What can you see on-chain that no equity market lets you see?

Here is the genuinely new thing in this course. For a public blockchain, the settlement layer itself is a public dataset. Every transfer, every balance, every timestamp, published continuously, in full, to everyone, for free.

Compare that to equities. You get price, you get volume, and you get ownership disclosures that are quarterly, threshold-based and filed with a delay measured in weeks. The trade blotter of the market as a whole is not available to you and never will be. On a transparent chain, an approximation of it is.

This is a real analytical primitive that did not previously exist. It is also routinely oversold, and this unit does both halves.

What is actually derivable

• Address activity — how many distinct addresses transacted, and how many were new
• Transaction counts and transferred value
• Fee levels — what users are paying to get included, a direct congestion read
• Supply distribution — how the coin supply is split across balance buckets
• Coin age — how long each unit has sat unmoved
• Issuance and miner or validator balances
• Entity flows — but only with a labelling layer, which is Lesson 2's entire subject

Active addresses, defined precisely

Active addresses = the number of distinct addresses appearing as a sender or a recipient in at least one confirmed transaction during the window.

Read that definition slowly, because every word in it is a limitation:

• Distinct addresses, not distinct people. One person may control thousands. One address may represent millions of people.
• Sender or recipient. A single payment activates at least two addresses, sometimes more.
• Confirmed on this chain. Anything settled elsewhere is invisible.

Worked: what a headline number is actually made of

A chain reports 900,000 active addresses today, flat year on year. A labelled-entity breakdown might attribute:

• 35% (315,000) to a handful of exchange clusters — where millions of real customers trade inside the exchange's database, producing no on-chain activity at all, while a few hot wallets churn deposits and withdrawals
• 20% (180,000) to automated programs, bots and consolidation traffic
• In a UTXO chain, a further slice to change addresses — a single wallet paying a bill can create a fresh address for its own leftovers, activating two addresses for one economic act
• And on chains with rollups, an entire growing user base settling on Layer 2, batched into a handful of Layer 1 transactions

The headline is flat. Underneath it, exchange share could have doubled, real users could have migrated to Layer 2, and bot traffic could have collapsed. A flat number is not evidence that nothing changed.

How to hold the metric

On-chain activity measures the settlement layer, not adoption, not demand, and certainly not price. It is a high-quality observation of a narrow thing, and its relationship to anything you care about is an inference you are making, not a fact the chain reported. Compare it across time on the same chain with the same definition; be extremely careful comparing it across chains with different architectures.

These are observations. Nothing here is a signal or advice.

Try it now

1. Open a free active-address series for bitcoin. Coin Metrics publishes one in its open community dataset on GitHub (coinmetrics/data, file csv/btc.csv, column AdrActCnt, one row per day); a block explorer's charts page is a second source.
2. The same asset's price over five years is below. Put your address series beside it and answer two questions in writing: do they move together, and does either consistently lead the other?

3. Then find one month where they clearly diverged. Measure the price over that month so you know how large the divergence was, and see whether you can explain it from the definition of an active address given above. Usually you can, and the explanation is that the metric counted something other than what you assumed.