> ## Content Index > Fetch the complete content index at: https://www.themarketsunplugged.com/llms.txt > Use this file to discover other available public pages before exploring further. # Bitcoin's four year cycle is a liquidity cycle wearing a halving costume. - URL: https://www.themarketsunplugged.com/bitcoins-four-year-cycle-is-a-liquidity-cycle-wearing-a-halving-costume/ - Published: 2026-09-18T16:09:28.000Z - Updated: 2026-09-18T16:45:33.000Z - Description: The interval is real and the halving is not what produces it. Four cycles measured low to low, what the last one did, and what I use in its place. - Author: Oz - Tags: oz-updates, macro-read The Markets Unplugged # Bitcoin's four year cycle is a liquidity cycle wearing a halving costume. Most arguments about the four year cycle are between people who think it is gospel and people who think it is dead. Both are arguing about the wrong thing. The interval is real. The halving is not what produces it. What the interval actually measures, why the cycle that just closed came in three months early, and what I use in its place. On this page[The separating measurement](#the-measurement-that-separates-the-two-e)[Honest limits](#the-honest-limits-on-that)[What a calendar can do](#what-a-calendar-can-and-cannot-do)[What I do instead](#what-i-do-instead)[A fixed horizon](#why-a-fixed-horizon-decided-in-advance)[The column everyone skips](#the-column-everyone-skips)[What the band tells you](#what-the-opening-band-actually-tells-you)[What this cannot do](#what-this-cannot-do)[A dated claim](#a-dated-claim-you-can-check) Start with the measurement, because almost nobody quotes it properly. Bitcoin has completed four cycle intervals, low to low. Measured in days rather than rounded to months, they run: 1,157 Days, first interval 1,430 Days, second 1,431 Days, third 1,338 Days, fourth The two middle intervals are one day apart. One day, across nearly eight years. That is a tighter result than almost anyone claiming the four year cycle actually cites, and it is why I have no patience for the argument that the cycle is dead. But look at the ones either side. The first interval ran 1,157 days, some nine months shorter. And the one that just completed, November 2022 to July 2026, ran 1,338 days. That is **92 days short** of the two that preceded it. So the pattern is not a clock. It is short, then locked for two cycles, then short again. ## The measurement that separates the two explanations Here is where it gets interesting, and it is the reason for the title. If the halving drives the cycle, the low should keep a stable distance from it. The halving schedule is known years ahead and it did not move. So measure the same lows against the halvings that followed them: 366 Days, low to Nov 2012 halving 547 Days, low to Jul 2016 517 Days, low to May 2020 517 Days, low to Apr 2024 That figure moves by 181 days across the four. Same lows, same chart, same method as the interval measurement. One of those two numbers behaved like a mechanism for two cycles and has now shifted. The other never held still at all. Neither behaves like something bolted to the block subsidy. And the cycle that just completed is the case that actually separates them. The halving schedule did not change. The low arrived three months early anyway. A supply driven story has nothing to offer there. A liquidity driven one does, because liquidity conditions are exactly the sort of thing that shifts a turn forward by a quarter. The rhythm people have spent a decade attributing to the halving looks far more like the rhythm of money getting cheaper and more expensive. The halving has simply been standing nearby when it happened. ## The honest limits on that Four intervals is four. I am not going to dress that up, and anyone presenting this as settled is overselling it. The 92 day figure also depends on July 2026 being the low. That is my read rather than a fact, it is a public call with a stated invalidation, and that window is open as I write this. If it breaks, the number changes and I will say so. What I would defend is narrower than the headline: the halving explanation is unproven rather than disproven, and the interval has now moved in a way that the halving cannot account for. That is enough to stop treating a halving countdown as a forecast. Which leaves the practical question. If the calendar is not bolted to anything, what do you actually use? ## What a calendar can and cannot do A cycle interval is a calendar. It tells you approximately when something is likely to happen. That is genuinely valuable, and it is also the only thing it does. It cannot tell you where you are standing right now, what the road between here and there looks like, or what you will have to sit through on the way. Four year cycle A date TMU framework Position now Past outcomes The drawdowns How far each one fell before it got there. That is the one that decides whether you hold. Everyone I have watched get hurt in this market got hurt in that gap. They knew the destination. They had no idea about the journey. So they bought, sat through a fall nobody had warned them about, decided the thesis was broken, sold, and watched the destination arrive on schedule without them. Being right about the date is worth nothing if you cannot survive the path. ## What I do instead Two questions, asked every week. ### Where does this sit right now, against its own history? Not against a target, not against a prediction. Against everything Bitcoin has ever done, is today cheap, ordinary, or stretched. Same question asked separately of Ethereum, and separately again of the broad altcoin market, because they are frequently in different places at the same time. ### And when it has sat here before, what happened over the following seven months? Not the average. All of them. The best one, the worst one, and critically, how far each of them fell before it got where it ended up. That produces something the calendar cannot: a position, and a range of paths from it. ## Why a fixed horizon, decided in advance Roughly seven months, and it has never changed. The length matters less than the fact that it is fixed. An open-ended horizon lets you stop measuring when the answer flatters you. Almost every impressive backtest you have ever seen contains a version of that choice, usually unintentionally. Fixing the window removes it. Whatever happened over that period is what gets recorded, good or bad, and I do not get a vote. It also means every comparison is like for like. Measuring one stretch for seven months and another for twelve tells you nothing, and it is the single easiest way to flatter a result without noticing you have done it. ## The column everyone skips When I show these comparisons, the number people read first is always the outcome. Up this much, up that much. The number that matters is the one underneath. Where each of the six ended Aug 2020 +131% Nov 2018 +42% Dec 2022 +42% Jul 2023 +26% Jan 2020 +18% May 2022 \-29% What they fell along the way, across the same six Typically 11% Deepest 31% Six windows. Descriptive, not predictive. Five of those six ended higher and one ended down 29%. But every one of them fell along the way, typically 11% and in the worst case 31%. Anyone sitting through that deepest one in real time would have been certain the thesis was broken. If you know in advance that the path from a position like this included a fall of a third, a fall of a third is information rather than an emergency. If you do not know it, the same fall is what makes you sell at the worst possible moment, with a perfectly sound four year cycle thesis still sitting in your notes. The calendar tells you where you are going. The path tells you what you have to be willing to hold through to get there. One of those is interesting. The other one is the reason people either make money or do not. ## What the opening band actually tells you This is the part I would keep if I had to throw the rest away. Eighteen completed windows, sorted by the band they opened in. **Every window that opened in the high band fell between 43% and 64% before it was done.** Every one, across three separate cycle tops, against a baseline of 14% for any window at random. Not most of them. All of them. **Windows that opened deep or low finished higher in dollar terms eight times out of nine**, against a base rate of 72% for any window at all. Those two findings are the practical output of this whole exercise. Not a price target and not a date. A conditional statement about what you are likely to have to survive, given where the window opened. If a window opens high, history says the fall is coming and says nothing about when. If it opens deep or low, history says the odds are good and still says nothing about when. Both are worth knowing before you size a position, and neither one tells you what to do next week. Positions also drift. Of those eighteen completed windows, only three stayed in the band they opened in, and the ones that opened cheap generally got cheaper before they recovered. So a reading getting cheaper is the ordinary path rather than evidence that something has broken. ## What this cannot do I would rather say this myself than have you find it out later. - **The samples are small.** Depending on the comparison, I am usually working with somewhere between four and twelve historical instances. Four is not a sample, it is an anecdote with a spreadsheet. I publish the count beside every figure, every time, because a median from four observations and a median from four hundred are completely different objects and they look identical on a page. - **The cadence gives the calendar. The analogues do not narrow it.** The window closing in December comes from the interval, not from the comparisons. What the comparisons give you is the range of what followed and where this one sits inside it. They do not tighten the date, because the paths spread wider as the window runs and are widest at the close. - **Individual altcoins mostly do not qualify.** An asset needs roughly six years of history before this kind of reading means anything, and tested on shorter histories it lands in the wrong place about two times in three. So I do not publish a position reading for individual alts on anything that runs unattended. A plausible wrong number is worse than a blank space, because people act on it. - **It is a record, not a prediction.** This is the sentence I repeat most often and I mean it literally. What similar starting points did is not what this one will do. ## A dated claim you can check Arguments about method are cheap. So here is one that resolves. The current window opened in May and closes in mid-December, which is where seven months from May lands. That date comes from the interval itself, fixed the day the window opened, not from anything I decided afterwards and not from the comparisons. Now Week 17 of 30 Opened May Date fixed on day one Closes mid-December Spread at its widest The comparisons do something different. They say what followed from this opening band before, and how wide the spread was. By December you will be able to check both: whether the window behaved anywhere inside that spread, and whether what was written down in May survived. That is the only test of any of this that means anything. Not whether the reasoning sounds good in September. Whether the thing that was written down in May survived December. Every call I make gets recorded the same way, with the date it was made, the condition that would prove it wrong, and the outcome when it arrives. Including the ones that go against me, and some have. That is the actual difference between this and a four year cycle countdown. Not that the calendar is wrong, because as far as anyone can measure, the calendar has been remarkably right. It is that a date on its own leaves you to face the journey with no idea what is normal, and that gap is where almost everyone loses money in a bull market. ### Every call, scored Every call I make goes into a public record: the date it was made, the condition that would prove it wrong, and what happened. The misses are in there too. Free to read, no account needed. [See the track record →](https://www.themarketsunplugged.com/every-call-scored/) [Join TMU →](https://www.themarketsunplugged.com/plans/) Members get the windows updated every Monday, for Bitcoin, Ethereum and the broad altcoin market, with the comparison set and the falls along the way printed beside every figure. This is part of my live analytical process, published with full context so you can follow the reasoning. Nothing here is a buy or sell instruction. Not financial advice. Always do your own research.