Bitcoinโs price story lately has been told like it only has one main character, the ETFs.
Money goes in, price goes up, money goes out, price goes down. Itโs a clean narrative, and itโs not wrong, but itโs incomplete, because Bitcoin is not just a ticker. The network has its own internal plumbing, and some of the best clues about where we are in the cycle are sitting in plain sight on-chain.
The charts Iโve been watching feel a bit like checking the pulse under the headline. Miners, long-term holders, and the broad mass of wallets donโt react the way ETFs do, they donโt flip direction on a whim, they grind, they hold, then they crack, then they recover.
That is why I decided to check in on a few cycle gauges that have kept me honest across the years, miner reserves, NUPL, and the percentage of UTXOs in profit.
Bitcoin miner reserves dwindle
We’ll start with miners, because miners are where the Bitcoin โreal economyโ meets the fiat world. They have bills to pay, they are constantly converting electricity into BTC, and when the math stops working they donโt get to be philosophical about it, they sell, they shut down, they restructure, they move, they hedge, they survive.
In the data here, miner reserves are sliding to levels we havenโt seen since the early era. Miners currently hold about 1.801 million BTC.
Over the last 60 days, theyโve shed roughly 6,300 BTC, just over 100 BTC per day on average. That is a steady leak, the kind you see when the business is under pressure, and the treasury becomes working capital.
In dollar terms, the picture gets more dramatic. Miner reserves in USD are around $133 billion, down a little over 20 percent in about two months. Some of that is price, some of that is coins leaving miner wallets, and the combination is what matters, because it tightens their margin of safety.
If BTC is falling while reserves are thinning, miners have less cushion to ride out volatility, and the market has one more source of potential supply if things get ugly.
This is where the ETF narrative collides with the on-chain narrative. The ETF tape can be brutal, and it can overwhelm everything else in the short run.
Looking at recent flow data, the net move over roughly the last 10 trading days shown is about negative $1.7 billion, around $170 million per day on average. That number matters because itโs big enough to dominate marginal demand, and itโs fast enough to change sentiment before most people even register the shift.
But the problem with staring at flows alone is that it tells you whatโs happening at the surface, not whatโs building underneath.
Net unrealized profit and loss chart
If Iโm trying to work out where we are in the cycle, I want to know whether the market is in a normal downturn that can snap back, or whether itโs approaching a deeper reset that needs a real washout.
Thatโs why I keep one eye on NUPL, net unrealized profit and loss. Itโs not perfect, nothing is, but it does a good job of showing when the market as a whole is sitting on profit, sitting on pain, or sitting somewhere in between.
In the latest data, NUPL is still positive, around 0.215, which keeps Bitcoin in the green zone. It has fallen hard over the last couple of months, down about 0.17, and that slope is the part that grabs me, because you can feel the mood shifting in that compression.
The line in the sand for me is when NUPL goes below zero, and especially if it pushes toward negative 0.2.
NUPL last dipped below zero in early 2023, and the last time it was below negative 0.2 was late 2022. Thatโs the territory where true capitulation lives, thatโs where the โbear bottom confirmationโ argument has usually been strongest.
We arenโt there right now, and that matters if youโre trying to call a bottom today. It doesnโt mean we cannot be close, it does mean we donโt yet have the kind of confirmation that usually comes with a classic cycle low.
How many trades are in profit right now?
Then thereโs the UTXOs in profit chart, and this one is quietly fascinating because it shows how the market has matured over time. At the bottoms in earlier cycles, almost nobody was in profit.
In 2011 the trough was around 8 percent, in 2015 around 15 percent, in 2018 around 49 percent. The COVID crash in 2020 is a weird outlier that I tend to treat as its own event.
In 2023, the trough was about 60 percent. In the current data, 2026 has already printed a low around 58 percent, and the latest reading is around 71 percent.
That pattern, those rising floor levels, tells a human story. Bitcoin has more long-term conviction than it used to, more holders with low cost basis, more people who have sat through enough cycles to know the game, and that changes how deep the pain can go before the market finds a buyer.
It also changes how fast a bottom can form, because you donโt need to wipe out as much profit to push a large cohort into discomfort.
Thatโs where the main question comes from, and itโs the question I think this whole story should revolve around.
If UTXOs in profit have already touched levels that look like prior bear lows, are we closer to the bottom than people think, even though the cycle is โtoo earlyโ by the usual four-year script.
The marketโs stress test is happening in public
If youโve ever watched miners during a real drawdown, you know the vibe. Itโs less about charts, more about logistics. Machines donโt care about your thesis, your power contract doesnโt care about your timeline, interest payments donโt care about narratives.
When the price slides and the network keeps moving, miners are the first group that has to make hard decisions.
Thatโs why miner reserves hitting extreme lows, at least in the long-term view, is psychologically important. Itโs a sign that miners have already been drawing down inventory over a long period, and itโs a reminder that the industry has matured into something that behaves like a real sector with real balance sheets.
If the reserve base is already thinned out, and profitability keeps getting squeezed, you can get moments where miner selling stops being discretionary and starts being forced.
There are also signs in the wider mining data that stress is real.
Big difficulty adjustments and hashrate drops tend to show up when the economics are tight, and when disruptions, weather, or marginal operators create a sudden shift in the networkโs rhythm.
We have just had one of the largest difficulty adjustments in history, tied to hashrate declines and operational disruptions, which fits the broader theme of pressure building in the mining sector.
This is why Iโm wary of treating the current selloff as purely an ETF story. ETF flows are powerful, and right now they are pointing the wrong way. But miners and on-chain holder behavior are the parts that determine whether a dip stays a dip, or whether it becomes something that leaves a mark.
I also think itโs worth putting the numbers in the same frame, because scale helps. Miner reserves fell by roughly 6,300 BTC over 60 days. At rough spot levels, thatโs hundreds of millions of dollars worth of net coins leaving miner wallets.
That sounds huge until you compare it to ETF flow regimes, where the market can see net moves in the billions in a couple of weeks. The ETF tape can swallow miner supply in a way retail used to struggle to do.
The more interesting point is how these forces interact.
When ETF flows are negative, and price is sliding, miners get squeezed, and miner reserves drift lower.
That can create feedback, because weaker price tightens mining margins, tighter margins increase the odds of treasury drawdown, and treasury drawdown adds supply into already weak conditions. That doesnโt guarantee a crash but it raises the probability of one if the trend continues long enough.
Where NUPL and UTXOs in profit start to disagree, the story gets good
If all the indicators lined up neatly, there wouldnโt be much to write. The whole reason this moment matters is because the signals are mixed in a way that forces you to think.
NUPL is still positive. Thatโs a restraint. Itโs telling you the market has not entered the kind of widespread underwater pain that typically defines the deepest bear lows.
You can still argue we are in a reset, and you can still argue the cycle is intact, but the indicator hasnโt crossed the threshold that historically screams โcapitulation confirmed.โ
UTXOs in profit are telling a different story, or at least a different timing story. We have already seen readings that match the 2023 trough levels. Thatโs early, if you take the four-year cycle literally.
It suggests the market has already front-loaded a lot of damage, and if enough holders are already near the edge of โnot feeling rich anymore,โ it doesnโt take much more selling to tip sentiment into full exhaustion.
This is where I think journalists often miss the human element.
A bottom is not a single candle. A social process where the last group of people who were certain they were right finally stops checking the price is where the true bottom lives.
Itโs when the market stops caring about narratives because itโs too tired to argue. Indicators like UTXOs in profit are a proxy for that fatigue, and the fact that the floor keeps rising cycle to cycle is basically a story about a market that has developed scar tissue.
So could the bottom be close? Yes, it could.
But the โcouldโ is doing a lot of work, and this is why I keep the NUPL threshold in mind, because itโs the difference between a sharp washout that resets the board, and a slower grind that keeps punishing impatience.
Three forward paths, and what would confirm each one
The first path is the one most people will hate, a choppy, frustrating range where ETF outflows slow down, miners stop bleeding reserves at the current pace, and NUPL stabilizes somewhere in the 0.15 to 0.30 area.
The market doesnโt collapse, it doesnโt rip, it just wears people down.
This is the scenario where the cycle holds without giving you the clean catharsis everyone wants.
The second path is the classic capitulation, ETF outflows stay heavy, price continues to slip, NUPL breaks below zero, and miners accelerate distribution because the economics force it.
If NUPL pushes toward negative 0.2, that would fit the historical playbook for a deeper bear confirmation, and it would likely come with the kind of volatility that makes everyone swear theyโre done with Bitcoin for good, right before it turns.
The third path is the early bottom thesis, the one implied by UTXOs in profit touching prior cycle floor levels sooner than expected.
In that scenario, ETFs flip from outflows to a sequence of inflow days, NUPL stays positive and starts rising again, and miner reserves stop draining. That would say the market took its pain fast, and it found buyers before the full psychological reset.
The tension between these paths is where we need to focus. People are trying to explain price in real time with one metric, and the chain is showing you that the system is more layered than that.
Macro is the backdrop, and it always sneaks back into the plot
The other thing I donโt want to ignore is macro, because macro is why the ETF narrative exists in the first place.
When institutions are involved, they bring their own rhythm, and that rhythm is tied to rates, liquidity, and risk appetite.
The Fedโs projections, and the marketโs expectations around policy, matter because they shape the environment where big allocators decide whether they want exposure, and how much, and when.
This is also why I think the best framing is not โETFs versus on-chain.โ ETFs are part of the ecosystem now, and they can set the pace in the short run.
On-chain data is where you look for the deeper cycle clues, and where you look for stress that can turn a routine downturn into a structural event.
If I had to sum up what the data is telling me, itโs that the market is closer to exhaustion than it looks if you only stare at flows, but we donโt have full capitulation confirmation yet.
Miners have been bleeding reserves, the USD value of those reserves has dropped sharply, NUPL is compressing but still positive, and UTXOs in profit are already flirting with levels that have marked prior bear lows.
That combination makes this moment worth paying attention to, because it suggests the cycle theory can still hold, while the timing can still surprise you.
The chain is giving us enough evidence to take the โbottom could be closer than expectedโ idea seriously, and enough restraint to avoid declaring victory too early.
We need to look at the market from the perspective of the groups who cannot pause the game, miners who keep running machines, holders who keep weighing conviction against fear, and institutions who follow policy signals and flow models. All of them are pulling on the same price from different directions.
The next big moment will come when the pressure on-chain either breaks, or releases, not after a headline about flows.
