On-Chain Analysis · Lesson 01 of 07
Reading a blockchain: addresses & transactions
What it is
Imagine if every bank wire on Earth were published in a public book — every payment, every account balance, every transfer, timestamped, forever, readable by anyone with an internet connection.
That's a blockchain. Not the buzzword version — the practical version. It's a public ledger, and the entire discipline of on-chain analysis is just… reading it.
Three terms and you're literate:
An address is an account number — a long string of characters that holds a balance. Crucially, it's pseudonymous: you can see everything an address has ever done, but not automatically who owns it. Masks on, behavior public.
A transaction is coins moving from one address to another. Amount, sender, receiver, time, fee paid — all visible, all permanent.
A block explorer is the free window into it all (mempool.space for Bitcoin, Etherscan for Ethereum). Paste any address or transaction in, and read.
On-chain analysis is what happens when you read at scale: grouping addresses into entities (this cluster is Binance, that one's an OTC desk, that one's a whale who bought in 2013), tracking how long coins sit still, and watching money move — in a market where the money can't hide its movements.
Why it matters
Nothing like this has ever existed in financial markets. Stock traders get quarterly filings, months late. You get the entire ledger, live.
You've already used it twice without the formal introduction. Exchange Flows (Flows 04) was on-chain analysis — watching coins move to and from tagged exchange addresses. The footprint-reading in Flows 06 leaned on desk wallets. This shelf turns those borrowed tools into a discipline: active addresses, fees, TVL, unlocks, staking — every one of them is a different question asked of this same public book.
And one honest boundary, set now, because it governs everything: the chain tells you what moved, never why. A transfer is a fact; a motive is a guess. Entity labels are detective work — probabilistic, sometimes wrong. The analysts who blow up are the ones who forget which of their statements are ledger facts and which are inferences wearing confidence.
The two readings, always taught together
Read bullish when
- Old coins staying asleep while price rises. Every coin's last-moved date is on the ledger, so you can watch whether long-term holders are tempted by the rally. When ancient coins don't move as price climbs, conviction is holding — supply stays off the market.
- Coins migrating to fresh cold storage. Withdrawals flowing to newly created addresses that then sit still — buyers taking delivery and locking the vault (Flows 04's exodus, seen at wallet level).
- Steady growth in new, real entities. More distinct participants transacting, holding, returning — a network being used, not just traded. (Lesson 02 turns this into a proper metric, warts and all.)
- Miners and old whales holding through stress. The entities with the most coins and the best information declining to sell into fear.
Never alone — confirm with Exchange Flows & price structure
Read bearish when
- Ancient coins waking up. Addresses dormant for five, ten years suddenly moving is the chain's loudest alarm. It doesn't guarantee selling — but coins don't wake up after a decade to do nothing, and the market watches every such move for a reason.
- Heavy flows toward exchange-tagged addresses at highs. Supply traveling from vaults to shelves (Flows 04) — visible at the individual-wallet level, often before it shows in aggregate.
- Hollow participation. Price rising while the count of genuinely active entities shrinks — a rally with fewer and fewer real people in it.
- Miners spending reserves during stress. The industry's forced sellers becoming active — operational selling that doesn't care about your support level.
Never alone — confirm with Exchange Flows & price structure
Visual explanation
Real market example
Summer 2024 — watching a government sell, transfer by transfer. In mid-2024, the German government began moving roughly 50,000 seized Bitcoin (from the Movie2k piracy case). Because the wallets were identified, the entire market watched it happen live: coins leaving the government's addresses, traveling to exchange-tagged wallets and OTC desks, week after week through June and July.
Every transfer was public. Analysts tracked the remaining balance like a countdown clock. The selling pressure was real — it coincided with the market's summer swoon toward $53,000 — but so was the transparency: everyone knew the size of the overhang, watched it shrink, and knew the exact moment the wallets hit zero in mid-July. An identified seller, a visible inventory, a live countdown — try getting that in any other market on Earth.
It's also the perfect specimen of this lesson's mechanics: dormant coins waking (bearish alarm), flows to exchange-tagged addresses (the shelf restocking), entity identification turning anonymous transfers into a readable story — and the limits, too, since observers still had to infer sales from transfers. The chain showed what moved. The why was hypothesis — a well-supported one, confirmed by the German authorities after the fact.
How RIX Intel uses this signal
The chain is the desk's verification layer. It's why publications carry evidence links — on-chain claims are checkable by any reader, which is the entire trust model of this site applied to data. Entity-level watching (seizure wallets, distribution events, treasury movements) feeds the desk's event-risk map: known overhangs get tracked, not feared vaguely.
And the boundary is enforced in writing: on-chain observations enter publications as what moved, with the why framed as hypothesis carrying an invalidation — never as certainty. Tagging is estimation; the desk treats provider labels as clues and cross-checks big claims across sources.
Common mistakes
Where this signal ruins people
Reading addresses as people. One person can run a thousand addresses; one exchange address serves millions of users. Address counts are not user counts — ever. (Lesson 02 lives in this minefield.)
Reading every big transfer as a sale. Wallets reshuffle, custodians rotate, exchanges reorganize internally. Flows 04's three questions apply to every single transfer before it's allowed to mean anything.
Treating labels as facts. "This wallet is X" is a probability, not a verdict — and providers disagree. Anchor conclusions to the strength of the tagging, not the drama of the transfer.
Thinking the chain reveals intent. It reveals behavior. The gap between the two is exactly where your hypothesis — and its invalidation — belongs. Analysts who skip that step aren't reading the chain; they're projecting onto it.