Crypto

Bitcoin May Not Have Bottomed Yet. Here’s Where the Data Says it Could.


This report was sent to Galaxy clients and counterparties on June 11. Trade or invest with Galaxy if you want to receive our research in the most timely fashion.

Introduction

During bitcoin’s 17-year existence, the price has moved in long waves. Roughly every four years, it climbed to a euphoric high, fell through a painful decline to a low, and then began to recover. This rhythm has historically been anchored around the quadrennial halving, the event that cuts regular supply issuance by half. Despite the declining impact of successive halvings, and many prognostications of a “supercycle,” the empirical data yet again shows evidence that the 4-year cycle is intact. This report is about those waves and the emergence of a pattern that has held across bitcoin’s modern history: each wave has been gentler than the last.

The October 2025 peak was the calmest top bitcoin has ever made, and the decline since has been unusually mild. If the high was so tame, should we expect the eventual cycle low to be unusually shallow too? And, if so, roughly where might that bottom land?

This report assumes that the current drawdown’s bottom is not in, and we present data to support that assumption. The data also suggests that the calmer top in October 2025 is likely to result in a shallower bottom. And the historical analogies suggest a base case bottom for the current drawdown between $40k-46k occurring sometime between now and Q4 2026. (Base case scenario is provided for illustrative purposes only. Actual results may differ materially.)

Crucially, this report exclusively relies on market and onchain data and timing analysis. The projected cycle bottom ranges do utilize or rely upon assessments of the likelihood, timing, or impact of exogenous events, such as regulatory, market, or geopolitical developments.

4-Year Cycles at a Glance

Each of bitcoin’s cycles has run from a prior low, through a halving, up to a top, and back down to the next low. Here are the four cycles, including the one underway:

The current cycle’s bottom has not formed; its drawdown and elapsed time are “so far” figures based on a report date of June 9, 2026. Notice the two patterns that this report builds on: each cycle’s peak-to-trough fall has shrunk (from 85% to 84% to 77%), and each top has historically been followed by a bottom about 12-13 months later. The current cycle is only eight months past the most recent top.

Because October 2025’s top was so subdued relative to prior cycle tops on an indexed basis, the average price the market paid for its coins (the realized price, effectively the “cost basis”) sits unusually close to the all-time high: about 43.7% of the prior ATH, versus roughly a third or less in past cycles. This is a crucial data point: the same kind of sell-off that ended past bear markets would, this time, end at a much higher dollar price. Comparing cycle timing, amplitude, and onchain indicators suggests that the current drawdown could bottom within these levels:

Bitcoin Scenario Drawdown Table

The levels above and the analysis in this report imply our view that this cycle’s bottom has not yet been found. Very few of our historical cycle bottom indicators have lit up, the decline is still younger than historical drawdowns on a time basis, and the cost basis itself can fall if a real panic emerges. Our thesis is that the 4-year cycle remains real on an empirical basis, but that the amplitude of the cycles has compressed. A calmer top has raised the floor, but it has not removed it.

bitcoin report headline numbers

Scoping the Analysis

It’s extremely difficult if not impossible to spot a top or a bottom while it is happening, but both are obvious in hindsight. We thus assess how many of the conditions that showed up at every past top and bottom are present at once. To establish a set of indicators by which to assess prior tops and bottoms, we look across five families of evidence: valuation (is the price high or low relative to what holders paid?), profit-taking (are holders selling into strength or capitulating into weakness?), miners (are the participants who produce bitcoin flush with profit or under stress?), trend (how far is the price above or below its long-run averages?), and sentiment (greed or fear?).

Applying that five-dimensional lens to both ends of the current cycle yields a clear picture: bitcoin’s swings are getting smaller. Each top has been less euphoric than the one before and the crashes that followed have grown shallower. If that “compression” in amplitude is real and works on both sides, it may provide information about the expected cycle low forthcoming in the current drawdown. We can estimate, within a range, where bitcoin may bottom in the current drawdown.

Two Ends of This Cycle

This analysis requires us to identify indicators and establish a baseline to identify cycle tops and bottoms. We score each end the same way: against the levels that every past top and every past bottom reached.

Mapping the Cycle Top

The top was real, but the quietest on record. At the October high, only two of 11 classic warning signs reached even the mildest prior-top level, and both only barely. The clearest valuation gauge, market value to realized value, or MVRV (how high the price is relative to the average price holders paid), peaked at just 2.29, versus 2.93 to 5.91 at the three previous tops. The whole “greed” cluster of indicators printed its lowest top-of-cycle readings on record, and the Pi Cycle Top, a timing signal that nailed the last three peaks within days, never lit up at all (a first in bitcoin’s history). Yet the timing was textbook: the top landed 1,062 days after the previous low, right about where the 2017 and 2021 peaks arrived. The twist is that the real euphoria came about 18 months early, around the launch of U.S. spot bitcoin ETFs, after which the price kept grinding higher even as enthusiasm faded. In retrospect, this looks like institutional buying rather than frenzied retail buying that leads to a blow-off.

bitcoin top euphoria decay

The graphic below maps our full topping indicator table for the current cycle (anchored around the October 2025 all-time high).

bitcoin topping scorecard table

Of the 11 magnitude signals: two were confirmed, two were only partial (clearing at least 85% of the bar), and seven did not light up. Of the two that did confirm (RSI and SOPR), each only marginally cleared its weakest-ever 2021 bar, and peaked back in 2023 and 2024, respectively, not at the October 2025 price high. Crucially, the Pi Cycle Top did not light up even though the cycle clock hit on schedule (these two metrics are treated separately because timing is a calendar fact, not a measure of how hot the top was). “What marked prior tops” is the range across the 2013, 2017, and 2021 cycle tops; the threshold is the least euphoric of the three (the 2021 peak), i.e. the easiest top-grade bar to clear. “Cycle peak” is each metric’s most extreme reading of the current cycle, with the month it occurred. Reserve Risk and the Pi Cycle ratio are on our internal scale.

Mapping the Cycle Bottom

Only four of 13 bottoming signs have hit during this drawdown, and three are among the softer indicators (fear; a trend gauge grazing its bottom range; and the first dip below the 200-week moving average). The fourth, which flipped in early June, is the first miner-side tell: the Hash Ribbons recovery cross, where the 30-day average of mining power has climbed back above the 60-day average after a stretch of capitulation, a signal that has historically preceded bottoms.

The strongest signs that have marked every true bottom (the price falling below the cost basis, holders sitting on aggregate losses, sustained loss-taking, a deep capitulation flush) have not happened. At −51% the decline is still far milder than the −77% to −85% lows that ended every past cycle, and milder than the −53% mid-2021 dip.

But the pace has shifted. Measured at the same point in the cycle (about eight months, or 242 days, after the peak), the recent slide has pushed the current fall just below where the 2013–15 cycle stood at the same stage (−48%, during a relief rally), so it is no longer the shallowest path on the board (which it was for most of this drawdown). The 2017–18 and 2021–22 cycles were far deeper (both near −68%) at this point. By the cycle clock, the window for a bear-market low does not open until roughly late 2026.

bitcoin drawdown from ATH + 400 days

Each line tracks one cycle’s fall from its peak, lined up at day zero. At ~242 days in (dashed line), the current cycle (orange, −51%) has slipped just below the 2013–15 cycle (−48%). That makes this cycle no longer the shallowest (which it was for most of the drawdown). The other two prior cycles were near −68% at this stage. All remain far above today’s level (green band = where past bear markets bottomed).

The graphic below maps our full bottom indicator scorecard for the current drawdown, using indicators that previously indicated a cycle bottom.

bitcoin bottom indicator scorecard

Of 13 target metrics, four have been reached, two are approaching, and seven have not yet been reached.

To demonstrate the indicative nature of this set of bottom indicators, the table below shows when they lit up during prior cycle bottoms compared to today. The same 13 signals tell a sharp story when lined up against the three prior cycles. At every past bear-market low, all 13 eventually reached their bottoming zone, the only difference being timing: some light up early and some lag. Today, only four have hit, and the one miner-side metric among them (Hash Ribbons) is the most recent. (A notable difference is that in this drawdown, the Hash Ribbon flip appears to have preceded a bottom rather than lagged it, as it did in prior cycles. This could be due to an externality relating to bitcoin miners transitioning to artificial intelligence, a phenomenon not seen during prior cycles).

bitcoin bottom indicator cycle comparison

Prior-cycle cells show how many days each indicator’s most bottom-like extreme led (-) or lagged (+) that cycle’s price low (within a 180-day window). Hash Ribbons = the recovery cross; cycle clock = the 12-month-after-top mark. Every indicator hit at all three prior bottoms, with the signal being which lit up early versus late. This cycle’s low does not appear to be in, so the current column shows only whether each box has been ticked since the October 2025 price high.

The Swings are Shrinking on Both Sides

Before drawing any conclusion, here is the plain fact the rest of the report rests on: bitcoin’s swings have narrowed at both ends. The tops have cooled every cycle (MVRV of 5.91, then 4.72, then 2.93, then 2.29) and the bottoms that followed have risen every cycle, from an MVRV of 0.56 in 2015 to 0.69 in 2018 to 0.75 in 2022. In other words, the distance between the most over-valued and the most under-valued points in each cycle keeps shrinking. The crashes tell the same story in price: −85%, then −84%, then −77%, and −51% so far this time.

bitcoin cycle amplitude compression

The price-to-cost-basis ratio (MVRV) at each top (red) and the bottom that followed (blue), closing in on “fair value” (1.0) from both directions. The data suggests that the current cycle has likely not yet bottomed (the hollow diamond is the deepest reading *so far*). This describes the cycle patterns but is not a guarantee of where this cycle bottoms.

Tops cooling while bottoms rise is a description of three completed cycles, not a law of nature. It does not, by itself, prove the next low will be shallow. But it does allow us to ask a precise question with an exact answer: if a bottom behaves like past bottoms, how much of the dollar fall is actually decided by how hot the top was?

A Calm Top Raises the Floor

MVRV is just today’s price divided by the onchain cost basis. If you flip it around, the cost basis is the all-time high divided by the MVRV at the top. So, a lower MVRV at the top means the cost basis sits closer to the peak.

Because October’s top was the calmest ever (MVRV 2.29), the cost basis ended up at 43.7% of the all-time high (compared to 34.2%, 21.2%, and 16.9% at the 2021, 2017, and 2013 tops, respectively). A calm top does not lower the floor; a calm top raises the floor by leaving the cost basis much closer to the peak (all else equal).

muted bitcoin top raises the floor

The cost basis as a share of each cycle’s all-time high. It has climbed every cycle, to 44% in 2025, because each top has been calmer. The note on each bar shows what a typical past-style bottom would have meant for that cycle’s dollar fall.

Now hold the behavior of the bottom fixed (assume every cycle bottoms at the same MVRV) and the dollar fall shrinks every cycle, purely because the cost basis started higher. The table makes the point without any forecast in it:

bitcoin cycle map table

Each cell is the fall from that cycle’s peak if it bottomed at the column’s MVRV, using that cycle’s own cost-basis-to-peak ratio. The bottom behaves identically across the rows; only the calmness of the top changes. A typical past-style bottom (MVRV 0.70) meant a −88% fall in 2013 but only −69% this cycle. This isolates the top’s effect; it is arithmetic, not a claim that a calm top causes a higher bottom.

How Far Below the Anchors Do Bottoms Reach?

Bottoms are not located by a round-number percentage but relative to two key anchors: the cost basis and the 200-week moving average (200w MA), which has served as long-term price support for bitcoin for its entire life. Measured by those anchors, the three last bear-market lows landed well below both: on average about −33% beneath the cost basis (as deep as −44% in 2015), and about −14% beneath the four-year average.

Two things jump out. First, the gap below the cost basis has narrowed every cycle (−44%, −31%, −25%), the same compression we saw at the tops. Second, today’s price has not reached that zone at all. Even after a 51% fall, bitcoin’s price is still 14% above the cost basis (it has never dropped below cost basis this cycle) and only 1.5% under the four-year average. By the yardstick that has located every past bottom, this one has not arrived.

bottoms form below cost basis

How far each past bear-market low sat below the cost basis (blue) and the four-year average (purple). Past lows ran well under both. Today’s price is still above the cost basis and only just below the 200w MA, and the gap below the cost basis has shrunk every cycle.

The anchors agree with the arithmetic. If we translate those past gaps onto today’s anchors, they point to the same zone: a −25% to −44% drop below the cost basis works out to roughly $30k–$40k, and the four-year-average gaps span about $41k–$62k. This suggests that a real bottom is likely below today’s price, but far above the old “down 75%–85%” levels.

Where the Floor Sits This Time

Turning the arithmetic into prices, off today’s cost basis of $53k, gives a set of scenarios rather than a single number; we lead with the middle one.

Our base case, in which the bottom simply continues the cycle-by-cycle rise toward fair value (an MVRV of 0.75–0.86), sits near $40k–$46k. A harsher washout to a 2018/2022-style level (MVRV 0.56–0.70) lands near $30k–$37k. A shallower outcome, where steady buying absorbs the decline near the cost basis (MVRV 0.95–1.01), is around $51k–$54k; and merely touching the rising four-year average ($62k) would be only about −51%. (Provided for illustrative purposes only. Actual results may differ materially.)

bitcoin drawdown scenario ladder

The scenarios drawn against price. The cost basis and the rising four-year average (the lines a bottom historically tracks) sit far above the old “down 75–85%” band (gray, retired). The colored bands translate past-style bottoms into today’s dollars. These are levels conditional on a bottom forming, not a prediction that one is near. Provided for illustrative purposes only. Actual results may differ materially.

The punchline is what this does to the old rule of thumb. A −77% to −85% fall (the right yardstick in past cycles) would put this bottom at $19k–$29k. But that rule effectively double-counts the calm top by applying a deep, peak-relative percentage to a peak that was already close to the cost basis because the top was calm.

The cost basis is the rising tide under this whole picture, and it is also the clearest sign that the “floor” can move. Over the past year, the cost basis climbed from about $47k to a peak near $56k late in 2025 (a 20% rise) as this cycle’s higher-priced buyers reset the average. That climb is the deepest reason a bottom now sits so far above the old rule. But realized price has since slipped about 5% to ~$53k as the decline drags some 2024–25 coins through at a loss. Into the late-2026 window realized price / cost basis is the swing factor: a calm, orderly decline lets it firm up and holds the base case in the mid-$40,000s; a genuine panic pushes it lower and drags the whole picture down with it.

The Catch: The Floor Can Move

There is reflexivity to the cost basis. Cost basis can look like a floor, but it is built from prices coins last traded at. In a real sell-off, coins change hands at a loss and pull that average down, so the floor chases the price lower instead of holding it.

This is the most important limit on the higher-floor case. The cushion is thin: the price is only about 14% above the cost basis today (an MVRV of 1.14), and it has never dropped below it this cycle. If a sell-off drags the cost basis down 10%, 20%, or 30%, a typical-style bottom could fall from about $40k to roughly $36k, $32k or $28k, putting it right back into the normal historical range.

reflexivity of the bitcoin floor

Holding the style of the bottom fixed and letting the cost basis fall during a sell-off. The implied floor slides from about $40k back toward $28k, re-entering the normal historical band (amber). A calm top raised the floor; a genuine panic can claw part of that gain back.

A steady, price-insensitive bid from spot ETFs and corporate treasuries is something past cycles did not have, and it leans toward a higher floor, but it can amplify a decline as easily as cushion it. The nature of their access to capital means that digital asset treasury companies (DATs) and corporate treasuries tend to buy into strength, not catch falling knives, and ETF flows have lately been net outflows in 2026. In real deep sell-off, fund redemptions could force selling rather than absorb it. The 2022 cycle, which saw the biggest forced-selling washout in crypto history, still only fell −77%. So, “less leverage this time” is not necessarily something to lean on. (These are supporting points, not the backbone of the argument.)

The higher floor and its risk of erosion in a panic are two halves of the same mechanism: the cost basis starts higher this cycle, and it can fall if real capitulation arrives. This is why we focus on ranges rather than a single output.

Combining the Analysis: Where the Numbers Say the Drawdown Will Go

The conclusion of our analysis cleanly suggests both how far a drawdown will go and how long it will take.

The calmer top has lifted the cost basis to 43.7% of the all-time high, so the dollar fall for any given style of bottom is mechanically milder than in any past cycle. We believe that the “bitcoin has historically fallen 75%–85%, so this cycle bottoms near $19k–$29k” rule is now obsolete as a literal price floor. Even a very harsh washout equivalent to prior similar washouts would now imply a much higher number. Hence, even our harsher washout is higher than that area and our base is in the mid-$40k region.

When comparing to data on prior cycle indicators and timing, it’s likely the bottom is not in. Only four of 13 bottom indicators have lit up, and the current drawdown is currently only about eight months old compared to the historical 12-13 months to bottom (and the cost basis itself can fall).

headline numbers

If the deep signals (price below the cost basis, holders in aggregate loss, sustained loss-taking, a four-year-average break that holds, a bear-depth fall) begin to flip at prices well above the old band, the compression was real on both sides. If, instead, a full capitulation arrives on the old schedule, the calm top will simply have delayed the pain rather than reduced it. Either way, the cost-basis arithmetic says the starting line for that judgment is far higher than the old four-year rule assumes.

This is a descriptive study of how a calm cycle top shapes the arithmetic of a cycle bottom, not necessarily a forecast of price direction or price target. Our levels apply historical data to analogize the current drawdown relative to today’s cost basis (which itself can move).

We include numerous supporting charts, grouped by theme. The first set frames the cycles; the second walks through the full bottoming checklist, one indicator at a time. In every indicator chart, the shaded band is the range that metric reached at the 2015, 2018, and 2022 lows, and the orange marker is the latest reading.

The Cycle in Pictures

bitcoin price tops

Price and its cycle tops. Bitcoin’s whole price history on a logarithmic scale, with the three past cycle tops (red) and the October 2025 high (orange).

apx price bottoms

Price and its cycle bottoms. The same history, marking the reference lows: the 2015, 2018 and 2022 bear bottoms (red) and the COVID and mid-2021 dips (gray).

apx cycle clock

The cycle clock. How many days after the prior low (circles) and the halving (squares) each top arrived. October 2025 landed squarely in the historical window.

apx euphoria timing

Euphoria came early. The cycle’s valuation peak arrived in early 2024, around the spot ETF launch; the price then rose another ~70% to its October 2025 high as onchain enthusiasm faded.

apx pi cycle

The signal that never fired. The Pi Cycle Top flagged the 2013, 2017 and 2021 peaks within days (stars). This cycle the trigger was never reached (a first for any cycle top).

The Bottoming Checklist, Indicator by Indicator

apx mvrv

MVRV. Price versus the average holder’s cost basis. Past bottoms drove it well below 1.0; this cycle’s low so far stopped at 1.14.

apx nupl

NUPL. The share of market value sitting in unrealized profit. Past bottoms pushed it below zero (aggregate losses); it remains positive today.

apx mvrv z

MVRV Z-Score. A standardized version of MVRV. Past bottoms printed deeply negative readings; this cycle it is still positive.

apx mayer

Mayer Multiple. Price divided by its 200-day average. It dipped into the bottom range, the most bottom-like of the trend signals.

apx price 200w

Price vs the four-year average. The 200-week moving average is bitcoin’s most durable support. Past bottoms touched or broke it; the price has now slipped just below it for the first time this cycle.

apx sopr

SOPR. The average profit/loss on coins that move. Past bottoms pinned it below 1.0 (loss-taking) for months; this cycle, it has only grazed below.

apx realized pnl

Net realized profit/loss. Dollars of profit (+) or loss (−) locked in each day, scaled by market size. The deep loss-taking spikes that mark bottoms have not occurred.

apx puell

Puell Multiple. Miner-revenue stress. Past miner-capitulation bottoms printed 0.30–0.41; this cycle’s low (~0.44) came close but did not reach it.

apx hash ribbons

Hash Ribbons. Mining-power momentum. Below 1.0, miners are capitulating; and it has already durably fallen below that threshold in 2026.

apx fear greed

Fear & Greed. Our proprietary 0–100 sentiment gauge has spent this decline deeper in fear than at the average past bottom. This is the one indicator that has definitively hit.

Appendix B: Glossary of Terms

Bitcoin cycle. Bitcoin’s roughly four-year rhythm of a multi-year climb to an all-time high, a sharp decline to a low, and a long recovery. Each cycle is usually framed around the halving.

Halving. About every four years, the rate at which new bitcoin is created is cut in half. It is a fixed feature of the protocol and has historically acted as the anchor of each cycle.

All-time high (ATH). The highest daily closing price bitcoin has ever reached. This cycle’s ATH was $124,824 on Oct. 6, 2025.

Drawdown. How far the price has fallen from its peak, expressed as a percentage. A −50% drawdown means the price is down half its all-time high.

Cost basis, a.k.a. realized price. An estimate of the average price the market paid for the coins it holds. Technically it is the value of every coin at the price it last moved onchain, divided by the number of coins. It is the single most important anchor in this report; we also call it the network’s cost basis.

Market capitalization (market cap). The total dollar value of all bitcoin at the current price (price × coins in circulation).

Realized capitalization (realized cap). The total value of all bitcoin priced at the level each coin last moved, rather than today’s price. Realized price is realized cap divided by the coin count.

MVRV ratio. Market cap divided by realized cap: equivalently, today’s price divided by the network’s cost basis. Above 1.0, the average coin is in profit; below 1.0, the average coin is underwater. It is the spine of this report.

MVRV Z-Score. A standardized version of the gap between market cap and realized cap, which makes extreme highs and lows comparable across bitcoin’s very different price eras.

NUPL (Net Unrealized Profit/Loss). The share of the total market value that is unrealized profit. High positive readings signal greed near tops; readings below zero (aggregate paper losses) accompany capitulation near bottoms.

SOPR (Spent Output Profit Ratio). The average profit or loss on coins that move on a given day. Above 1.0, coins are being sold at a profit; pinned below 1.0, holders are realizing losses (a bottoming tell).

Mayer Multiple. Price divided by its 200-day moving average. A simple gauge of how stretched price is above or below its medium-term trend.

200-day/200-week moving average. The average closing price over the last 200 days (medium-term trend) or 200 weeks (about four years, bitcoin’s most durable long-term support line).

Puell Multiple. The dollar value of newly mined bitcoin versus its one-year average: a gauge of miner-revenue stress (low) or blow-off (high). Named after David Puell, an analyst at ARK Invest.

Reserve Risk. A measure of the confidence of long-term holders relative to the price. It is shown on a scale, used only in a relative sense in this report.

Pi Cycle Top. A timing indicator that fires when the 111-day moving average crosses above twice the 350-day moving average. It pinpointed the 2013, 2017 and 2021 tops within days; this cycle, it has never fired.

Hash Ribbons. Compares the 30-day and 60-day averages of mining power. When the shorter average falls below the longer one, the highest-cost miners are switching off (capitulation); the recovery cross has historically preceded bottoms.

Fear and Greed Index. A 0–100 sentiment gauge built from onchain, derivatives, and flow data. Low readings signal extreme fear (near bottoms); high readings indicate greed (near tops).

RSI (Relative Strength Index). A momentum oscillator from 0 to 100; high readings indicate an overbought market, often near tops.

Cycle clock. The number of days from a cycle’s starting low, or from its halving, to its top or bottom. Bitcoin’s last three tops arrived about 1,060 days after the prior low; bottoms happened about 12–13 months after the top.

Reflexivity. The idea, popularized by George Soros in his 1987 book “The Alchemy of Finance,” that a measure used as a yardstick is itself moved by price. Here, the cost basis looks like a floor, but in a genuine sell-off it falls as coins change hands at a loss. The floor is a moving target, not a fixed line.



Click to comment

Leave a Reply

Your email address will not be published. Required fields are marked *

Most Popular

To Top