Every forward projection assumes the old high returns

Every outlook and review on the site prices a machine's future coins at bitcoin's previous all-time high, not today's, and runs that number to the halving expected in April 2028. That looks generous. The quieter alternative — assuming a price bitcoin already reached is never reached again — is the assumption nobody notices they'd be making instead.

A carved stone high-water mark set into an old canal-side wall, lit by low evening light

A number that stops at a known date

Every outlook and review on the site carries one figure that reaches further than the rest of the page: what a machine mines between today and the next halving, expected on 20 April 2028. Every other number on the page describes something that already happened. This one describes something that has not, and it stops at the halving rather than running on indefinitely, because that is the one future date on the page that is actually known.

To get there, the figure makes two calls rather than one. It holds hashprice — what a terahash earns in a day — flat at today's level for every day between now and the halving, instead of letting it drift the way it always has. And it prices the coins that arithmetic produces not at today's price, but at bitcoin's previous all-time high.

The first call is a simplification, clearly labelled as one: nobody is claiming hashprice will actually sit still for a year and a half. The second is the one worth examining on its own, because it looks like the generous assumption and it is actually the more defensible one.

The choice is between two guesses, not one guess and a fact

A forward figure has to put a dollar value on bitcoin a machine has not mined yet, and there is no way round choosing a number to do it with. Leaving the number out is the only neutral option, and it is not the option a reader asking "what does this machine do next" actually wants answered.

The choice that looks neutral is today's price, held flat all the way to the halving. It feels conservative because it appears to add nothing to what the market says right now. But bitcoin is trading well below its own record as this is written, and holding that gap flat for a year and a half is not a null assumption — it is a bet, just a quiet one: that the record the market has already set is never reached again before the clock this figure runs on stops.

The number this site uses makes the opposite bet, out loud: that a price bitcoin has already printed gets printed again before April 2028. Both are guesses about a future nobody can see. Only one of them can be checked against what has actually happened every other time this question came up, and that is the test worth running.

Holding today's price flat to the halving is not the neutral choice. It is the quieter bet.

What happened after the last three halvings

Bitcoin's daily price is on record back to 2013, which covers three of the four halvings so far, start to finish. Each one is a clean test of the same question: once a halving lands, how long does it take for the record standing on that day to be broken again?

After the 9 July 2016 halving, the record was $1,137, set in December 2013. It was broken on 24 February 2017 — 230 days later.

After the 11 May 2020 halving, the record was $18,912, set in December 2017. It was broken on 23 November 2020 — 196 days later.

After the 20 April 2024 halving, the record was $72,743, set a month earlier on 13 March 2024. It was broken on 6 November 2024 — 200 days later.

Three halvings, three records broken within about seven months of the subsidy cut that followed them. None of them needed anywhere near the two and a half years this site's forward number allows itself, which makes the number's own deadline look less like optimism and more like a wide margin.

Chart: Distance from bitcoin's running record price, every day since 2013, with each halving marked. Every drawdown on this chart has eventually closed back to zero. Open the chart

Two and a half years is room, not a promise

Those three examples measure from the halving forward. The fairer comparison for where bitcoin stands today measures from the record itself, because the current one was not set at a halving at all — it was set mid-cycle, on 6 October 2025, at $124,493.

Widen the lens to every point in the record where a high was followed by a real drawdown rather than an immediate continuation of the same rally, and three longer cycles appear. The December 2013 high took 3.2 years to be broken again, in February 2017. The December 2017 high took 2.9 years, broken in November 2020. The November 2021 high took 2.3 years, broken in March 2024.

The 2.5 years between October 2025's record and the halving now expected in April 2028 sits inside that range — shorter than the two oldest cycles needed, a touch longer than the most recent one.

It is not the outer edge of what bitcoin has done before. It sits close to the middle of it.

A carved stone high-water mark set into a stone canal wall at the waterline, evening light — A high-water mark is kept because the water has reached it before, not because it is high today. Bitcoin's own record has worked the same way so far: revisited, not written off.

Where this cycle stands as it is written

Bitcoin traded down from October's record to $59,014 on 30 June 2026, a fall of about 53%, then recovered part of the way back: $86,608 by 22 September 2026, about 30% below the record. The site's own chart of distance from the all-time high shows this stretch next to every earlier one on the same axis.

The three earlier cycles above went a good deal further before they turned. The December 2013 high was not revisited until the market had fallen 85%, in January 2015. The December 2017 high waited out an 82% fall, in December 2018. The November 2021 high waited out a 76% fall, in November 2022. This cycle's deepest point so far, 53%, is shallower than any of the three, for whatever that turns out to be worth.

None of that predicts anything about what comes next, and this piece is not putting a date on it either. Cycles have shrunk every time so far — the gap between a record and its replacement has gone from 3.2 years to 2.9 to 2.3 — and a fourth data point is not a law. What the record shows is a pattern repeated three times running, not a guarantee of a fourth.

The other half of the balance

A halving does not just sit in the background of this number — it is the reason the number stops where it does at all. Every 210,000 blocks the reward paid to miners halves: 50 bitcoin, then 25, then 12.5, then 6.25, then 3.125, and 1.5625 from around April 2028. Hashprice steps down with it, because the network is suddenly paid half as much bitcoin for the same work, on the same day, whatever the price is doing. What that does to a machine's own break-even line, and why efficiency is what decides which machines shrug it off, is covered in Efficiency decides how many halvings a machine survives.

The forward number does not model that step down day by day. It holds hashprice flat straight through the cut, which on its own would understate what actually happens on the day the subsidy halves. What balances it is the other half of the same assumption: the recovery that has followed every previous halving is also what has put hashprice back to where it started, and usually higher, within roughly a year of every previous cut. One assumption is being asked to carry both sides of the ledger — the cut and the recovery — rather than counting the good news twice or the bad news not at all.

One assumption is asked to carry both sides of the ledger: the cut and the recovery.

What it means for a miner deciding whether to trust the numbers

None of this makes the forward figure a prediction, and nobody here is calling a price or a date. It is a number built on a rule that is written down, checked against the whole of bitcoin's price history rather than the convenient parts of it, and honest about which halvings it has and has not actually been tested against.

The rule could still be wrong. A cycle that has kept shrinking every time so far could eventually shrink to nothing, and the fourth halving's record could stand for longer than the previous three managed. That is exactly why every page carrying the number says plainly what it assumes, rather than printing a total and letting a reader mistake it for a fact the way a spot price is a fact.

What a miner deciding whether to trust it can actually do is check the two things holding it up, because both are answerable from the record rather than from anyone's opinion of it: has a level bitcoin has already reached ever failed to be reached again, and has the market ever needed as long as two and a half years to do it once a halving had already landed. So far, on both counts, it has not — and a number that only asks for what has already happened three times running is a more careful use of an assumption than most forecasts get credit for being.

Disclosure

Firsthand Bitcoin sells and hosts mining hardware, including this machine. No manufacturer, distributor or affiliate programme paid for or reviewed this page and we take no commission on the links above. Historical figures are computed from daily bitcoin price and network hashprice, each day valued at its own prices. Nothing here is investment advice.