Coin Days Destroyed (CDD)

What Is Coin Days Destroyed?

Coin days destroyed is an on-chain metric that measures the economic significance of Bitcoin transactions by accounting for how long the coins sat idle before being spent. Every unspent Bitcoin accumulates one "coin day" per day it remains untouched. When that Bitcoin finally moves, those accumulated coin days are "destroyed" and tallied.

The concept was first proposed by a user named ByteCoin on the BitcoinTalk forum in April 2011. ByteCoin observed that raw transaction volume is a poor indicator of genuine economic activity because anyone can inflate it by sending coins back and forth between their own addresses. CDD solves this by giving greater weight to coins that have been dormant for longer periods: a transaction moving coins held for years carries far more weight than one moving coins received minutes ago.

As a simple example: if 10 BTC sits untouched for 100 days and then moves, that transaction destroys 1,000 coin days. By contrast, 1,000 BTC moving after just one day also destroys 1,000 coin days. This weighting makes CDD resistant to wash trading and artificial volume inflation, which is why on-chain analysts consider it a more reliable gauge of meaningful economic activity than raw transaction counts.

How It Works

CDD is calculated directly from Bitcoin's UTXO model. Each unspent transaction output has a creation timestamp (when it was received) and a value in BTC. When a transaction spends that output, the formula is straightforward:

CDD = BTC_amount × days_since_last_spent

Daily CDD = SUM over all spent outputs that day:
  (value_of_output × lifespan_of_output)

For a concrete example: a transaction with two inputs would be calculated as follows:

Input 1: 5 BTC held for 200 days  → 5 × 200 = 1,000 coin days
Input 2: 3 BTC held for 50 days   → 3 × 50  =   150 coin days
                                     Total CDD = 1,150 coin days

UTXO-Level Calculation

Computing CDD requires indexing every spent output in each confirmed block. For each UTXO consumed as a transaction input:

1. Look up the block height (or timestamp) when the UTXO was created
2. Determine the current block height (or timestamp) when it is being spent
3. Calculate the holding period in days
4. Multiply the BTC value of that UTXO by the holding period
5. Sum across all spent outputs in all transactions confirmed that day

Running this calculation requires a full Bitcoin node with a spent-output index, making it computationally intensive but fully verifiable by anyone running their own node. Analytics platforms like Glassnode and CryptoQuant run these calculations continuously and expose the results through dashboards and APIs.

Why CDD Filters Noise

Consider two scenarios that produce identical raw transaction volume of 100 BTC per day:

Raw volume treats both scenarios identically. CDD reveals the second as 1,825 times more economically significant, correctly flagging that long-dormant coins are on the move.

Derived Metrics

Supply-Adjusted CDD

Because Bitcoin's circulating supply grows over time (through block rewards), raw CDD naturally trends higher as more coins exist and accumulate more coin days. Supply-adjusted CDD normalizes for this:

Supply-Adjusted CDD = CDD / Circulating Supply

This allows meaningful comparison of CDD values from 2012 (when roughly 10 million BTC existed) with values from 2026 (with over 19.8 million BTC in circulation). Without this adjustment, historical CDD readings appear artificially low relative to recent ones.

Binary CDD

Binary CDD simplifies the signal to a 0 or 1 for each day:

• Returns 1 when the day's supply-adjusted CDD exceeds its long-term historical average, indicating above-average old-coin spending
• Returns 0 when below average, indicating normal or low old-coin activity

Sustained streaks of 1s correspond to distribution phases where long-term holders are actively selling. Sustained 0s indicate accumulation phases. Alternating "barcode" patterns of 0s and 1s are typical of uncertain, range-bound markets.

Average Coin Dormancy

Formalized by researcher Reginald D. Smith in a 2018 paper published in the journal Ledger, average coin dormancy divides CDD by on-chain transaction volume:

Average Coin Dormancy = CDD / On-Chain Volume (BTC)

This yields the average number of days each transacted coin was dormant before moving. High dormancy means older coins are being spent (potential distribution by long-term holders); low dormancy means recently acquired coins are trading (normal market activity or accumulation).

Dormancy Flow

Created by on-chain analyst David Puell, dormancy flow relates market capitalization to annualized dormancy:

Dormancy Flow = Market Cap / (365-day MA of Daily Dormancy × Price)

When dormancy flow is high, the market's valuation is large relative to annualized spending behavior: price is rising faster than old coins are being sold. When dormancy flow drops to historically low levels, it signals full capitulation where heavy old-coin spending overwhelms market cap, which has historically marked strong buying opportunities.

Use Cases

Market Cycle Analysis

Analysts tracking Bitcoin market cycles use CDD as one of several indicators to identify phase transitions:

• Accumulation phases show persistently low CDD as long-term holders sit on their positions and new buyers acquire coins for holding
• Distribution phases show elevated CDD as early buyers and long-term holders begin moving coins to exchanges, often for profit-taking near cycle peaks
• CDD spikes exceeding 20 million coin days in a single day are rare events that have historically aligned with periods of high volatility

Whale Monitoring

Because CDD naturally amplifies the signal from large, old UTXOs, it serves as a proxy for whale activity. When a wallet holding thousands of BTC acquired years ago suddenly moves its coins, the CDD spike is unmistakable. This complements chain analysis tools that track specific addresses by providing an aggregate, network-wide view of long-term holder behavior.

UTXO Health Assessment

CDD contributes to understanding the age distribution of Bitcoin's UTXO set. Periods of low CDD mean the UTXO set is aging: more coins are sitting idle for longer. Periods of high CDD mean the set is "rejuvenating" as old outputs are spent and replaced by new ones. This age profile is relevant for UTXO management strategies and understanding network-wide holding patterns.

Filtering Artificial Volume

Exchanges and market makers regularly move large amounts of Bitcoin between internal wallets for operational reasons. These movements inflate raw transaction volume but destroy minimal coin days because the coins are typically only held briefly. CDD strips away this noise, revealing the underlying pace of genuine economic activity driven by holders who are making deliberate decisions to move long-held coins.

Risks and Considerations

False Signals from Internal Transfers

Not every CDD spike indicates selling intent. Large holders may move coins between their own wallets for security reasons (migrating to a new cold storage setup), for UTXO consolidation during low-fee periods, or as part of custodial reorganizations. These transfers destroy coin days without any change in ownership or selling pressure.

No Directional Signal

CDD measures that old coins moved, not why they moved. A spike could indicate profit-taking, loss-cutting, inheritance transfers, exchange cold wallet maintenance, or simply an owner upgrading their key management. Analysts should combine CDD with exchange flow data, price action, and other on-chain metrics to determine the likely motivation behind coin movements.

Supply Growth Distortion

As noted above, raw CDD increases over time simply because more coins exist and accumulate more coin days. Comparing raw CDD values across different eras without supply adjustment can lead to incorrect conclusions. Always use supply-adjusted variants when analyzing long-term trends, especially when comparing pre- and post-halving epochs.

Lag as a Predictive Tool

CDD is a coincident or lagging indicator: it tells you that old coins moved, after they have already moved. By the time a significant CDD spike appears in daily data, the market impact may already be priced in. It is most useful as a confirmation signal alongside leading indicators rather than as a standalone prediction tool.

Why It Matters

For anyone analyzing the Bitcoin network, CDD provides a lens that raw transaction volume cannot: it distinguishes routine activity from economically meaningful movements. When combined with its derived metrics, CDD offers a framework for understanding market cycles, holder behavior, and network health that has proven valuable since ByteCoin first proposed it in 2011.

On-chain analytics platforms like Glassnode and CryptoQuant include CDD and its variants as core metrics, and researchers continue to build on the concept. The 2018 formalization of average dormancy by Reginald D. Smith connected CDD to queueing theory (specifically Little's Law), estimating the size of Bitcoin's active trading pool from the relationship between coin days destroyed and transaction volume.

For Bitcoin layer-2 protocols and off-chain systems, CDD remains an exclusively on-chain metric. Transactions on the Lightning Network or other layer-2 solutions do not destroy coin days because the underlying UTXOs remain unspent on the base layer until channels are closed. This means CDD captures only layer-1 activity, making it a complement to (not a replacement for) off-chain analytics.

This glossary entry is for informational purposes only and does not constitute financial or investment advice. Always do your own research before using any protocol or technology.