Loss Volume in Bitcoin

What is it?

When a coin moves, the chain records both what it is worth now and what it was worth when it last moved. If the second figure is higher, that coin changed hands at a loss. This chart counts those coins, and the choice of unit is the entire point: it counts them in bitcoin, not in dollars. The reason is that dollar-denominated loss charts silently rewrite history. A given dollar amount of realised loss represented an enormous number of coins in the early years and a modest number today, because the price rose by orders of magnitude in between. Plot losses in dollars across the full history and every early cycle flattens into nothing next to the recent ones, not because less was happening but because each coin was worth less. Counting units removes both price inflation and network growth from the comparison, and the cycles become genuinely comparable. Two cohorts are shown separately. Long-term holders are those who have held through time and whose selling at a loss represents genuine conviction breaking. Recent holders turn over far more volume, an order of magnitude more, because their activity includes the ordinary churn of exchanges and custody. Mixing them into one line would let the churn drown the quieter but more meaningful behaviour of the patient cohort.

How to read

The vertical axis is bitcoin per day. The long-term cohort appears as a thick thirty-day mean with a thinner seven-day mean over it. The thirty-day line gives the underlying trend, the seven-day line gives the current impulse, and the distance between them tells you whether pressure is building or fading. What matters most is not the absolute level on any given day but the comparison with the same measure in previous cycles, which the long history makes possible. A reading that looks alarming in isolation may be modest against past stress, and the reverse. Rising floors between successive waves are worth more attention than any individual peak. When each successive trough sits higher than the last, the cohort is not simply reacting to events but progressively giving up, which is a slower and more structural process than a single spike. The logarithmic scale is available and often preferable, because loss volume spans several orders of magnitude between calm periods and stressed ones.

Key zones

There are no absolute thresholds, and any that were published would age badly as the network grows. The useful reference is always the historical distribution of the same cohort. Three comparisons carry most of the information: • The current thirty-day level against the highest thirty-day levels reached in past stressed periods. This places the present episode on a scale that spans the whole history. • The seven-day mean against the thirty-day mean. Above it, pressure is accelerating. Below it, the wave is receding. • The trough level between waves against the troughs of earlier waves in the same episode. Rising troughs indicate accumulating fatigue in the cohort. The long-term cohort is the one worth watching most closely. Its volume is smaller by construction, but a coin held for years that moves at a loss carries a different meaning from a coin bought last month.

What to observe

• Rising troughs between waves within the same episode, which describe a cohort progressively giving up rather than reacting to a single event. • Divergence between the two cohorts. Recent holders capitulating while the long-term cohort stays quiet is a very different market from both giving up together. • How the current episode compares in magnitude with past stressed periods, which only a unit-denominated measure allows. • Whether volume rises while price barely moves. Selling at a loss into a flat market means supply is being absorbed without a price concession, which is not the same as a decline. • The shape of the decay after a peak. A sharp spike that fades quickly and a plateau that persists for weeks are different behaviours. • Coincidence with the regime map. Loss volume rising while the structural state is already weak is a different situation from loss volume rising during a constructive configuration.

Historical context

Denominating in units is what allows the early cycles to remain visible at all. In dollar terms the first cycles are a flat line at the bottom of any chart that also contains the recent ones. In coin terms they are fully comparable, and it becomes apparent that some early episodes moved a larger share of the circulating supply at a loss than anything since. The long-term cohort has a natural upper bound that grows only slowly, since it can only contain coins that have already been held for a long time. The recent cohort has no such constraint and its volume tracks overall market activity, which is why its peaks are so much larger and why the two need separate scales. Because the measure counts transferred volume rather than net selling, it includes movements that are not sales in any economic sense. This inflates the absolute level, particularly for the recent cohort, without affecting the comparison over time as long as the same definition is applied throughout, which it is.

Expert notes

This measure is taken directly from a unit-denominated source series rather than derived by dividing a dollar amount by a price. That distinction matters more than it sounds. Converting a realised loss in dollars back into coins would require knowing the average purchase price of the coins that moved, which is not the same as the cohort average, and the result would be an approximation carrying an error that varies with market conditions. Reading a natively denominated series avoids the question entirely. The seven and thirty day windows are conventional smoothing choices rather than fitted parameters. Raw daily values are extremely noisy, dominated by individual large transfers, and unreadable without smoothing. The two windows are shown together rather than one alone so that the reader can see both the trend and its current deviation. The two cohorts partition all coins between them: every coin is either above or below the holding-period boundary. Their sum is therefore the total loss volume, which is why no third total line is drawn.

Common mistakes to avoid

• Reading transferred volume as net selling. A coin moving between wallets under the same control counts here, and no economic transaction needs to have occurred. • Expecting both cohorts on the plot. Only the long-term cohort is drawn; the recent one moves far more volume and would flatten it. • Treating a peak as a bottom. Elevated loss volume has accompanied lasting lows and has also occurred in the middle of extended declines that continued afterward. • Converting mentally back into dollars. Doing so reintroduces exactly the distortion the unit denomination was chosen to remove. • Expecting the level to be comparable with unit-denominated measures published elsewhere. Cohort boundaries and volume definitions vary between conventions, and levels are only comparable within a consistent definition. • Reading a low level as calm. Low loss volume can mean holders are comfortable, or simply that few coins are moving at all, and the two are different situations.

Programmatic access

REST API

curl -sS \
  'https://api.trinityinsights.io/api/v1/cycle-intelligence/metrics/cycle-loss-volume-btc/data?timeframe=90d' \
  -H 'X-API-Key: $TRINITY_API_KEY'

MCP server

{
  "tool": "get_chart_value",
  "metric_id": "cycle-loss-volume-btc",
  "timeframe": "1y"
}

Required tier: pro. See the pricing grid for the tier list and the MCP documentation for multi-client configuration.

Related metrics

• cycle-cost-basis-regime-map
• cycle-upper-half-overhang
• realized-loss-total
• sopr-lth
• sopr-sth

Institutional disclaimer

Trinity Insights is an educational and analytical tool. The metric above does not constitute investment advice. Trinity Insights is not a Crypto-Asset Service Provider (CASP) registered under MiCA Regulation (EU) 2023/1114. See the full disclaimer.