An equity trader can see price and volume. What they cannot see is how many shareholders are sitting on losses, how long the average holder has held, or whether large holders are moving stock to a broker in preparation to sell.
In crypto, all of that is public. Every transaction is recorded on a ledger anyone can query. This is the single biggest analytical advantage the asset class offers, and most traders ignore it in favour of drawing lines on a chart.
The catch is that on-chain data is noisy, frequently misread, and almost never a timing tool. Used as a measure of conditions rather than a source of entries, it is extremely valuable.
Exchange flows
The most intuitive place to start. Coins sitting in a personal wallet cannot be sold on an exchange. Coins deposited to an exchange usually can.
Exchange inflows rising sharply means holders are moving coins somewhere they can be sold. Sustained heavy inflows have historically preceded periods of weakness.
Exchange outflows mean coins are moving to self-custody, which removes them from immediate selling supply. Extended outflows suggest accumulation by people who intend to hold.
Exchange balance, the total held on exchanges, is the slower and more reliable version. A multi-month decline in exchange balances reduces the supply available to sell into demand.
The misreadings are worth knowing. A single large inflow may be an exchange rebalancing its own wallets rather than a whale preparing to dump. Institutional custody moves can look like exchange flows. And as more trading moves to derivatives, spot exchange balances say less about selling pressure than they used to.
MVRV
Market value to realised value compares the current market cap to the realised cap, which values every coin at the price it last moved rather than the current price.
Realised cap is effectively the aggregate cost basis of the network. MVRV is therefore the ratio of what the market is worth now to what holders collectively paid.
- MVRV above 3.5 historically corresponds with market tops. Holders are sitting on very large unrealised gains, and the incentive to take profit is high.
- MVRV below 1 means the average holder is underwater. This has historically marked major bottom regions.
- MVRV around 1 to 2 is unremarkable middle ground.
MVRV Z-Score normalises this against historical volatility, which makes comparisons across cycles more meaningful. Use the Z-Score version if you can get it.
The important caveat: these thresholds come from a small number of past cycles. Bitcoin has had four. Drawing confident conclusions from four observations is statistically thin, and the ranges have drifted lower each cycle as the asset matured.
SOPR
Spent output profit ratio measures whether coins moving on-chain are being moved at a profit or a loss. A value above 1 means coins are moving in profit. Below 1 means they are moving at a loss.
The useful signal is not the absolute value but what happens when SOPR approaches 1 in a trend.
In a bull market, SOPR repeatedly dips towards 1 and bounces. Holders refuse to sell at a loss, so selling pressure stops at breakeven. When SOPR breaks decisively below 1 and stays there, that behaviour has changed and it is a meaningful warning.
In a bear market, the mirror applies: SOPR rallies towards 1 and gets rejected as trapped holders exit at breakeven.
Adjusted SOPR filters out coins moved within a short window, which removes a lot of internal transfer noise. Prefer it where available.
Supply in profit
The percentage of circulating supply whose last movement was at a lower price than today. Simple and effective.
When supply in profit exceeds roughly 95 percent, nearly everyone is winning, which historically has not lasted. When it falls below 50 percent, most of the network is underwater, which is capitulation territory.
What makes this metric useful is that it maps directly to a behavioural reality: people sell differently depending on whether they are up or down.
Holder age bands
Coins can be grouped by how long they have sat unmoved. The usual split is short-term holders, under 155 days, and long-term holders, over 155 days.
Long-term holder supply rising means coins are maturing into strong hands. Long-term holder supply falling sharply means experienced holders are distributing, which has typically happened into strength rather than weakness.
Watching long-term holder behaviour around a rally tells you something a price chart cannot: whether the people who bought low are using the rally to exit.
Metrics that get overrated
Whale wallet counts. Wallet clustering is imprecise. Exchange and custodian wallets distort the picture, and one entity can hold many addresses.
Active addresses. Once a reasonable proxy for network usage, now heavily affected by layer 2 activity, batching and automated systems.
Hash ribbons and miner metrics. Miner selling was a much larger share of supply in earlier cycles than it is now. The relationship has weakened considerably.
Anything presented as a single indicator that called every top. With enough parameters, any metric can be fitted to four historical events.
How to actually use this
On-chain data works on a timescale of weeks to months. It is a conditions tool. Trying to use MVRV to time an entry this afternoon is a category error.
A sensible workflow looks like this. Once a week, check where the market sits on two or three structural metrics: MVRV Z-Score, supply in profit, and exchange balance trend. Use that to decide how aggressive your position sizing should be and whether you are looking for longs or shorts.
Then trade your normal technical process inside that bias. On-chain sets the weather. Technical analysis picks the moment.
Data sources and their limits
Several platforms publish on-chain analytics, some free and some by subscription. They do not always agree, because each makes different choices about address clustering and entity labelling. Two providers can show materially different exchange balances for the same day.
This is worth knowing before you build conviction on one provider's chart. Where a signal matters, check it against a second source.
The other limit is coverage. On-chain analysis is strongest for Bitcoin, reasonably good for Ethereum, and thin for most other assets. Tokens that live mostly on exchanges or across many chains produce data too fragmented to read confidently.
Starting out
Pick three metrics, not fifteen. Exchange balance, MVRV Z-Score and supply in profit will cover most of what a trader needs from on-chain data.
Chart them against price over several years and look at what each one did around the moves you already know about. That exercise is worth more than reading another list of metrics, because it teaches you how much lead time each one actually gives, which is usually less than its advocates suggest and more than a price chart offers.