order book at the relevant size all determine how much intent becomes a trade.

    When plumbing improves, previously excluded capital can act, and the effect can look like a demand shock even though preferences did not change. When plumbing breaks — a payment rail withdrawn, a venue frozen — supply and demand can be unchanged while the price moves sharply, because the two sides can no longer meet.

    What does not belong on this list

    Two things get more credit than they deserve.

    Individual statements. A prominent person saying something favourable moves price when it changes what someone can or will do with capital. When it does not, the effect decays within days. The test is whether a mechanism follows the sentence.

    Hash rate. Mining capacity responds to price far more reliably than price responds to mining capacity. Miners expand when revenue is high, which means hash rate is largely a lagging indicator of the thing it is often used to predict.

    How to use this

    When you next read that bitcoin moved for a reason, sort the reason into one of the six. If it fits, you have learned something about magnitude and probably about persistence. If it fits none of them, you are reading a narrative constructed after the fact to explain a move that was really about positioning.

    And keep the honest caveat in view: knowing which force is acting tells you about the character of a move, not its direction. These are the levers, not a forecast. Nobody at Coinpric is going to tell you where the price goes next, because we do not know, and neither does anyone who says they do.

    Key takeaways

    • Bitcoin has no cash flow, so its price is a clearing level between competing preferences, not the output of a valuation model.
    • Net new buying moves price durably; exchange volume mostly measures the same coins circulating and says little about direction.
    • Forced selling is the fastest mover, because a liquidation is price-insensitive in a way a voluntary sale never is.
    • Leverage sets amplitude rather than direction — which is why the biggest moves often follow quiet stretches.
    • Supply is fully scheduled and public, making it weak material for surprises; the halving is knowable years ahead.
    • Hash rate follows price more reliably than it leads it, so it is a poor predictor of the thing it is used to predict.
    • Identifying the active force tells you about a move’s character, not where price goes next. This is not advice.