← BitMine Immersion Technologies, Inc.
BMNR · Forward model · Bear case
The Bear case, 20 quarters out
Model as of
Each segment is projected from its own operating driver, rolled up into consolidated cash flow, and discounted back to a fair value per share. The assumptions below are editable — change them and every number on this page moves with them.
READ THE SCENARIO SPREAD AS A WARNING, NOT AS COMFORT. This is a digital-asset treasury company, so 93.7% of the base case's $26.92 is the ETH mark sitting in net cash, and net cash is the one input the scenarios cannot move. Bear to Bull spans $26.27 to $27.89 - six percent - while the thing that actually decides the answer sits outside that range entirely: hold everything else and put ETH at $2,000 and fair value is $21.95; put it at $3,000 and it is $31.73. A narrow spread here means the model is honest about what it can compute, not that the stock is safe. WHAT NET CASH CONTAINS. $15.22bn: the $15.6bn of crypto, cash and marketable securities and 'moonshot' stakes the company disclosed as of 30 August 2026, less the $350m stated amount of the 9.50% Series A Perpetual Preferred (3.5m shares at $100), less $30.1m of total liabilities from the 31 May 2026 balance sheet. There is no debt of any kind. Deducting the preferred at stated amount is deliberately conservative - its perpetual $33.25m coupon discounted at 15% is worth $222m, and the shares quoted around $86 in mid-July, roughly $301m. THE DOUBLE-COUNT QUESTION, STATED OPENLY. The model values the ETH at its mark AND the staking rewards it earns. A direct holder of ETH could stake it too, so this only adds value if the spot price does not already capitalise the holder's own yield. The position taken here is that BitMine's equity is worth ETH plus yield MINUS the costs a direct holder never pays - roughly $150m a year of overhead and a $33m preferred coupon, both of which are in the projection. Disagree with that and the answer is nearer $25 than $27. It is written down so it can be argued with rather than swallowed. FISCAL LABELS ARE FISCAL. The year ends 31 AUGUST, so 2026 Q1 ended 30 November 2025 and the basis quarter 2026 Q3 ended 31 May 2026. And on 9 July 2026 the board moved the year end to 31 December after the year ending 31 August 2026, with a four-month transition period on Form 10-KT, so projected quarter two stands in for September-December 2026 and labels from 2027 Q1 are calendar quarters. WHAT IS REPORTED AND WHAT IS NOT. Every quarterly figure in both verticals is a reported number: the fiscal 2026 Q1 and Q2 10-Qs each print the staking line and the individual legacy lines, and the Q3 10-Q prints all five disaggregated lines on the face of the income statement. They reconcile to consolidated revenue exactly in all seven quarters - $2,293, $11,041 and $46,535 thousand for the three fiscal 2026 quarters. Nothing is apportioned and nothing is estimated. The one thing that is NOT disclosed is the split of the basis quarter's $5,726 thousand cost of sales between the two lines; the 89.7% staking margin is derived from the Q1 and Q2 splits (staking cost was 3.0% of staking revenue in both) adjusted upward for the Q3 MD&A's statement that the cost ratio rose after the Pier Two acquisition, and the legacy line's -33.3% margin is measured directly from the Q2 disaggregation. NO SEASONALITY, AND THAT IS A FINDING. Ethereum pays rewards every 6.4-minute epoch with no calendar mechanism, and the staking line has only three non-zero quarters, so no centred four-quarter moving average can be formed. The legacy line has seven quarters but yields at most one moving-average ratio per quarter index and none for Q2, so signal cannot be separated from spread; its visible shape is contract timing - a machine lease that expired 31 December 2025 and a consulting contract that ran to 15 May 2026 - not a season. Neither vertical carries factors. TERMINAL VALUE. The present value of the terminal is 51.8% of the $1.02bn operating enterprise value but only 3.3% of fair value, because net cash dwarfs it. Exit multiple from 4x to 16x moves fair value $26.48 to $27.80. EXCHANGE. Common stock and the BMNP preferred both trade on the New York Stock Exchange. The common uplisted from the NYSE American on 9 April 2026; the 8-A12B of 10 June 2026 registers the preferred, not the common.
The yield the run-rate is built on is a seven-day spot rate, and the realised one is lower. Fiscal 2026 Q3's $45.7m annualises to roughly 2.14% on the staked base, a fifth below the 2.67% headline, and the 10-Q says the direction of travel is structural: as more ETH is staked network-wide the per-validator reward rate falls. This adds 1.5% a quarter of compression on top of the base case's 1%, so the effective yield grinds toward the realised rate over five years, cuts margin six points because the custody bill scales with the treasury rather than with the yield, and takes the exit to 6x. Fair value $26.27, still above the $25.32 price - because the ETH is still there. What this case does NOT do is mark the ETH down; it cannot, and that is the real bear case.
Latest: $57M (2031Q3E)
| Period | Value |
|---|---|
| 2025Q1 | $1M |
| 2025Q2 | $2M |
| 2025Q3 | $2M |
| 2025Q4 | $1M |
| 2026Q1 | $2M |
| 2026Q2 | $11M |
| 2026Q3 | $47M |
| 2026Q4E | $84M |
| 2027Q1E | $83M |
| 2027Q2E | $81M |
| 2027Q3E | $79M |
| 2027Q4E | $78M |
| 2028Q1E | $76M |
| 2028Q2E | $75M |
| 2028Q3E | $73M |
| 2028Q4E | $72M |
| 2029Q1E | $70M |
| 2029Q2E | $69M |
| 2029Q3E | $68M |
| 2029Q4E | $66M |
| 2030Q1E | $65M |
| 2030Q2E | $64M |
| 2030Q3E | $62M |
| 2030Q4E | $61M |
| 2031Q1E | $60M |
| 2031Q2E | $59M |
| 2031Q3E | $57M |
What drives each segment
ETH staking and validation
Capacity × utilisation × price5,067,309 of BitMine's 5,901,112 ETH are staked through MAVAN, and every epoch they pay the company rewards in ether. Revenue is three numbers multiplied together: how many coins are staked, what annualised yield the protocol pays, and what an ETH is worth. The company publishes all three every week. At 5,067,309 staked, a 2.67% seven-day yield and a $2,511 mark that is $340 million a year, and it has said what full staking would add: $396 million. The line went from nothing to 98% of revenue in three quarters. The threat is in the yield, not the volume - the 10-Q says plainly that as more ETH is staked network-wide the per-validator reward rate falls.
Latest: $57M (2031Q3E)
| Period | Value |
|---|---|
| 2025Q1 | $0.00 |
| 2025Q2 | $0.00 |
| 2025Q3 | $0.00 |
| 2025Q4 | $0.00 |
| 2026Q1 | $980,000.00 |
| 2026Q2 | $10M |
| 2026Q3 | $46M |
| 2026Q4E | $84M |
| 2027Q1E | $82M |
| 2027Q2E | $81M |
| 2027Q3E | $79M |
| 2027Q4E | $78M |
| 2028Q1E | $76M |
| 2028Q2E | $75M |
| 2028Q3E | $73M |
| 2028Q4E | $72M |
| 2029Q1E | $70M |
| 2029Q2E | $69M |
| 2029Q3E | $67M |
| 2029Q4E | $66M |
| 2030Q1E | $65M |
| 2030Q2E | $63M |
| 2030Q3E | $62M |
| 2030Q4E | $61M |
| 2031Q1E | $60M |
| 2031Q2E | $58M |
| 2031Q3E | $57M |
Assumptions & reasoning
- Every quarter above is a reported figure, not an allocation. The fiscal 2026 Q1 and Q2 10-Qs each print 'Revenue from staking' as its own line ($980 and $10,201 thousand) and the Q3 10-Q prints 'Revenue from staking and validation $45,743' on the face of the income statement. The four zeros are equally reported: native staking did not begin until November 2025.
- The capacity unit is an ETH the company OWNS, and utilisation is the share of them that is staked. That is what makes the model checkable: 5,901,112 x 85.87% x $16.76 a quarter is $84.9 million, which annualises to the $340 million the company disclosed on 31 August 2026, and taking utilisation to 100% gives $98.9 million a quarter, or the $396 million it disclosed for the fully-staked stack. Both to within 0.1%.
- Revenue per ETH is the ETH price times the annualised yield, divided by four: $2,511 x 2.67% / 4 = $16.76. The base case holds the ETH price FLAT at the 30 August mark, because a base case that embeds an ETH forecast is not a base case. The -1% quarterly drift is yield compression alone, and it is sourced: the 10-Q states that as the total amount of Ethereum staked on the network increases, the per-validator reward rate generally decreases.
- Utilisation glides toward a disclosed 100% at a rate of zero, so the base case does not assume the unstaked 834,000 coins ever get staked. That is deliberate: the staked count has been frozen at exactly 5,067,309 in every weekly release since 16 August while the stack kept growing. Moving the glide off zero is the Bull case, and it is worth 16.5%.
- Accumulation is nearly over, which is why capacity adds only 50,000 ETH a quarter and decays 20% from there. The company is 133,888 coins short of 5% of the ETH supply, weekly purchases have fallen from 53,501 to 7,430 and back, and about 23,000 ETH a quarter now arrive as staking rewards in kind whether or not anything is bought.
- Revenue is recognised GROSS: the company is the principal node operator and books the whole protocol reward, with amounts passed to delegators and network participants sitting in cost of sales. That inflates both revenue and cost against a net presentation, so this line is not directly comparable to a staking peer that reports net.
- MAVAN is still a captive platform. Nothing here assumes a single third-party client - the 10-Q concedes the company has 'limited operating history as a commercial staking-services provider'.
Self mining, consulting and leasing
Growth pathWhat is left of the company BitMine was before June 2025 - immersion-cooled bitcoin mining, hosting, a machine lease and a consulting contract. $792 thousand in the basis quarter against $45.7 million of staking, and shrinking on a published schedule: the machine lease expired 31 December 2025 and took leasing to zero, the KULR consulting agreement ran to 15 May 2026, equipment sales stopped altogether. What remains is a bitcoin mining operation the company says only that it is 'maintaining'. It is modelled because it is reported, not because it matters.
Latest: $171,805.78 (2031Q3E)
| Period | Value |
|---|---|
| 2025Q1 | $1M |
| 2025Q2 | $2M |
| 2025Q3 | $2M |
| 2025Q4 | $1M |
| 2026Q1 | $1M |
| 2026Q2 | $840,000.00 |
| 2026Q3 | $792,000.00 |
| 2026Q4E | $561,686.40 |
| 2027Q1E | $511,134.62 |
| 2027Q2E | $469,221.58 |
| 2027Q3E | $433,748.43 |
| 2027Q4E | $403,177.84 |
| 2028Q1E | $376,413.29 |
| 2028Q2E | $352,658.90 |
| 2028Q3E | $331,328.06 |
| 2028Q4E | $311,982.27 |
| 2029Q1E | $294,289.48 |
| 2029Q2E | $277,995.05 |
| 2029Q3E | $262,901.32 |
| 2029Q4E | $248,852.93 |
| 2030Q1E | $235,726.24 |
| 2030Q2E | $223,421.56 |
| 2030Q3E | $211,857.44 |
| 2030Q4E | $200,966.40 |
| 2031Q1E | $190,691.82 |
| 2031Q2E | $180,985.46 |
| 2031Q3E | $171,805.78 |
Assumptions & reasoning
- Reported, not apportioned. The fiscal 2026 Q1 10-Q prints self-mining $2, consulting $199 and leasing $1,112 thousand; the Q2 10-Q prints $219, $197 and $424; the Q3 10-Q prints $624, $168 and nil. Each set sums to the value above and, added to the staking line, to reported consolidated revenue exactly - $2,293, $11,041 and $46,535 thousand.
- The step change is a dated contract expiry, not a growth rate: the KULR consulting agreement ran 'from May 16, 2025 to May 15, 2026' and contributed $168 thousand of the basis quarter's $792 thousand, so the level drops 21.2% and it applies at the basis quarter because the contract had already lapsed sixteen days before that quarter closed.
- The margin here is measured rather than assumed. The fiscal 2026 Q2 10-Q disaggregates cost of sales - staking $306 thousand of a $1,426 thousand total - which leaves $1,120 thousand of legacy cost against $840 thousand of legacy revenue. Self-mining, hosting rent and electricity cost more than the bitcoin they produce.
- There is no operational driver left to model even if one were wanted. The company stopped publishing hashrate and energy metrics entirely, saying they are 'no longer decision useful' after the pivot to an asset-light ETH model.
Where each case comes from
Bear case — primary sources
The primary sources this case is built from — filings, calls and posts. Where they are claims by an interested party rather than disclosures, every number in the Bear column is what happens if they are taken at face value.
The disclosed yield risk
- Jul 14, 2026 As the total amount of Ethereum staked on the network increases, the per-validator reward rate generally decreases, which could result in a material reduction in the yield we earn on our staked assets over time.
- Jul 14, 2026 For the three months ended May 31, 2026, revenue from staking and validation was $45.7 million, representing 98% of total revenue of $46.5 million for the quarter.
Bull case — primary sources
The primary sources this case is built from — filings, calls and posts. Where they are claims by an interested party rather than disclosures, every number in the Bull column is what happens if they are taken at face value.
What full staking is worth, in the company's own numbers
- Aug 31, 2026 At scale (when Bitmine’s ETH is fully staked by MAVAN and its staking partners), the projected ETH staking reward is $396 million on an annualized basis (using 2.67% 7-day BMNR yield)
- Aug 31, 2026 Annualized staking revenues are now projected at $340 million. And this 5.1 million ETH is 86% of the 5.90 million ETH held by Bitmine.
Alchemy of 5% case — primary sources
The primary sources this case is built from — filings, calls and posts. Where they are claims by an interested party rather than disclosures, every number in the Alchemy of 5% column is what happens if they are taken at face value.
How close the 5% goal is
- Aug 31, 2026 Bitmine is 98% of the way to the ‘Alchemy of 5%’ in just 15 months
- Aug 31, 2026 Bitmine owns 4.9% of the total ETH coin supply of 120.7 million
- Aug 31, 2026 As of August 30, 2026 at 3:00pm ET, the Company’s crypto holdings are comprised of 5,901,112 ETH at $2,511 per ETH (per Coinbase NASDAQ: COIN), 211 Bitcoin (BTC), $180 million stake in Beast Industries, $81 million stake in Eightco Holdings (NASDAQ: ORBS) (“moonshots”) and total cash & marketable securities of $541 million.
From cash flow to fair value
| Present value of free cash flow, 20 quarters | $369M |
| Terminal-year revenue | $237M |
| Terminal-year EBITDA | $86M |
| Exit multiple, on ebitda | 6.0x |
| Terminal value | $516M |
| Discounted at 15.0% a year, terminal value becomes | $257M |
| Enterprise value | $626M |
| Net cash | $15.22B |
| Equity value | $15.85B |
| Shares | 0.60B |
| Fair value per share | $26.27 |
| Against the deployed price of $25.32, as of | +4% |
8x terminal EBITDA on a validator business, and it barely matters. Almost none of this company's value is in the projection: net asset value is $15.22bn - the $15.6bn of ETH, bitcoin, cash and private stakes the company disclosed on 31 August 2026, less the $350m preferred stated amount and $30m of other liabilities - which is $25.23 a share, or 93.7% of the base case's $26.92. Everything the discounted cash flow adds is MAVAN: about $85m a quarter at 89.7% gross margin, less corporate overhead running at 43.5% of revenue, discounted at 15% to an 8x exit. That is $1.02bn of enterprise value, or $1.69 a share. Take the exit multiple all the way from 4x to 16x and fair value moves from $26.48 to $27.80 - $1.32, less than 5%. Take the discount rate from 20% to 10% and it moves 55 cents. Now move the ETH price: at $2,000 fair value is $21.95 and at $3,000 it is $31.73. Every slider on this page argues over about a dollar; the asset none of them touches is worth twenty-five. The 15% discount rate is not a guess - BitMine sold perpetual preferred at $80 against a $100 stated amount and a 9.50% coupon on 10 June 2026, an 11.875% cost of capital observed in a real transaction, and the common ranks behind it.
Read the other way round: at $25.32 the market is paying -7.4x terminal-year revenue, holding every other assumption on this page fixed. That is the number to argue about.
The projected path
| Quarter | ETH staking and validation | Self mining, consulting and leasing | Revenue | YoY | EBITDA | Capex | FCF | R40 | PV of FCF |
|---|---|---|---|---|---|---|---|---|---|
| 2026 Q4E | $84M | $561686 | $84M | — | $33M | $609721 | $32M | — | $31M |
| 2027 Q1E | $82M | $511135 | $83M | — | $32M | $621175 | $31M | — | $29M |
| 2027 Q2E | $81M | $469222 | $81M | — | $31M | $629403 | $30M | — | $27M |
| 2027 Q3E | $79M | $433748 | $79M | +71% | $30M | $634859 | $29M | +108 | $26M |
| 2027 Q4E | $78M | $403178 | $78M | -7% | $29M | $637939 | $29M | +29 | $24M |
| 2028 Q1E | $76M | $376413 | $76M | -7% | $28M | $638984 | $28M | +29 | $23M |
| 2028 Q2E | $75M | $352659 | $75M | -8% | $28M | $638293 | $27M | +29 | $21M |
| 2028 Q3E | $73M | $331328 | $73M | -8% | $27M | $636122 | $26M | +28 | $20M |
| 2028 Q4E | $72M | $311982 | $72M | -8% | $26M | $632694 | $26M | +28 | $19M |
| 2029 Q1E | $70M | $294289 | $70M | -8% | $26M | $628200 | $25M | +28 | $18M |
| 2029 Q2E | $69M | $277995 | $69M | -8% | $25M | $622805 | $25M | +28 | $17M |
| 2029 Q3E | $67M | $262901 | $68M | -8% | $25M | $616652 | $24M | +28 | $16M |
| 2029 Q4E | $66M | $248853 | $66M | -8% | $24M | $609865 | $24M | +28 | $15M |
| 2030 Q1E | $65M | $235726 | $65M | -8% | $24M | $602552 | $23M | +28 | $14M |
| 2030 Q2E | $63M | $223422 | $64M | -8% | $23M | $594803 | $23M | +28 | $13M |
| 2030 Q3E | $62M | $211857 | $62M | -8% | $23M | $586700 | $22M | +28 | $13M |
| 2030 Q4E | $61M | $200966 | $61M | -8% | $22M | $578311 | $22M | +28 | $12M |
| 2031 Q1E | $60M | $190692 | $60M | -8% | $22M | $569697 | $21M | +28 | $11M |
| 2031 Q2E | $58M | $180985 | $59M | -8% | $21M | $560909 | $21M | +28 | $11M |
| 2031 Q3E | $57M | $171806 | $57M | -8% | $21M | $551993 | $20M | +28 | $10M |
Every row is projected. A year-over-year change is shown only where the quarter it compares against exists — an em dash means there is no comparable quarter, not a flat year.
Model revisions
Assumptions are marked to reality as each quarter prints. Every change is appended here, with the fair value the model produced at the time, so the model's own history stays visible.
| Date | Fair value then | Note |
|---|---|---|
| 2026-09-01 | $26.92 | First model. Built on the fiscal 2026 Q3 10-Q filed 14 July 2026 and the weekly Regulation FD disclosure of 31 August 2026. |