Comparing Crypto-Native On-Chain Valuation Frameworks

Research question and scope

Kaavon's indicator articles each explain NVT, MVRV, Realized Capitalization, and SOPR individually — what each measures, how it's calculated, and its own specific limitations. This article asks a different question: how do these four metrics relate to each other as a combined toolkit, what distinct question does each one actually answer, and where do they genuinely overlap versus diverge? It does not introduce a new metric, present live values, or claim any metric predicts future price.

Background and context

Realized Capitalization — valuing each coin at the price it last moved on-chain, rather than at the current market price — was introduced by the Coin Metrics team (Nic Carter and Antoine Le Calvez). MVRV was built directly on top of that work: Murad Mahmudov and David Puell published the ratio in an October 2018 analysis, framing a reading above 1 as the aggregate holder base sitting on unrealized gains and a reading below 1 as the aggregate holder base sitting on an unrealized loss. NVT and SOPR each have their own separate origins and mechanics, covered in Kaavon's own dedicated guides for those two metrics; this article's background focus here is MVRV and Realized Cap specifically, since that is the pairing independently re-verified for this pass.

Methodology

This remains Kaavon's own analytical, structural synthesis of four already-documented metrics, not a presentation of live readings. For this pass, Glassnode's own MVRV metric-guide documentation was fetched and read directly (not found only via search synthesis) to ground the stock/flow distinction and the historical threshold claims below in a named technical source, rather than resting solely on Kaavon's own prior editorial framing. NVT's and SOPR's specific calculation mechanics were not independently re-verified in this pass — they rely on Kaavon's existing dedicated guides for those two metrics, cross-referenced in the References section below.

Key findings

• All four metrics are genuinely distinct, not redundant restatements of the same idea — NVT relates valuation to transaction activity; MVRV and Realized Cap relate valuation to aggregate cost basis; SOPR relates to realized profit/loss on coins actively moving.
• MVRV and SOPR ask a related but structurally different question: MVRV reflects the entire outstanding supply at once (a stock), while SOPR reflects only coins that moved during a specific period (a flow).
• No single metric in this family is documented as more reliable than the others — each has its own independently-documented limitations, and combining them narrows, but does not eliminate, the interpretive gaps any one metric carries alone.

What each metric actually answers

Stock versus flow: the key structural distinction

Realized Capitalization and MVRV are both "stock" measures — they describe the entire outstanding supply at a single point in time, regardless of whether any of it is actually moving. SOPR is a "flow" measure — it only reflects the subset of coins that were actually spent during the period in question. This is a genuine, structural difference, not a matter of degree: a market can show an elevated MVRV (aggregate unrealized gains sitting across the whole supply) while SOPR remains near 1 (coins currently moving aren't being sold at a notable profit or loss), if the holders sitting on gains simply aren't transacting.

Where the four metrics tend to agree, and where they can diverge

Historically, NVT, MVRV, and SOPR have tended to move in broadly the same direction during large, sustained market cycles, since all four ultimately respond to the same underlying relationship between price and network activity or cost basis. They can diverge meaningfully over shorter windows — for example, NVT can rise (valuation outrunning transaction activity) while MVRV stays moderate (the supply's aggregate cost basis hasn't yet caught up to the price move), since the two metrics are driven by genuinely different inputs (transaction volume versus realized cost basis).

Evidence: documented historical threshold readings for MVRV

Glassnode's own MVRV metric-guide documentation, read directly for this pass, states that a reading above 3.5 "has generally served as a strong signal for late stage bull cycles," and separately that readings below 1 "have typically provided strong signal of market capitulation and late stage bear accumulations." The same documentation describes MVRV as a "mean reversion style model," with Realized Cap — defined there as the "aggregate market cost basis" of the supply — functioning as the mean price tends to revert toward. These are the two concrete, named thresholds most commonly cited for this metric family, and they come from the metric provider's own documentation rather than from Kaavon's prior editorial framing alone.

Independent critical analysis

Glassnode's own documentation states in one place that "MVRV Ratio signals leave very little room for ambiguity," while describing the same >3.5 and <1 thresholds elsewhere on the same page using consistently hedged language — "historically," "generally," "typically." Those two framings sit in tension: a threshold described as holding only "historically" and "generally" is, by definition, exactly the kind of signal that does leave room for ambiguity in any specific instance, since hedge language is how a source signals that a pattern is not guaranteed to repeat. This article's own stock/flow distinction sharpens the point: the documentation concedes a low MVRV "may signal both undervaluation, or poor demand dynamics" — two readings with opposite investment implications — and a market can simultaneously show an elevated, "historically bearish" MVRV while SOPR sits near 1 because the gain-sitting holders simply aren't transacting. In that concrete combined-metric scenario, which this article's own framework predicts can occur, the "very little room for ambiguity" claim does not hold; the metrics are genuinely complementary, but reading them together surfaces real interpretive disagreement in some states of the market rather than resolving it outright.

Limitations of this comparison

• This article describes how the four metrics relate conceptually and structurally — it does not present or imply any current numeric reading for any of them.
• Each metric's own independent limitations (detailed in its dedicated guide) still apply in full when the metrics are used together, not just individually.
• Historical co-movement between these metrics in past cycles is a documented pattern, not a guarantee of how they'll relate in any future cycle.
• The >3.5 / <1 MVRV thresholds above are drawn from one provider's (Glassnode's) own documentation; they are historical observations the provider itself hedges with "historically" and "generally," not fixed, provider-independent rules, and NVT's and SOPR's own threshold conventions were not re-verified against primary documentation in this same pass.

Research conclusion

NVT, Realized Capitalization, MVRV, and SOPR form a genuinely complementary, non-redundant toolkit for on-chain valuation — each answering a structurally distinct question (transaction-relative valuation, aggregate cost basis, aggregate unrealized gain/loss, and realized profit/loss on active coins, respectively) rather than four versions of the same idea. Reading them together provides a fuller on-chain picture than any single metric alone, without eliminating the real, independently-documented limitations each one still carries — and, as the >3.5-MVRV/near-1-SOPR case above shows, combining metrics can surface genuine disagreement between them rather than always resolving to a single clean signal.

References

• Primary (read directly in full for this pass): Glassnode docs, "MVRV Ratio" metric guide (docs.glassnode.com/guides-and-tutorials/metric-guides/mvrv/mvrv-ratio) — formula, Realized Cap definition, >3.5/<1 threshold language, "very little room for ambiguity" claim.
• Secondary (found via WebSearch, not independently fetched in full): Forklog and Woobull accounts crediting Murad Mahmudov and David Puell with MVRV's October 2018 publication, built on Coin Metrics' (Nic Carter, Antoine Le Calvez) prior Realized Cap work.
• Internal cross-reference: Kaavon's own dedicated NVT, MVRV, Realized Capitalization, and SOPR indicator guides, for each metric's full individual mechanics and limitations (not re-verified against new primary sources in this pass for NVT and SOPR specifically).