Bitcoin weekly close: $62,900, inside the range

The week, from open to close

Bitcoin traded between a weekly low of $62,535 and a high of $65,391, and finished -3.1% from Monday's open. The number that settles the range is the close, and this week the close was $62,900.

Measured against the centre of the projected band, that landed -0.4% away. The verdict: inside the band.

The track record this range joins

6 of the 7 projected weekly ranges published so far resolved inside the band. Every one of them was frozen before its week, and every result went up afterwards, including the ones that went wrong.

There is a trap in that number worth naming. A band that contains the close every single time is not a good model, it is a wide one. The published band is the middle half of the distribution, so a properly calibrated engine should land inside about half the time. A hit rate far above that means the bands are wider than they should be, which is a miscalibration too, just the flattering kind.

Why the timestamps are the whole point

This range was emitted on August 9, 2026 and resolves against the weekly close of August 16, 2026. Those two dates are not decoration. They are what separates a forecast from a comment, because once the first one passes the number cannot be edited.

Almost nobody in this corner of the internet publishes their errors. That is not a moral failing, it is an incentive: an unmarked forecast is always right in hindsight. The full record, including the ranges that missed, is public and dated, and it can be read as data rather than taken on trust.

The range and the person buying every month

Most people reading this are not trading the week. They are buying Bitcoin gradually, dollar cost averaging into a position they intend to hold for years, and a seven day band is not a signal to act on.

What it is good for is calibration. A band of $62,699 to $63,610 tells you what the engine currently considers an ordinary week, so when price moves inside that span you know you are watching noise rather than a change of story. The decision of whether to buy stays yours, and it should be driven by a plan written when you were calm, not by a week.

What the width of the band says about Bitcoin volatility

The band is 1.4% wide between its edges. That single figure is the model stating, in advance, how much weekly movement it considered ordinary for Bitcoin right now. Wider bands mean the engine sees a jumpier market; narrower ones mean it is willing to commit.

It is a more useful reading than it looks, because it is the part of a forecast that people usually skip. A narrow band that holds is a real result. A wide band that holds proves very little, and the width is published precisely so nobody has to take the hit rate at face value.

The Bitcoin model behind the range

The band is not one forecast dressed up as a range. It comes out of a combined engine of four models (a power law over a decade of price history, a Monte Carlo simulation driven by GARCH volatility, an Ornstein-Uhlenbeck residual model and an implied density read from the Deribit options surface), pooled and reweighted every week against real history. Each one is wrong in its own way, which is the point: pooling them cancels part of the error that any single Bitcoin model carries alone.

Around fourteen live data sources feed the bias layer on top, among them MVRV-Z, SOPR, NUPL, the Puell multiple, Reserve Risk, the Bitcoin realized price, funding rates and global liquidity. What comes out is a distribution of thousands of simulated price paths, and the published band is its middle half, from the 25th to the 75th percentile. Roughly one path in two ends inside it.

How far Bitcoin travelled to get nowhere

Between the weekly low of $62,535 and the high of $65,391 there is 4.6% of ground, and the close ended 3.8% below the top of that span. The drawdown from the week's high is the number that tends to be felt rather than read.

Ranges are resolved on the close, not on the wick, and the two can tell very different stories. A week can spend days outside the band and still settle inside it, which is why the log records one number and not the most dramatic one available.

What this engine does not use

Worth being explicit, because the most searched Bitcoin models are not in this engine. There is no stock to flow here and no rainbow chart. Both are readable, both are popular, and both have spent the last cycle being argued about precisely because they were published as certainties and then had to be quietly re-drawn.

The power law is in, as one of the four generators, and it earns its place by being reweighted against real history every week rather than by looking convincing on a chart. Leaving the famous models out is not a dig at them. It is the same standard applied here: if a component cannot be scored against what actually happened, it does not get to move the band.

Where Bitcoin sits in its four year cycle

Counting days since the April 2024 halving puts this piece on day 849. The next halving is estimated by block height for April 10, 2028, a date that drifts with the real cadence of mined blocks rather than sitting on a calendar.

That count is the honest way to compare cycles, because what matters is not the date but how far in we are. It is also worth saying plainly that four halvings means four cases. Enough to see a shape, nowhere near enough to prove one, and whether the four year cycle still holds is an argument that is very much open. The cycle is context here, never the reason for a weekly number.

The next number is already live

The next projected range is drawn over the live Bitcoin chart in the app, alongside the dated zones for the longer horizons, and it is recalculated on the server every day so that every visitor sees the same numbers. Nothing here is computed in your browser and nothing is personalised.

None of this is investment advice. It is one statistical model keeping its numbers where anybody can check them, including the weeks it gets wrong.