I keep two horizontal lines on my Bitcoin chart that are not drawn from any candle. They come from the chain itself, and every few months I watch price walk right up to one of them, stall, and either bounce or break with a conviction that a hand-drawn trendline never earns. They are the average cost basis of long-term holders and the average cost basis of short-term holders, and once you understand where they come from, they are hard to unsee.
The whole thing hangs on one arbitrary-looking number. On-chain analysts split every coin into two buckets based on how long it has sat unmoved: coins that last changed hands within roughly the last 155 days are short-term holder supply, and everything older is long-term holder supply. The 155-day threshold is not magic. It is roughly the point where, historically, the probability of a coin being spent drops sharply. Coins that survive past that age tend to keep sitting still. So the split is a rough proxy for two different populations of owners, the recent tourists and the people who have already decided to sit through whatever comes next.
What the two levels actually measure
Cost basis here means realized price, which is the average price at which the coins in that bucket last moved on-chain. It is not what anyone paid on an exchange with leverage. It is the price stamped on the last time each coin changed wallets, weighted by how many coins moved at each price. When you take that average across only the short-term holder bucket, you get the STH cost basis. Do it across the long-term bucket and you get the LTH cost basis.
Those two numbers behave completely differently, and that is the entire point. Short-term holder cost basis moves fast, because the bucket is constantly being refilled with coins that just transacted at recent prices. In a rally it climbs as new buyers step in near the top. In a slow grind it drifts along just under spot. Long-term holder cost basis moves slowly and usually sits well below spot during a bull phase, because it is anchored by coins that were acquired cheaply and have not moved since. In a deep bear it can actually sit above spot, which is a strange and useful signal I will come back to.
Why they act like support and resistance
The mechanism is behavioral, not mystical. STH cost basis works as support because of loss aversion in recent buyers. When spot is above the STH cost basis, the average short-term holder is in profit, and a pullback toward that level lands them near break-even. Historically that is where dip-buying shows up and where sell pressure from recent buyers has not yet turned into panic. When spot slices below STH cost basis, the average recent buyer is now underwater, and that same level flips to resistance. Rallies back up to it run into people who just want to get out flat. You can watch the level change character the moment price crosses it.
LTH cost basis works on a longer clock. Historically it has marked the neighborhood of cycle bottoms. When spot trades down to or below the long-term holder cost basis, it means the patient money, as a group, is underwater, and that has been a rare and usually short-lived condition. It does not tell you the exact bottom tick. It tells you that you are in the part of the range where the strongest hands are being tested, which historically has not lasted long before mean reversion.
Trading the crossovers, not the levels
The levels alone are just lines. The signal is in how spot, STH cost basis, and LTH cost basis are arranged relative to each other, and in the moments they cross. A few configurations I actually watch for:
- Spot above STH above LTH. The normal bull arrangement. Recent buyers are in profit, patient buyers are deep in profit. STH cost basis is your first line of defense on a pullback. Losing it is a caution, not a catastrophe.
- Spot dips below STH but holds above LTH. This is the classic bull-market correction. Recent buyers are underwater, older holders are fine, and the LTH level below you is the deeper floor. A lot of the scariest-feeling corrections live entirely in this band.
- STH cost basis crosses below LTH cost basis. This inversion is worth flagging. It tends to happen after a sustained decline where recent buyers, on average, now hold coins cheaper than the long-term cohort did. Historically this has clustered near late-stage bear conditions rather than tops.
- Spot below both, spot below LTH. Rare, uncomfortable, and historically the vicinity of capitulation. Not a level to short into.
My rule of thumb is simple. I treat a reclaim of STH cost basis after a period below it as a meaningfully more reliable bullish tell than the same-size move that happens up in open air with no level nearby. And I treat a clean rejection at STH cost basis from below, on the second or third attempt, as confirmation that the level has flipped to resistance and the recent-buyer cohort is still trying to exit.
How I plot them and the ways this goes wrong
You do not compute these yourself. Any competent on-chain data provider publishes STH and LTH realized price as time series. I plot both as lines directly on the price chart, log scale, so I can see spot crossing them in real time rather than eyeballing three separate panels. On Blockcircle I keep them pinned as overlays alongside spot so the crossovers are visible at a glance instead of something I have to reconstruct after the fact.
The failure modes are worth stating plainly, because these levels get oversold. First, they are averages, so they say nothing about distribution. A cluster of supply acquired at one specific price can matter more than the mean, and the average can sit in a zone where very few coins actually changed hands. Second, they lag. The 155-day boundary means the buckets update on a delay, so around sharp regime changes the lines can be describing a cohort that has already shifted. Third, and this is the big one, they are context, not a trigger. I have watched people short a rejection at STH cost basis into a market that was clearly re-accumulating and get run over, because they treated a probabilistic level as a hard wall. It is not a wall. It is a place where the behavior of a known group of holders becomes predictable enough to lean on, and only slightly.
The practical version of all this fits on an index card. Plot STH and LTH realized price over spot. Note which of the three is on top. Watch what happens the moment spot crosses one of the holder levels, and give more weight to reclaims and rejections that happen at a level than to the same move in empty space. That is most of the value, and it costs you two lines.