Cost Basis Regime Map
Three acquisition prices (the coins moved one day to one week ago, the coins held under roughly 155 days, and the active market average) admit only six possible orderings. Each ordering describes a distinct structural state, from the one where marginal demand pays the most to the one where recent buyers sit above the market while new money enters below it. The background colour is that state; the three lines are what produce it, so every colour change lands exactly on a crossing. A five percent deadband keeps the state from flickering when two lines graze each other. No weights, no hidden parameters: the state is a ranking, and the spot price never takes part in it.
What is it?
Every bitcoin in circulation was bought at some price. Group the coins by how recently they last moved and you get an average price paid per group, which is what the industry calls a cost basis. This chart follows three of them. The first covers coins that moved between one day and one week ago: that is the marginal buyer, the money arriving right now. The second covers coins held for less than roughly 155 days, the conventional boundary for recent holders. The third is the average of the actively traded market, which excludes long-dormant coins so that lost or forgotten wealth does not drag the average down. Three numbers can only be ranked in six ways, and each ranking describes a different market. When the money arriving today pays more than everyone else, demand is leading. When recent buyers sit above the market average while new money enters below it, a large group is holding losses at the same moment the marginal buyer refuses their price. Those are structurally different situations, and the ranking distinguishes them without any weighting or judgement. The deliberate omission is the spot price. It plays no part in the ranking. Cost bases move slowly because they are averages over large groups of coins, so a ranking built from them changes only when something structural changes, not when the market has a volatile afternoon.
How to read
Three lines on a logarithmic price scale, plus the BTC price for reference, over a background band whose colour is the current state. Because the state is nothing more than the ranking of the three lines, every colour change lands exactly on a crossing. If you see a band change without a crossing, something is wrong with the chart, not with the market. The six states, from the strongest structure to the weakest: • Expansion: new money above recent holders, both above the market. The marginal buyer is paying the most. • Recovery: new money above the market, recent holders still below it. Fresh demand is returning while the recent cohort has not caught up. • Stalling: recent holders above new money, both still above the market. Recent buyers paid more than today's demand is willing to. • Basing: everything below the market, but new money paying more than recent holders. The purge is already advanced. • Exhaustion: everything below the market, new money at the bottom. Demand is entering at the cheapest point of the three. • Capitulation: recent holders above the market while new money sits below it. The widest gap between what a large group paid and what anyone will pay now. The heading reports the state, how many days it has lasted, and the three distances in percent, so the reading requires no hovering.
Key zones
The boundaries are the crossings themselves, and a five percent deadband guards each one. A state only changes when the smallest of the three distances clears that margin, which means two lines can graze each other without the background flickering. Without this filter the state changes roughly three times more often, and the chart becomes unreadable noise. Three crossings matter, and each answers a different question: • New money against the market average decides the side. Above it, the marginal buyer is validating current levels. Below it, the marginal buyer needs a discount. • Recent holders against the market average tells you whether the recent cohort is carrying gains or losses relative to the whole investor base. • New money against recent holders tells you whether today's demand is stronger or weaker than the demand of the past few months. The six states are not a ladder to be climbed in order. Markets skip states, revisit them, and can sit in one for a very long time.
What to observe
• The width of the band, not just its colour. A state entered by a hair and a state entered decisively are not the same situation, and the percentages in the heading tell them apart. • Which crossing produced the change. A state change driven by new money moving is a demand event. A state change driven by the market average drifting is a slower, structural event. • How long the state has lasted, which the companion duration chart puts in historical perspective. • Sequences rather than single states. Passing through the weakest states and back out is a different story from oscillating around a boundary for months. • Disagreements with price. The most informative moments are when price is calm and the ranking changes anyway, or the reverse: the ordering of what people paid can shift while the quoted price barely moves.
Historical context
Measured over the full available history, the six states are far from equally common. The strongest one occupies the largest share of time by a wide margin, which is what a long-term rising asset looks like from the inside. The weakest states are comparatively rare and comparatively brief, but they cluster around the periods that later get described as bottoms. The cost basis of coins that moved in the last week is the fastest of the three by construction, since it reprices completely every seven days. The market average is the slowest, because it carries the accumulated purchases of years. Recent holders sit between the two. This ordering of speeds is why the crossings happen in recognisable sequences rather than at random: the fast line crosses first, the slow line confirms later. The conventional 155-day boundary for recent holders is inherited from the observation that spending behaviour changes markedly around five months of holding. It is a convention, not a law, and its exact placement matters less than its consistency across the whole history.
Expert notes
The cost basis of the newest cohort is rebuilt from its raw components, the group's realised capitalisation divided by the group's supply, rather than read from any precomputed ratio. The classification is a pure ranking with no weights, no fitted parameters and no thresholds beyond the deadband, which means it cannot be overfitted to past cycles: there is nothing to fit. The deadband applies to the smallest of the three distances rather than to each one separately. This is deliberate. Near a triple crossing, where all three cost bases converge, several boundaries are in play at once and any per-boundary rule would let the state jump around while the market is genuinely undecided. A companion chart in this catalogue ranks the spot price inside a fan of cost bases. It answers a related but distinct question, and the two often disagree by design. This one asks who is paying the most right now. The other asks where the current price sits relative to what people paid. Neither supersedes the other, and comparing them is more informative than picking one.
Common mistakes to avoid
• Reading the states as a forecast. They describe a structure that exists now. Nothing here says what happens next, and the weakest state has both preceded lasting lows and persisted for months without resolving. • Assuming the order is a cycle to be walked through. States are skipped and revisited constantly. • Expecting the state to change when price moves sharply. It usually will not, and that is the design: a violent day barely moves an average taken over an entire cohort. • Confusing the newest cohort with retail. Coins that moved in the past week include exchange operations, custody transfers and rebalancing, not only new buyers. • Treating the market average as a floor or a target. It is the average price paid by the active market, which carries no obligation to be revisited. • Comparing state frequencies across different history windows. The share of time spent in each state depends heavily on the period measured, and the early years of the asset behaved differently from the later ones.
Programmatic access
REST API
curl -sS \
'https://api.trinityinsights.io/api/v1/cycle-intelligence/metrics/cycle-cost-basis-regime-map/data?timeframe=90d' \
-H 'X-API-Key: $TRINITY_API_KEY'MCP server
{
"tool": "get_chart_value",
"metric_id": "cycle-cost-basis-regime-map",
"timeframe": "1y"
}Required tier: pro. See the pricing grid for the tier list and the MCP documentation for multi-client configuration.
Related metrics
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.