In traditional markets, you cannot see who moved money where. You get prices, volumes and quarterly filings – the plumbing stays hidden. Public blockchains flip that: every transfer of bitcoin since 2009 is sitting in an open database anyone can download. On-chain analysis is the practice of reading that database for clues about market behaviour. It is a genuinely useful lens, and also one of the easiest places in crypto to fool yourself with a chart that looks scientific.
What on-chain analysis actually is
On-chain analysis means studying the data recorded on a blockchain itself – transactions, addresses, balances, fees, block contents and smart-contract activity – rather than prices on exchanges. A full node holds all of it. Analytics platforms process it into charts and metrics, and many traders simply use those dashboards.
What makes it distinctive:
- It is complete. Every on-chain transaction is recorded; nothing is sampled.
- It is verifiable. Anyone running a node can check the raw data.
- It is pseudonymous. You see addresses, not names. Everything about "who" is inference.
That last point shapes everything that follows.
Key on-chain metrics traders watch
Some of the most common on-chain metrics, and what they try to measure:
| Metric | What it measures | Common interpretation |
|---|---|---|
| Exchange inflows/outflows | Coins moving to or from addresses linked to exchanges | Inflows can mean selling intent; outflows can mean moving into custody |
| Exchange reserves | Total coins held at known exchange addresses | Falling reserves are read as less supply readily available to sell |
| Active addresses | Unique addresses transacting in a period | A rough proxy for network usage |
| Transaction fees | What users pay to get into blocks | Demand for block space |
| Coin age / dormancy | How long coins sat unmoved before being spent | Old coins moving may signal long-term holders selling |
| Realised price / MVRV | Average price at which coins last moved, compared with market value | Whether holders on average sit on profits or losses |
| SOPR | Whether coins spent were moved at a profit or loss | Short-term profit-taking or capitulation |
| Stablecoin supply on exchanges | Stablecoins available on trading venues | "Dry powder" that could be used to buy |
On UTXO-based chains such as Bitcoin, every coin carries a record of when it last moved, which is what makes metrics like realised price and coin age possible. Account-based chains like Ethereum lend themselves more to tracking contract activity, token flows and DeFi usage.
What blockchain data analysis can tell you
Used carefully, on-chain data helps with:
- Context for big moves. A sharp drop alongside heavy exchange inflows from long-dormant coins is a different story from a drop driven by derivatives liquidations with little on-chain movement.
- Holder behaviour over cycles. Aggregates of coin age and realised profit or loss show how different cohorts have historically behaved.
- Network health. Fees, hash rate and active usage show whether a chain is actually being used.
- Tracing specific events. Hacks, exchange wallet movements and large transfers by known entities are visible in real time.
- DeFi positioning. On smart-contract chains, you can see loan sizes, collateral levels and where large liquidations would trigger.
What on-chain data can't tell you
This is where most beginner mistakes happen.
- Addresses are not people. One person can control thousands of addresses; one exchange address can represent millions of customers. "Whale wallets" are often exchanges, custodians or ETF custody addresses.
- Labelling is guesswork. Analytics firms cluster addresses using heuristics. Labels can be wrong or outdated, and different providers can disagree.
- Most trading never touches the chain. Trades inside a centralised exchange are entries in its internal database. Derivatives activity – often the main driver of short-term price – is largely invisible on-chain.
- Internal shuffling looks like flows. When an exchange moves funds between its own wallets, naive metrics can read it as inflows or outflows.
- Layer 2s and other chains split the picture. Activity on rollups, sidechains or other networks may not show up in a single-chain metric.
- Correlation is not prediction. A metric that "called" past tops in hindsight can stop working once traders watch it, and there are very few full market cycles to test against.
A practical way to use on-chain analysis
- Start with a question, not a dashboard: "Are long-term holders selling into this rally?"
- Pick metrics that answer it, and read their methodology – what counts as an exchange, how cohorts are defined.
- Compare against price and derivatives data (funding rates, open interest) to see whether the on-chain story is the main driver.
- Check against a second data provider for any surprising reading.
- Treat conclusions as probabilities, and size any decision as though you might be wrong.
- Watch for regime changes. New structures, like ETFs holding coins with custodians, shift what old metrics mean.
For teams that want their own view, on-chain data can be pulled from a node or an indexing service into a custom dashboard – the kind of data tooling Mraviko builds – but the interpretation problems above apply no matter how polished the charts are.
A lens, not a crystal ball
On-chain analysis gives crypto traders something no other market offers: a complete, public ledger of every settlement. That makes it excellent for understanding what happened and who might be moving. It does not reveal identities, it misses most trading activity, and its metrics are interpretations built on assumptions. Use blockchain data to add context to your thinking – not to replace it.
This article is for education only and is not financial or investment advice.
