Bitcoin runs on a public ledger. Every transaction, every active address, every hash is visible. A set of derived metrics (active addresses, hash rate, MVRV, SOPR, NVT) describes network health at any moment. These are the tools serious analysts use. They are not trade signals. They are health checks.
READING TIME: 7 MIN
Not financial advice. On-chain metrics are context, not predictions. Thresholds marked [VERIFY] change over time and should be confirmed against current data before you rely on them.
Network value frameworks ask a simple question: does the network's activity justify the market cap? Active addresses, transaction volume, hash rate, and a handful of derived ratios (MVRV, SOPR, NVT) let you answer that. Hash rate is a miner conviction signal. MVRV and SOPR are cycle-position signals. NVT is a usage-to-price signal. Use them as context. Anyone treating them as trade triggers has missed more upside than they have saved.
Metcalfe's Law states the value of a network is proportional to the square of the number of users (n squared). It was originally a telecom observation: a phone network with 10 users has 45 possible connections; 100 users gets you 4,950. The network's utility grows super-linearly with adoption.
Applied to Bitcoin, more holders and more active addresses should drive the market cap higher at a super-linear rate. Empirically, BTC market cap has tracked a Metcalfe-style curve reasonably well over its first 15 years. The correlation is imperfect but persistent.
Daily active addresses (unique addresses that sent or received BTC in the past 24 hours) is the simplest network-health metric. Historically, BTC market cap correlates strongly with active addresses. When active addresses rise faster than price, the network is described as "undervalued" by this lens. When price outruns active addresses, "overvalued."
Caveat: Lightning usage, exchange internal ledgers, and custodial holdings do not show up as on-chain active addresses. The metric undercounts modern activity. Treat it as directional, not absolute.
Total on-chain transaction volume (with change outputs and internal-shuffle noise filtered out) is another network-health metric. Services like Coin Metrics publish an "adjusted" volume series. High adjusted volume combined with a low price is a bullish divergence. Low volume during a price rally is a bearish divergence.
Raw volume is easy to spoof and easy to misinterpret. Adjusted series are more useful but not trivial to compute from scratch. Most analysts rely on Coin Metrics or Glassnode for the clean series.
Hash rate is the total computing power pointed at Bitcoin's proof-of-work. It reflects miner willingness to spend electricity to secure the network, which is a forward-looking bet on future BTC price.
Sustained hash rate increases through bear markets historically precede bull markets [VERIFY historical pattern]. The logic: miners with long-duration capex do not add rigs in bear markets unless they believe the next cycle will be higher. Hash rate all-time-highs during price drawdowns have been a reliable "institutional miners are committed" signal across multiple cycles.
Hash rate is harder to fake than almost any other metric. It represents real electricity spent and real hardware deployed. When hash rate rises during a price drawdown, it is a strong signal that the marginal producer is not capitulating.
MVRV compares current market cap to the "realized cap" (the sum of every coin valued at the price it last moved on-chain). A high MVRV means the average coin is worth far more than its last-moved price, meaning a lot of unrealized profit sits in the market. A low MVRV means the opposite.
MVRV above 3.7 has historically marked cycle tops. MVRV below 1 has marked cycle bottoms [VERIFY thresholds]. The metric lags and the thresholds drift over time, but the general pattern has held for several cycles.
SOPR compares the price at which coins are sold to the price at which they were last bought. SOPR above 1 means holders, on average, are selling at a profit. SOPR below 1 means selling at a loss.
The interesting signal is declining SOPR crossing 1.0 from above. That is the point where "holders giving up" becomes visible in the data. It often marks late-stage capitulation. It is not a guaranteed bottom marker, but it has preceded several major lows.
NVT is BTC market cap divided by daily transaction volume, analogous to a P/E ratio. High NVT means price is outpacing network usage. Low NVT means usage is outpacing price.
NVT has a known flaw: Lightning usage and off-chain batching both reduce on-chain volume without reducing actual economic activity. The metric has drifted upward over time for reasons unrelated to speculation. Use it with the Lightning caveat in mind.
Not as trade signals. As health checks. The investors who do best with on-chain data use it to understand what the market is currently pricing, not to time entries and exits. If hash rate is rising, network security is strengthening regardless of price. If MVRV is below 1, you are in a historical "buy zone" for long-term holders. Neither fact guarantees the next move.
The history of "I'll sell at MVRV greater than 3.7" has caused more missed upside than captured profits. In cycles where the metric topped earlier and the price kept running, disciplined sellers missed the last double. Treat thresholds as context, not rules.
Glassnode, CryptoQuant, Coin Metrics, and mempool.space (free) publish the metrics covered here. Glassnode has the most complete free tier for casual users [VERIFY sources still active]. Coin Metrics has the cleanest adjusted-volume series. mempool.space is a free, non-commercial public-good resource.
Do not pay for premium dashboards until you can explain, from memory, what each metric measures and its known flaws. Paying for data you do not understand is how people build confident, wrong models.
These metrics are useful context. They are not predictions. They describe the state of the network today. They do not tell you what tomorrow brings. The honest stance: check them quarterly, use them to sanity-check your thesis, and keep doing what you were going to do anyway if the broad picture confirms it.
Last updated 2026-04-14. Not financial advice.