I. Introduction: The Invisible Hand of Time
Bitcoin’s price is determined at the margin — the last satoshi bought and sold. But beneath the surface ticker lies a deeply stratified supply structure where coins behave radically differently based on when they were last moved. This concept, visualized most clearly by Glassnode’s HODL Waves, reveals that Bitcoin is not a homogeneous asset traded freely by rational actors, but rather an onion with layers of conviction, dormancy, and behavioral inertia.
For vintage investors, understanding HODL waves is not academic. The age distribution of supply directly determines:
- How much “real” liquid supply exists at any price level
- When old coins might unlock and overwhelm demand
- Whether the 2010–2012 vintage premium is structurally justified
- How long-term holder behavior shifts across market cycles
This article examines the on-chain evidence, vintage by vintage.
II. The Current State of Supply Age
As of Q1–Q2 2026, Bitcoin’s circulating supply of approximately 19.8 million BTC is distributed across age bands as follows:
| Age Band | % of Supply | Approximate BTC | Vintage Period |
|---|---|---|---|
| < 1 month | 3–5% | 600k–1M BTC | Current cycle active |
| 1–3 months | 2–4% | 400k–800k BTC | Short-term |
| 3–6 months | 3–5% | 600k–1M BTC | Speculative hold |
| 6–12 months | 4–7% | 800k–1.4M BTC | Medium-term |
| 1–2 years | 8–12% | 1.6M–2.4M BTC | Recent accumulation |
| 2–3 years | 6–10% | 1.2M–2M BTC | Post-2023 |
| 3–5 years | 10–15% | 2M–3M BTC | 2021–2023 |
| 5–7 years | 8–12% | 1.6M–2.4M BTC | 2019–2021 |
| 7–10 years | 6–10% | 1.2M–2M BTC | 2016–2019 |
| 10+ years | 12–16% | 2.4M–3.2M BTC | Pre-2016 vintage |
Source: Glassnode HODL Waves, CoinMetrics Supply by Age (synthesized from public data, Q1 2026).
The critical observation: ~30% of all Bitcoin has not moved in 5+ years. This is the highest level of long-term dormancy ever recorded. When cross-referenced with historical HODL wave data from 2021, the supply held 5+ years has increased from approximately 20% to over 30% today — a 50% relative increase in just five years.
III. Vintage-by-Vintage Behavioral Analysis
Each vintage stratum behaves differently. The following table summarizes movement likelihood during bull market peaks (when selling incentives are highest):
| Vintage Period | % of Total Supply | Peak Bull Movement | Supply Stickiness |
|---|---|---|---|
| 2010–2012 | ~2–4% | Near-zero | Maximum — mostly lost or held indefinitely |
| 2013–2016 | ~5–8% | <5% of cohort | Very high — “true believers” |
| 2017–2018 | ~10–13% | 10–20% at peaks | High — accumulated at highs |
| 2019–2020 | ~12–15% | 15–25% at peaks | Medium-high |
| 2021–2024 | ~30–35% | 30–50% at peaks | Medium |
| 2025–2026 | ~10–15% | 50%+ of cohort | Low — current cycle traders |
Source: Glassnode Vintage Analysis, CoinMetrics Age Consumed; behavioral estimates based on on-chain movement correlation with price peaks.
III.A. The 2010–2012 Vintage: Geological Time
Coins mined in Bitcoin’s first three years are effectively geological. Of the approximately 1.2–1.6 million BTC mined in 2010, only fragments have ever moved since 2013. Chainalysis estimates that 60–70% of early-mined coins are permanently lost (Satoshi’s 1M+ BTC, unrecovered wallets, discarded hard drives). The remaining liquid supply of 2010 coins is estimated at less than 120,000 BTC.
These coins do not respond to price signals. Even during the 2021 bull run — Bitcoin’s highest price ever — on-chain data showed minimal movement of pre-2013 UTXOs. This creates a structural price floor: the oldest coins are permanently removed from the liquid supply curve, meaning that each new market cycle faces a tighter supply of truly old coins.
III.B. The 2013–2016 Vintage: Strong Hands
The 2013–2016 vintage represents the first wave of “post-mania” accumulation. These holders bought during or after the 2013 bubble and held through the multi-year bear market of 2014–2016. Their behavioral signature is distinct:
- Average holding period: 10–13 years
- Movement at 2021 peak: <5% of cohort
- Current state: >95% dormant
These coins move almost exclusively through inheritance events (estate transfers), security upgrades (migration to new wallets), or extreme price dislocations. They are best thought of as generational holdings rather than investment positions.
III.C. The 2017–2020 Vintage: The Conviction Layer
This vintage covers the 2017 bull, the 2018–2020 bear, and the COVID-era accumulation. Holders in this band have been tested by two severe drawdowns (>80% in 2018, >50% in 2020). Their behavior:
- Average holding period: 5–9 years
- Movement at 2021 peak: 10–20% of cohort
- Current state: 80–90% dormant
This layer provides the bulk of “long-term holder supply” that analysts track. When these coins begin moving en masse (the “old whale distribution” signal), it historically marks the late stages of a bull market.
III.D. The 2021–2024 Vintage: Institutional Accumulation
The 2021–2024 vintage is unique in Bitcoin’s history — it includes the first major wave of institutional and corporate buying (MicroStrategy, ETF flows, sovereign wealth funds). Behavioral characteristics:
- Average holding period: 2–4 years
- Movement at recent peaks: 30–50% of cohort
- Current state: 50–70% dormant
This cohort is more reactive to price than older vintages but far less reactive than short-term traders. The institutional component introduces a new variable: regulated selling windows, tax-loss harvesting, and SEC-compliant custody changes that create movement patterns unrelated to pure market sentiment.
IV. The Liquidity Spiral: How Vintage Hardening Works
As each vintage stratum ages, it undergoes a process we call vintage hardening — coins become progressively less likely to move with each year they remain dormant.
The mechanism is intuitive: the longer a UTXO stays unmoved, the more likely it is that:
- The private key is lost (device failure, death of holder)
- The holder is indifferent to price (wealth effect)
- The holding has emotional/dynastic value (family inheritance)
- The coin is in cold storage with no active trading strategy
Data from CoinMetrics’ Coin Days Destroyed (CDD) confirms this pattern. While the metric spikes during major bull peaks (indicating some old coins move at the exact top), the baseline level of CDD has been declining steadily since 2017, even as total network transaction volume has grown. This means old coins are moving less frequently, even in absolute terms, as a percentage of their cohort.
The Estimated Hardening Curve
| Years Since Last Move | Probability of Moving Next Year |
|---|---|
| < 1 year | 40–60% |
| 1–3 years | 20–35% |
| 3–5 years | 10–20% |
| 5–7 years | 5–12% |
| 7–10 years | 3–8% |
| 10+ years | < 2% |
Source: Estimated from CoinMetrics Age Consumed and Glassnode HODL Waves decay rates.
After 10 years of dormancy, a UTXO has less than a 2% annual probability of moving. This is not because these holders are more “committed” — it is because the combination of lost keys, generational transfers, and wealth insensitivity creates near-permanent supply absorption.
V. Market Implications
V.A. Effective Liquid Supply Is Smaller Than Believed
The common narrative — “19.8 million BTC in circulation” — is misleading for price analysis. The effective liquid supply — coins available for purchase within a reasonable timeframe — is far smaller.
| Supply Category | Estimate | % of Total |
|---|---|---|
| Permanently lost (pre-2016) | 3–4M BTC | 15–20% |
| Dormant 5+ years (not lost) | 3–4M BTC | 15–20% |
| Slow-moving (1–5 yr dormancy) | 6–7M BTC | 30–35% |
| Liquid (< 1 yr dormancy) | 5–6M BTC | 25–30% |
When ETF flows, institutional accumulation, and retail buying compete for the active liquid supply of 5–6 million BTC, the real supply-demand balance is far tighter than headline circulation suggests.
V.B. Vintage Strata Create a Natural Price Floor
Each vintage layer acts as a progressively higher floor for price. The oldest coins that do move — typically 2013–2016 vintage — only transact at prices far above their acquisition cost. This creates a “tail risk absorber” effect: price can fall significantly without triggering old-coin liquidation, because holders of those coins simply do not sell at current market prices.
V.C. The Next Major Unlock Risk
The largest risk to vintage premium is a sudden unlock event — a lost wallet recovery, an estate distribution, or a regulatory seizure of old coins. Historical examples include the 2020 PlusToken distribution (though not vintage-specific) and periodic Silk Road coin auctions. Each such event creates temporary price suppression in the specific vintage layer affected.
VI. Conclusion
HODL waves are not just a visualization — they are the on-chain signature of Bitcoin’s supply architecture. The data is clear:
- Older vintages are almost entirely dormant and becoming more so each year
- The 5+ year dormant supply has increased 50% in five years (from ~20% to ~30%)
- Vintage hardening is a self-reinforcing process: each year of dormancy reduces the probability of movement
- Effective liquid supply (~5–6M BTC) is roughly one-quarter of headline circulation
For vintage investors, the thesis is straightforward: the supply of truly old coins is structurally shrinking, while demand for provenanced, time-stamped assets is institutionally expanding. The intersection of these two trends — declining liquid vintage supply and rising demand — is the fundamental driver of vintage premium.
As Glassnode’s HODL Waves show us, Bitcoin supply is not a fixed pie. It is a slowly freezing lake — and each year, more of the surface turns to ice.
⚠️ Investment Risk Disclaimer The information provided on VintBTC.com is for educational and informational purposes only. It does not constitute financial advice, investment recommendation, or solicitation. Vintage cryptocurrency markets are illiquid, unregulated, and carry high risk including total loss of capital. Past performance of vintage coins does not guarantee future returns. Always conduct your own research (DYOR) and consult a licensed financial advisor before making investment decisions.