Blog

  • How to read a crypto chart without fooling yourself

    How to read a crypto chart without fooling yourself

    The short version

    A chart tells you what price has already done and where people previously reacted. It does not tell you what happens next. Learn candles, volume and timeframes so you can evaluate other people’s claims — and treat any read that does not say what would prove it wrong as a guess with a chart attached.

    There is a good reason to learn chart reading even if you never trade: an enormous amount of crypto commentary is delivered in this vocabulary, and you cannot assess a claim made in a language you do not speak. That is a different goal from learning to predict prices, and it is a far more achievable one.

    What a candle actually encodes

    Each candle covers one period — a minute, an hour, a day — and records four numbers: where price opened, where it closed, and the highest and lowest points reached in between. The body spans open to close; the thin wicks reach out to the extremes.

    The useful information is in the relationship between those four. A small body with long wicks in both directions means price travelled a long way and finished roughly where it started: a lot of activity, no resolution. A large body with almost no wicks means one side controlled the entire period. Same duration, very different meaning.

    What a candle cannot tell you is the order in which things happened inside it. A daily candle that opened low and closed high looks identical whether the rise was steady or whether price collapsed first and recovered in the final hour. Shorter timeframes contain that detail; the daily view has thrown it away.

    Timeframe is the setting that changes the conclusion

    This matters more than any pattern, and it is where most beginners go wrong. A level that is significant on a weekly chart is irrelevant on a five-minute one. A downtrend on the hourly can sit inside an uptrend on the daily, and both descriptions are accurate.

    The practical consequence is that a read is meaningless without a stated horizon. When you see analysis that does not say whether it is talking about days or months, you cannot act on it — and you cannot judge it afterwards either, which is often the point. Our own technical analysis states the horizon up front, and you should notice when other analysis does not.

    Related and less obvious: the same price history can be made to look like a breakout or a failure depending on three choices that are rarely disclosed — which exchange the data came from, whether the vertical scale is linear or logarithmic, and where the visible window begins. A chart presented as evidence is also a set of framing decisions.

    Volume is the part people skip

    Volume is how much traded during a period, and it is the closest thing a chart has to a confidence measure. A large move on heavy volume means many participants were involved. The same move on very little volume means it took hardly anything to produce — and in thin conditions it frequently reverses.

    This connects directly to liquidity. In a liquid market there are plenty of resting orders near the current price, so a large trade barely shifts it. In an illiquid one, a modest order moves the price several percent. Smaller assets often look exciting on the way up precisely because thin depth exaggerates every move — and the same thinness means there may be no bid at all when you want out.

    If you only add one habit from this article, make it checking volume before believing a move.

    Support, resistance, and the honest version of it

    Support and resistance describe price areas where the market has previously reversed. The reasoning is behavioural rather than mystical: people remember what they paid, orders cluster at round numbers and prior extremes, and those clusters get tested again.

    The honest version has two properties. First, a level you cite should be visible in the price history rather than drawn to support a conclusion — if you have to squint, it is not a level. Second, levels are zones, not lines. Price does not respect a specific figure to the cent, and treating it as though it does produces stops that get taken out by noise before the idea has been tested.

    Indicators, and what they cannot contain

    Moving averages, RSI and MACD are all transformations of past price. That is worth sitting with, because it has a firm implication: no indicator contains information the price did not already have. They compress and smooth, which can make a pattern easier to see, and they lag by construction.

    Moving averages

    An average of recent closes, used to describe trend direction. Because it is an average, it turns after price does — always. A moving average crossover is a description of something that already happened.

    RSI

    Compares the size of recent gains to recent losses on a 0–100 scale. The convention that above 70 is “overbought” and below 30 “oversold” is where beginners lose money: in a strong trend RSI can sit at an extreme for weeks while price keeps going. It measures momentum, not exhaustion.

    MACD

    The relationship between two moving averages. It inherits their lag and adds sensitivity to the settings chosen, which is why the same chart yields different signals under different parameters.

    None of this makes indicators useless. It makes them descriptive. Trouble starts when a description is treated as a forecast.

    Key takeaways

    • A candle gives you open, close, high and low — but not the order events happened in.
    • No read means anything without a stated timeframe; the same chart supports opposite conclusions across horizons.
    • Volume is the confidence check. A big move on thin volume frequently reverses.
    • Levels are zones, and a level worth citing is visible without squinting.
    • Every indicator is a transformation of past price and cannot contain new information.
    • Exchange, scale and window are undisclosed framing choices in any chart shown to you.

    Where the order book fits

    A chart shows what has traded. The order book shows what is currently available — bids, asks, and the volume stacked at each level. For the practical question of whether you could exit a position at anything near the quoted price, the book tells you far more than the chart does. A quoted price is simply the last trade; it is not a promise about the next one.

    Practising without risking anything

    Read the charts on our coin pages across the 7-day, 1-month, 3-month and 1-year ranges and notice how the story changes with the window — that single exercise teaches timeframe dependence faster than any explanation. Cross-check the figures on the market table, and look up anything unfamiliar in the glossary.

    When you move from reading charts to acting on them, the sequence matters: read risk management first, and size any position with the position size calculator. Chart skill without sizing discipline is the combination that empties accounts.

    Frequently asked questions

    Does technical analysis work?

    It works as a description of where participants have previously reacted. As a forecasting method its record is far weaker than the confidence with which it is usually presented. Both statements can be true at once.

    Which timeframe should I use?

    The one that matches how long you intend to hold. The error is mixing them — analysing on one and acting on another.

    Which indicator is best?

    None of them is best, because they all derive from the same price data. Adding more indicators does not add information; it adds ways to see what you already believe.

    Notes and further reading

  • Crypto market state: what the data actually shows

    Crypto market state: what the data actually shows

    The short version

    A snapshot of the crypto market as at 4 Aug 2026 20:38 UTC, built entirely from figures read out of our own data pipeline at the moment of writing. Bitcoin trades at $64,116, Ethereum at $1,871, and total market capitalisation stands at $2.17T. This is a description of conditions, not a forecast, and nothing here is a recommendation.

    This piece does one thing: record what the market looked like at a specific moment, with every number attributed and timestamped, so it stays useful when read later. It contains no predictions, because we do not make them.

    Where the majors stood

    At 4 Aug 2026 20:38 UTC, Bitcoin was trading at $64,116 (+0.29% over 24 hours) with a market capitalisation of $1.28T. Ethereum was at $1,871 (-0.06%), capitalised at $228.96B.

    Across the whole market, total capitalisation was $2.17T on $98.01B of 24-hour volume. Bitcoin accounted for 59.0% of that total and Ethereum 10.5% — the figure our market pulse reports as dominance. Dominance is easy to over-read: it rises both when Bitcoin gains and when everything else falls harder, and those are very different situations.

    Top assets at the time of writing

    Asset Price 24h 7d Market cap
    Bitcoin (BTC) $64,116 +0.29% +1.01% $1.28T
    Ethereum (ETH) $1,871 -0.06% -0.72% $228.96B
    Tether (USDT) $0.9984 -0.15% -0.08% $183.92B
    Binance Coin (BNB) $593.02 +0.15% +4.91% $82.54B
    USD Coin (USDC) $1.00 -0.02% -0.05% $74.02B
    XRP (XRP) $1.08 -0.44% +1.89% $64.31B
    Solana (SOL) $74.05 +0.06% +1.03% $40.22B
    TRON (TRX) $0.3286 -0.16% +0.91% $31.11B

    Figures supplied by CoinLore. A blank cell means we could not source that value and left it out rather than estimating it — our methodology explains the provider chain and when that happens.

    How broad was the move?

    Within the top 50 by market capitalisation, 22 advanced and 25 declined over 24 hours. That ratio is worth more than any single asset’s percentage, because it separates a market moving together from one where a couple of large names are doing the work. When advancers and decliners are close to even, a headline index move usually says more about the largest two assets than about the market.

    The strongest 24-hour performers we could measure were Zcash (ZEC) at $503.17, +3.55%; Bittensor (TAO) at $197.85, +3.45%; Polkadot (DOT) at $0.8441, +2.87%. The weakest were TON Coin (TON) at $1.39, -21.81%; Lido DAO (LDO) at $0.2787, -15.77%; Mumu the Bull (SOL) (MUMU) at $0.000703, -6.42%.

    A caution on both lists: the extremes of a movers table are almost always smaller assets, and large percentage moves in thin markets are a function of liquidity as much as of interest. A 20% move on an asset whose order book is a few hundred thousand dollars deep is not comparable to a 2% move in Bitcoin, and treating them as equivalent is a common way to misread a session. See slippage for why the quoted price on such a move may not be obtainable.

    Sentiment

    The Fear & Greed Index reads 25 (extreme fear), +1 points against its reading thirty days earlier, published by Alternative.me. The index compresses volatility, momentum, volume, social activity and Bitcoin’s market share into a single number between 0 and 100.

    It describes mood, not direction. The frequently repeated idea that extreme fear marks a bottom and extreme greed a top is a narrative fitted to history afterwards — the index can sit at an extreme for weeks while price continues the same way. Where it is genuinely useful is as a check on your own behaviour: if a decision feels urgent, knowing the whole market is at an emotional extreme is worth a pause. Our Fear & Greed page shows the current reading with the last thirty days.

    Key takeaways

    • Every figure here was read from our live pipeline at 4 Aug 2026 20:38 UTC and attributed to its provider.
    • Bitcoin $64,116, Ethereum $1,871, total market capitalisation $2.17T.
    • Breadth matters more than a headline move: 22 of the top 50 advanced, 25 declined.
    • Extreme movers are usually thin markets, where percentage moves overstate real interest.
    • Sentiment readings describe mood, not direction. They are not entry signals.
    • Prices here are indicative reference figures, not quotes you can trade on.

    What 24 hours and seven days each tell you

    The two change columns in the table above answer different questions, and reading them together is more informative than reading either alone. The 24-hour figure captures the current session — who is active now. The seven-day figure captures whether that session fits a direction or contradicts it.

    Of the top 50 assets where we can measure both, 33 are moving the same way over 24 hours as over seven days, and 17 are not. A high disagreement count is the more interesting reading: it means today is pushing against the week, which is what a genuine turn looks like early and also what an ordinary bounce inside a trend looks like. The data cannot distinguish those two, and neither can we — anyone claiming otherwise from a single day’s numbers is guessing.

    The other thing worth noting is how many assets are barely moving at all. 37 of the 50 measurable assets in the top 50 changed by less than one per cent over 24 hours. That count includes stablecoins, which should barely move by design, but a high figure across the rest indicates a market waiting rather than one deciding.

    What this data cannot tell you

    Being explicit about the limits is part of reporting honestly, so here they are.

    These are aggregated reference prices, not venue quotes. The figure shown for an asset is a composite, and the price available to you on any particular exchange will differ — by the spread, by fees, and by whatever the book looks like at the moment you act. Do not use anything here for an execution decision.

    Volume is the weakest number on the page. Reported crypto volume has historically included a great deal that is not economically meaningful, and different providers compute it differently. Treat volume as a rough indication of activity rather than a precise measure, and be especially sceptical of volume figures on smaller assets.

    Market capitalisation assumes every unit could be sold at the current price. For an asset with thin liquidity, or one where a large share of supply is locked or held by a few parties, that assumption is simply false — the figure describes an accounting total, not an amount of money that could be realised. Our market capitalisation entry covers this.

    A snapshot has no causal information. We can tell you what changed. We generally cannot tell you why, and where a cause is not identifiable we say so instead of supplying a plausible-sounding one. Most single-session narratives in crypto commentary are constructed after the fact.

    How to use a snapshot like this

    Its value is as a reference point, not a trigger. Three reasonable uses: as context for a decision you had already planned, as a record you can return to later, and as a check on whether your impression of the market matches its actual state — which it often does not, because attention is drawn to whatever moved most rather than to whatever matters most.

    An unreasonable use is treating any figure here as a reason to act. Nothing in this article is a recommendation, and the breadth and sentiment readings in particular are descriptions of the present rather than indications of the future. If you are considering a position, the sequence that protects you is risk management first and sizing second, via the position size calculator — not a market snapshot.

    What this article deliberately does not do

    It does not tell you where price goes next, name a target, or suggest a position. We do not publish calls, and any forecast we report elsewhere belongs to whoever made it and is attributed to them. If you are sizing anything, read risk management and use the position size calculator rather than acting on a snapshot.

    The Priced at Publication band at the top of this article records Bitcoin’s price at the moment it was published and shows the live price beside it. That is a timestamp, not a track record — it exists so that reading this months from now tells you immediately what market it was written into.

    Sources and method

    • Prices, market capitalisations, 24-hour and 7-day changes: CoinLore, retrieved 4 Aug 2026 20:38 UTC.
    • Total market capitalisation, volume and dominance: CoinLore, retrieved 4 Aug 2026 20:38 UTC.
    • Fear & Greed Index: Alternative.me, retrieved hourly.
    • Breadth (advancers and decliners) computed by us from the top 50 by market capitalisation in the dataset above.
    • Full provider chain, fallback order, refresh intervals and how the price freeze works: Coinpric methodology.
    • Live equivalents of every figure here: market table and the individual coin pages.
  • How to size a crypto position so one trade cannot ruin you

    How to size a crypto position so one trade cannot ruin you

    The short version

    Decide what percentage of your account you can lose on one trade before you open it, put your stop where the idea is actually wrong, and let those two numbers determine the size. Do it in that order and position sizing stops being a feeling. The arithmetic of recovery is unforgiving enough that avoiding large losses matters more than catching large gains.

    Most people who lose money in crypto do not lose it because their analysis was bad. They lose it because a position was too large for the account it sat in, so an ordinary adverse move became an unrecoverable one. That is a solvable problem, and solving it requires no forecasting ability at all.

    Start with the arithmetic, because it decides everything else

    A loss and the gain required to undo it are not symmetrical, and the gap widens fast:

    Loss taken Gain needed to recover
    20% 25%
    50% 100%
    60% 150%
    80% 400%
    90% 900%

    At the top of that table you are inconvenienced. At the bottom you are effectively finished, because a 900% gain is not something you can plan around. This is why drawdown is the number experienced traders watch rather than returns: a strategy that makes 40% in good conditions and loses 70% in bad ones is a losing strategy, however impressive the good years look.

    Crypto makes this sharper than other markets because volatility is larger. A daily move that would be a notable session in equities is an ordinary afternoon here. The same position size therefore carries substantially more risk, and position sizing habits imported from slower markets tend to be too aggressive.

    Size from your stop, not from your conviction

    The common approach is to decide how much you want to put in — often a round number, often driven by how good the idea feels — and then place a stop somewhere that looks reasonable. That is backwards, because the size is being set by emotion and the stop by convenience.

    The alternative has three inputs and one output:

    1. Account risk

    What percentage of your trading capital are you willing to lose if this trade is wrong? Not your net worth — the capital you are actually trading. Many experienced traders keep this at or below 1% per position. We are deliberately not going to name a number for you, because it depends on circumstances we cannot see. What we will say is that larger percentages shorten how long you survive a losing streak, and losing streaks are not optional.

    2. Stop distance

    Where does price have to go for the idea to be wrong? Not “where does a loss start to hurt” — those are different questions, and conflating them is how people end up with stops that get hit by noise. If the level that invalidates your thesis is far away, that is information: it means this trade requires a smaller position, not a tighter stop.

    3. Entry

    Where you actually get in. The distance between entry and stop is your risk per unit.

    Then the size falls out of it: position size = (account × risk percentage) ÷ (entry − stop). Our position size calculator does this, and also gives you the reward-to-risk ratio if you supply a target.

    Worked through: a $10,000 account, 1% risk, entry at $60,000 and a stop at $57,000. You are willing to lose $100. Risk per unit is $3,000. So you can hold 0.0333 units — about $2,000 of exposure. If you had instead decided to “put $5,000 in”, the same stop would cost you $250, or 2.5% of the account, and you would not have known that until it happened.

    Reward to risk, and why it filters more than it seems

    Once you know your risk per unit, a target tells you what you are being paid to take it. A setup offering $2 of upside for $1 of risk is 2:1. Below roughly 1.5:1 you need to be right a great deal of the time simply to cover fees and the occasional bad fill — and most people are not right that often.

    This is a useful filter precisely because it is unemotional. It rejects trades that feel compelling but do not pay enough, which is a category most traders have too many of.

    Key takeaways

    • Recovery is asymmetric: a 60% loss needs a 150% gain. Avoiding large losses beats chasing large gains.
    • Decide account risk and stop placement first; let them determine size. Never the other way round.
    • A far-away invalidation level means a smaller position, not a tighter stop.
    • Reward-to-risk below about 1.5:1 requires a hit rate most people do not have.
    • A stop is not a guarantee — in fast markets it can fill well below where you placed it.
    • Crypto assets correlate in a selloff, so several positions can fail together as if they were one.

    The assumption that can still hurt you

    Everything above assumes your stop fills at your stop price. In crypto that is not guaranteed. Books thin out precisely when you need them, prices gap, and during a violent move a stop can fill considerably worse than where it sat — so a real loss can exceed the figure any calculator gives you. Read our entries on slippage and liquidity, and treat the calculated risk as a floor rather than a ceiling.

    Leverage compounds this. A leveraged position can be liquidated — closed by the venue, at the venue’s speed, with your collateral gone — before any recovery is possible. No position-sizing arithmetic protects you from that, because liquidation is not a stop you control.

    Two habits that cost nothing

    First, decide in advance what would make you exit for reasons other than price: a broken assumption, a change in the project, a rule change. “It is down” is not information, and without a non-price exit condition you will hold things for the wrong reasons.

    Second, size on the assumption that several positions could go wrong simultaneously. Crypto assets correlate heavily in a selloff — a portfolio of ten ideas that look unrelated frequently behaves as one during a bad week. If every position is sized at your maximum individually, your portfolio risk is a multiple of what you think it is.

    Frequently asked questions

    What risk percentage should I use?

    That is genuinely your decision and depends on your circumstances, your timeframe and how many positions you hold at once. We will not name a figure. The relationship worth knowing is that the larger it is, the shorter the losing streak you can absorb.

    Does this work for long-term holding rather than trading?

    The sizing arithmetic applies to anything where a total loss is possible, which includes holding. The stop-based version is trading-specific, but the underlying question — how much of this can I lose without it mattering — is the same.

    Should I use leverage if I size correctly?

    We do not make recommendations. What we can tell you is that leverage introduces liquidation, which removes your control over the exit, and that this is a different category of risk from the one position sizing addresses.

    Notes and further reading