oz and sats per share

What the market is paying for Akoko

Take the whole company at its market price, subtract the treasury it has disclosed, and what is left is what the market is putting on everything else. No other treasury tracker can publish this, because no other issuer has a gold project sitting behind the treasury.

Compare the two charts. The lines above move every session, because they are prices. The line below moves only where a marker sits — the days the company bought or sold bitcoin, or issued shares — and is flat for months at a time in between. Over this period the share price fell 80.4% while sats per share fell 19.7%: the two answer different questions, and only one of them is about what management did.

And the line is not only going up. Sats per share peaked at 6.018 on 31 Mar 2026, the day of the last disclosed bitcoin purchase, and has fallen 40.9% since. Both halves of the fraction moved against it. The shares issued since would have taken it down 14.7% on their own; the change in bitcoin held, 30.8% on its own. They compound rather than add, which is why neither is the whole fall. That step down is the clearest thing on this page and the one a scoreboard is most tempted to leave off.

The series starts on 2025-10-09, which is the first session covered by both the price record and the share-capital ledger. Nothing before it is charted, because a market capitalisation needs a share count and the ledger's earliest stated total is 2025-10-09.

How much of the resource is which

One hue in two steps, not two colours: these are two confidence levels in one resource, not two different things. The brighter step is the one to weight more. Inferred is the lowest confidence category a resource statement has, and it cannot support a reserve or be used in a feasibility study — so 51% of the headline ounce count is material that a mine plan could not yet be built on. Every per-ounce figure above is struck on the whole 210,430 oz, which is why the Measured & Indicated version is printed beside it rather than instead of it.

What that residual is, next to what is in the ground

The crudest number anyone can attach to a gold resource is its gross in-situ value: contained ounces multiplied by the gold price, as if the metal were already refined and sold. Before mining, crushing, leaching, recovery, royalty, tax, or a single dollar of capital spent reaching it. At $4,194 an ounce that is $882,543,420 across the whole resource, and $432,820,800 across the Measured & Indicated half.

The market residual is 0.149% of the first figure and 0.303% of the second. A fraction of a percent is the ordinary case, not a finding. In-situ value is the number that ignores every cost of turning rock into metal, so almost nothing undeveloped trades at a meaningful share of it. It is here because it is the number people reach for when they ask what the ground is worth, and giving it with the right name and the right order of magnitude beats leaving someone to multiply it out for themselves without either.

This places a number. It does not judge one, and the bands are weaker evidence than they look. They are a third party's generalisation about other companies at three stages of development, not a peer set matched to this project — and Hamak is at none of those stages cleanly: the PEA is under way and unpublished, the option over Akoko is unexercised, and 51% of the ounces are Inferred and cannot support a reserve. The marker sits below every band shown, which is drawn rather than clamped to the left edge so it cannot be misread as the bottom of the first one. A figure under a peer band is not a statement that anything is mispriced, and the words that would make it one are barred here as they are in the room. The residual is also struck on market capitalisation rather than a true enterprise value, because the company publishes cash only in its periodic accounts and there is no current figure to net — so it is not comparable like-for-like with peer EV/oz figures computed the usual way.

There is no discounted case on this page, and that is deliberate. Running the arithmetic forwards — a production rate, a recovery, a royalty, a margin, a net present value — would take four numbers nobody has published and produce one that would be quoted back as what Akoko is worth. The PEA that would supply them is under way and unpublished. So the comparison here runs the other way: it takes what the market is already paying and says how that sits against the largest number the ground could support, which needs no invented input at all.

This is a residual, not a valuation of Akoko. It is everything the market is pricing that is not the disclosed treasury — the option over the project, the licence, the team, whatever cash it holds today, which it reports only twice a year — less whatever it is discounting for debt and dilution. Hamak holds an option over Akoko and has not exercised it. The Preliminary Economic Assessment is under way and unpublished. Nothing is in production, and roughly half the resource is Inferred, the lowest confidence category, which cannot support a reserve. The figure can be negative. It is not cash flow, and it is not a view on whether anything is priced correctly.

Per-ounce figures here are struck on market capitalisation, not a true enterprise value — the company does not publish a daily cash figure, so there is nothing to net off. Peer EV/oz figures are computed differently and are not directly comparable.

What a raise would do

The best-known rule in this corner of the market is that a company trading above an mNAV of 1.0 makes its holders better off by issuing shares and buying more bitcoin. That rule is correct — for a company where every pound raised buys bitcoin. Hamak also funds a gold project and is repaying a loan, so only a fraction of what it raises reaches the treasury, and the threshold moves with that fraction: break-even mNAV = 1 ÷ [ α × (1 − fee) × (1 − discount) ], where α is the share of net proceeds that actually buys treasury assets. At α = 0.5 the break-even is above 2.0, not 1.0 — so on this company the slogan does not merely lose precision, it can give the opposite answer.

Everything in this section is struck against an mNAV of 1.772× and a 0.450p share, the closes of 12 Oct 2026. Unlike the tiles above it, it is not refreshed in the browser: the break-even figures contain no price at all and never move, while the verdicts beside them turn on the mNAV printed here.

The threshold, before any of Hamak's own terms

Fee 3.0% throughout — the rate on the company's own ATM facility rather than a placing commission, so this grid and the table below are on the same footing. Shaded cells are the ones today's mNAV of 1.772× clears. A raise on terms in an unshaded cell would lower treasury value per share, however favourable the headline looks.

What the announced terms actually allow

α is not a guess here. Two announcements fix it between them. The £30 million at-the-market facility pays the company 97% of sale proceeds, and the amended funding package of 2 Jul 2026 lets the company keep the first £110,000 of ATM proceeds, with 50% of the remainder applied to the loan. Both are RNS-grade and both are on the open balance-sheet ledger, so the table below is folded from announcements rather than from an assumption.

The break-even does not rise with the size of the raise — it turns. Half of everything above £110,000 goes to the lender until the loan is gone, so the fraction reaching the treasury falls as a draw gets bigger, and then climbs once the repayment runs out. On the last announced loan balance of £1,430,000 the turn comes at a draw of about £3,062,000. The consequence, at today's mNAV: a draw below about £693,000 or above about £3,525,000 would raise treasury value per share, and anything in between would lower it. Both of those figures are arithmetic on announced terms, not a view about the company.

The same draws, against the scoreboard

Shares derived from the draw at the 0.450p close, because an ATM sells at the market price — the size of the raise and the number of shares it takes are the same decision. New buying is split 99.6% bitcoin and 0.4% gold, the proportion the treasury is already held in, because the company has published no allocation policy and that is the least invented assumption available. Units bought are converted at today's quoted prices; no future price is assumed anywhere, and no projected value per share is published.

Watch the £1,000,000 row: sats per share and ounces per million shares both edge up while treasury value per share falls. That is not an inconsistency. The treasury also holds 6,500,000 Vaultz Capital shares, which a raise does not add to — so the Vaultz stake per share is diluted by the full size of the issue, and on that row it outweighs the bitcoin and gold bought with the proceeds. Two per-share measures can move in opposite directions, and here they do.

Four things this section is not. It is not a forecast: every figure is arithmetic on inputs chosen here and prices quoted on 12 Oct 2026, which is what Tier B means on this site, and there is no fitted projection anywhere in the engine. The α column is a ceiling, not an allocation — retained cash still has to fund Akoko and running costs before any of it buys an asset, so every break-even above is a floor. The loan balance is the one number that would sharpen all of this and it is published only now and then: £1,430,000 is the figure last announced on 25 Sept 2026 , and the loan amortises monthly, so the true balance is lower — which moves the turn to a smaller draw and makes raises mildly more accretive than shown. And the announcement says “ATM proceeds” without saying whether that is before or after the facility's fee; the tables read it as after, which is the less flattering reading. One more, on the arithmetic's own terms: median daily trading in HAMA is about £6,119 of stock. The draws at the bottom of these tables are years of volume, and the sizes that clear the break-even today are precisely the implausible ones.

Definitions

Sats per share

bitcoin held × 100,000,000 ÷ shares. A satoshi is one hundred-millionth of a bitcoin.

Gold oz per 1,000,000 shares

gold kilograms × 32.1507 ÷ shares × 1,000,000. Shown per million shares because per single share it is around a ten-millionth of an ounce — true, and unreadable.

Shares outstanding

issued share capital − shares held in treasury. Shares a company holds in itself carry no vote and no economics, so they are excluded here and from market capitalisation.

FDSO and ADSO

Fully diluted counts only instruments currently in the money; all-dilution counts every live instrument whatever its strike. Both are shown because at this share price they differ by a quarter of a billion shares.

mNAV

market capitalisation ÷ treasury value. Above 1.0 the market pays more than the disclosed assets are worth; below, less. Assets and market value are different kinds of number and there is no rule that they should agree — the gap is an observation.

Market residual

market capitalisation − treasury value. Also shown per contained ounce and per Measured & Indicated ounce. Not a valuation of Akoko.

Source grades

Only figures from an announcement or a published filing are used in the numbers above. Figures that exist only because someone said them in an interview are recorded in the ledgers, marked, and never divided by.

Prices: Yahoo Finance, delayed data. Share and Vaultz prices are delayed and quoted in pence; bitcoin and gold in US dollars, converted at the sterling rate of the same session. The gold series is the COMEX front-month future, which tracks spot closely but is not spot.