Stock Analysis · Bulls vs Bears

Bitdeer Technologies Group BTDR

🐂 Bulls vs Bears 🐻Report #212 Oct 2026
★ Willem owns this stock · 1.8% of the portfolio
71
high risk▲ +4 vs last
🐂 9vs11 🐻
Bears lead
20%
▲ Chance of the biggest tailwind the forecasts miss
20%
▼ Chance of the biggest headwind the forecasts miss
The verdict

The assets look better than the stock price suggests. The Tydal rent of $202/kW/month is at the top of the peer range. JPMorgan is syndicating a $1.3B tranche that cash-backs the tenant's letter of credit. Enterprise value is only ~4.9x trailing sales, against 17–56x for AI-pivot peers. The funding side is worse than report 1 showed: H1 free cash flow was about −$463M against $457M of cash, related-party debt grows…

The Bulls vs Bears podcast is being made
Two AI hosts debate this report, NotebookLM-style. It appears here automatically within a few hours.

Written by Claude (Anthropic’s AI) from public sources · Willem owns BTDR · AI can make mistakes — numbers in brackets point to the sources · Not financial advice

🐂 Bull arguments 9

  1. 1Tydal rent sits at the top of the peer rangeStrong
  2. 2The market values Bitdeer like a miner, not an AI landlordMedium
  3. 3About 1.75 GW of power online, much of it cheap hydroMedium
  4. 4AI cloud revenue grew tenfold in ten monthsMedium
  5. 5Own chips cut the cash cost per bitcoin to about $39kMedium
  6. 6Bitcoin and hashprice have recovered from the June lowsMedium
  7. 7Insiders bought, a buyback started, and analysts stayed positiveWeak
  8. 8A Big Four auditor and remediated controls improve reportingWeak
  9. 9Heavy short interest could squeeze on a financing winWeak

🐻 Bear arguments 11

  1. 1Cash burn outruns cash: about two quarters of runwayHigh
  2. 2Dilution is relentless and the overhang is hugeHigh
  3. 3The founder's other company is lender, custodian and counterpartyMedium
  4. 4Net debt is about 12.6 times trailing EBITDAMedium
  5. 5Mining lost money at the gross line in H1High
  6. 6Missed dates and a retracted press release hurt credibilityHigh
  7. 7Tydal's tenant is borrowing at 11% and has a year-10 exitHigh
  8. 8Revenue rests on one pool, one country and one leaseMedium
  9. 9"Secured" Malaysia capacity has no customers yetMedium
  10. 10Power sits in jurisdictions that are tightening rulesMedium
  11. 11The chip lead over Bitmain is within measurement errorLow

Bull arguments show how well they survive the devil’s advocate (strong / medium / weak); bear arguments show the damage they could do after the counter-argument (high / medium / low).

Bull case

🐂 Bull #1Tydal rent sits at the top of the peer range

  • The point: Bitdeer leased Tydal at a price above most AI leases signed by bigger miners. The tenant's lenders are now putting cash behind the rent.
  • Evidence:
    • Tydal terms: 121 IT MW, 16 years, ~$4.7B, $202/kW/month average, 3% escalators, power passed through, NOI margin ~90% (company estimate) [9].
    • Peer monthly rents per kW (derived from public deal values) [41,42,43,44]:
      • TeraWulf/Anthropic ~$197.
      • Hut 8/Fluidstack ~$159.
      • Applied Digital Polaris Forge 3 ~$139.
      • Galaxy/CoreWeave ~$222.
    • New since report #1 (Bloomberg, 2026-10-05): JPMorgan is marketing a $5B loan to Volta in two tranches [25]:
      • $3.7B for ~36,000 Nvidia GPUs.
      • $1.3B that cash-collateralises the lease letter of credit.
      • Lender commitments are due 2026-10-14.
  • Why it matters: At $202/kW, Tydal produces ~$264M of yearly NOI (Benchmark estimate) [26]. KBW capitalised Tydal at a 6.7% cap rate [28]. On those figures Tydal alone is worth ~$3.9B, about equal to Bitdeer's entire EV of ~$3.99B [31]. Interpretation: this is my arithmetic. It ignores the ~$500M of capex still to spend and the year-10 exit.
  • 😈 Devil's advocate:
    • Peer rents are mostly triple-net (the tenant pays all running costs), and some are quoted per gross MW. Tydal is "modified gross" with a services agreement, so part of its premium covers costs that peers don't carry.
    • The high rent compensates for a tenant founded in 2026.
    • The Volta loan prices at ~11% yield, unrated, at 97–98 cents [25]. Lenders are demanding a junk-style return, and the loan has not closed yet.
  • Verdict: Strong The rent is verified in a filing, and a cash-collateralised letter of credit would make the early years close to bank risk once the loan closes.

🐂 Bull #2The market values Bitdeer like a miner, not an AI landlord

  • The point: AI-pivot peers trade at many times Bitdeer's sales multiple. A re-rating needs only proof of delivery, not new assets.
  • Evidence:
    • Market cap ~$2.50B (271.8M A+V shares × $9.19, 2026-10-12 intraday). EV ~$3.99B [31,32].
    • EV/sales (trailing) [31]:
      • BTDR 4.95x.
      • Core Scientific 17.1x, IREN 22.8x, Cipher 53.6x, Hut 8 56.1x, TeraWulf 56.4x.
    • EV per online MW is ~$2.3M (EV $3.99B / 1,752 MW; derived).
    • Sell-side targets average $22.14–23.45, against a share price of $9.19 [29,30].
  • Why it matters: If Tydal is delivered on time and one more large lease is signed, the gap to peers is the upside.
  • 😈 Devil's advocate:
    • Peers earned their multiples with investment-grade tenants: IREN/Microsoft, Cipher/AWS, TeraWulf/Anthropic directly [41,58].
    • Bitdeer's revenue is still ~84% variable mining revenue [1], which deserves a miner's multiple.
    • The low multiple also reflects dilution: per-share value is shrinking even as assets grow (Bear #2).
  • Verdict: Medium The discount is real, but much of it has a reason.

🐂 Bull #3About 1.75 GW of power online, much of it cheap hydro

  • The point: Grid power is the scarce input for AI. Bitdeer already holds a lot of it, at an average cost of $44/MWh.
  • Evidence:
    • Power: 1,752 MW online plus 1,228 MW pipeline, as of 2026-07-31. Q2 electricity cost $44/MWh [3].
    • ~$100M bought 200 acres next to Rockdale (Milam County, TX) in August [2,10].
    • The Gelephu (Bhutan) AI LOI covers 30 MW initially, with possible expansion to 500 MW. It is non-binding [7].
    • Texas paused approvals for new loads of 75 MW or more on 2026-08-03 [51]. That makes already-energised capacity like Rockdale's 563 MW scarcer.
  • Why it matters: Each MW converted from mining to a lease at Tydal-like pricing earns ~$2.2M of NOI a year (121 MW → $264M) [9,26].
  • 😈 Devil's advocate:
    • Only Tydal is contracted.
    • Bhutan (600 MW) and Ethiopia (50 MW) are poor AI locations.
    • Rockdale has no AI decision. Management is "preparing in the background", and the site is "pre-batch 0" [13]. Any new AI load there hits the Texas pause.
    • Clarington (570 MW) is in discovery in the AHP lawsuit [13].
  • Verdict: Medium The power is real. The part that is AI-ready and unencumbered is far smaller than 1.75 GW.

🐂 Bull #4AI cloud revenue grew tenfold in ten months

  • The point: Bitdeer's own GPU cloud is a second contracted business, and it is growing fast.
  • Evidence:
    • AI cloud ARR grew from ~$8M (Oct 2025) to ~$86M (Aug 2026), on 4,328 GPUs at 92% utilisation [7,17].
    • ARR counts only contractually obligated GPU orders [7].
    • A102 (9.5 MW) is 100% signed: "more than $800 million" over five-year commitments, with revenue from Q1 2027 [7].
    • Customer prepayments target ~50% of GPU capex [7].
  • Why it matters: $800M over 5 years from 9.5 MW is ~$16.8M of revenue per MW per year (derived). That is about 7x Tydal's ~$2.4M per MW, because Bitdeer supplies the GPUs too.
  • 😈 Devil's advocate:
    • AI cloud lost money at the gross line: −$2.3M in Q2 and −$7.3M in H1 [1].
    • TheEnergyMag (secondary) reported A102 as a ~$400M contract for about half the site [56]. That conflicts with the company's "$800M+ for 100%". The company figure wins as the primary source, but it implies a second unnamed deal.
    • Customers are unnamed. Each MW of GPU cloud needs roughly $25–30M of GPU capex (estimate), which brings more dilution.
  • Verdict: Medium The growth is filed, but there is no profit yet, and the capital intensity is very high.

🐂 Bull #5Own chips cut the cash cost per bitcoin to about $39k

  • The point: Building its own rigs and running them on cheap power gives Bitdeer one of the lowest cash costs per coin in the industry.
  • Evidence:
    • Fleet efficiency 15.8 J/TH; hashrate 73.0 EH/s in June and 79.9 EH/s in August [3,7].
    • Q2 self-mining costs: electricity $84.7M, depreciation and stock pay $79.8M, other $6.5M [1].
    • Derived cost per self-mined coin, splitting BTC between self-mining and co-mining by revenue share (an assumption):
      • Cash cost: ~$54.8k in Q1-26, ~$38.9k in Q2-26.
      • Including depreciation: ~$94.7k in Q1-26, ~$72.9k in Q2-26.
    • Correction to report #1: Q2 revenue per self-mined coin was ~$71.8k, not $62.5k. The earlier figure divided by total BTC, which included co-mining.
  • Why it matters: With bitcoin at ~$83.4k (2026-10-12) [38], the cash margin per coin is ~$44k on Q2 costs. Mining generates cash for the AI build-out even when the accounting shows a loss.
  • 😈 Devil's advocate:
    • Depreciation is real: rigs are written off over 2–3 years [1]. At ~$73k all-in cost, the margin over the cycle is thin.
    • The 2028 halving (~April 2028 [40]) halves revenue per EH/s overnight.
    • The cost split is my estimate. Bitdeer does not disclose self-mined BTC separately.
  • Verdict: Medium The cash edge is real, but full-cycle economics depend on bitcoin staying above ~$75k.

🐂 Bull #6Bitcoin and hashprice have recovered from the June lows

  • The point: Q2 was struck near the bottom of the cycle, and the mining backdrop is now clearly better.
  • Evidence:
    • Bitcoin closed June at $59,005. Hashprice hit a record low of $27.74/PH/day on 2026-06-06 [37].
    • On 2026-09-28: bitcoin $83,479, hashprice $39.87, network hashrate ~952 EH/s [36]. Bitcoin was ~$83.4k on 2026-10-12 [38].
    • Self-mining gross loss narrowed from −$34.4M (Q1) to −$2.7M (Q2) [1].
    • BTC mined: 2,694 in Q2; July 1,190 and August 1,310, against a Q2 average of ~898 a month [7,8].
  • Why it matters: Consensus FY2026 revenue was cut to $932.6M (from ~$1.03B in July) [30]. That implies H2 revenue of ~$515M against H1's $417.7M. A Q3 with ~3,800 BTC at higher prices could beat it (interpretation).
  • 😈 Devil's advocate:
    • Bitdeer holds almost no bitcoin: 61 BTC on 2026-08-31 [7], so it gains nothing on inventory.
    • It also owes BIT Group ~3,669 BTC (Bear #3), so a rally raises a debt.
    • Monthly updates have stopped. Q3 data arrive only with results [7].
  • Verdict: Medium This is a real near-term help, but it is a market variable.

🐂 Bull #7Insiders bought, a buyback started, and analysts stayed positive

  • The point: People with inside knowledge, and most of the sell side, think the shares are too cheap.
  • Evidence:
    • Open-market buys: CFO Potter 25,000 shares at $9.04 (2026-08-12) and CSO Basit 25,000 at $8.84 (2026-08-14), per Form 4s [21]. No insider sales in 3 months [29].
    • A new $100M buyback runs 2026-10-08 to 2028-10-07. The 2025 $40M programme was fully used [4].
    • Ratings [29,34,55]:
      • Cantor upgraded to Overweight on 2026-08-05 (PT $18). Correction to report #1, which called Cantor's last move a downgrade.
      • Barclays initiated Overweight on 2026-08-20 ($15) and raised to $17 on 2026-10-09.
      • Citizens JMP has $35.
      • 11 Buy, 2 Hold, 1 Sell.
  • Why it matters: Buying back stock at ~$9 retires shares at a price well below the ATM's $16 average in H1 ($484.5M / 30.4M shares) [1].
  • 😈 Devil's advocate:
    • The insider buys total ~$447k.
    • A buyback alongside a ~$990M unused ATM is contradictory: Bitdeer sold ~4.4M shares in Q3 [1] and now says it may buy shares back.
    • KBW cut its target to $10 (Market Perform) on Tydal credit [28]. Needham, B. Riley and Benchmark cut targets after Q2 [29].
    • The Northland $23 raise (2026-10-05) appears only on one aggregator [30].
  • Verdict: Weak The signal is small and mixed.

🐂 Bull #8A Big Four auditor and remediated controls improve reporting

  • The point: Moving from a small audit firm to Deloitte, with no open material weakness, makes the numbers more trustworthy.
  • Evidence:
    • MaloneBailey was dismissed on 2026-08-03. Deloitte & Touche (Singapore) was appointed for FY2026. No disagreements were reported [6].
    • The FY2025 20-F reports the FY2024 material weaknesses as remediated [11].
    • Bitdeer switched to US GAAP from 2026-01-01 [14].
  • Why it matters: It lowers the risk of a surprise restatement, and it matters to the lenders Bitdeer needs for project debt.
  • 😈 Devil's advocate:
    • The switch came one week before Q2 results, with a securities class action open [54], and no reason was given.
    • The US GAAP recast turned Q4-25 net income of +$70.5M into a −$191.5M loss [12]. The new framework shows much worse earnings.
  • Verdict: Weak to Medium. It is good for the long term, but the timing invites questions.

🐂 Bull #9Heavy short interest could squeeze on a financing win

  • The point: Many shares are sold short. A clean Volta loan close or Tydal handover could force short sellers to buy.
  • Evidence:
    • FINRA short interest on 2026-09-30: 61.2M shares, +6.2M on the previous print, 5.9 days to cover [33].
    • Reported as 22.5% of shares outstanding [33] or 26.9% of float [29]; the denominators differ.
    • The stock fell ~20% from $11.51 (09-25) to ~$9.19 (10-12) [32].
  • Why it matters: Positioning can magnify a good surprise.
  • 😈 Devil's advocate: Shorts may be hedging convertibles (the Feb-2026 notes convert at $9.93) and the ATM, not betting on collapse. Hedge-related shorts don't need to cover on good news.
  • Verdict: Weak

Bear case

🐻 Bear #1Cash burn outruns cash: about two quarters of runway

  • The point: Bitdeer spends roughly its whole cash balance every two quarters and relies on constant new funding.
  • Evidence:
    • H1 2026 cash flows [1]:
      • Operating cash flow −$505.4M, including $544.9M of wafer and rig purchases.
      • Proceeds from selling mined bitcoin +$402.4M, booked in investing.
      • Capex −$359.7M.
    • True H1 free cash flow ≈ −$463M (derived: −505.4 + 402.4 − 359.7), about −$231M a quarter.
    • Cash at 2026-06-30 was $456.8M, plus $39.4M restricted. Bitcoin held was 61 BTC (08-31) [1,7].
    • Commitments: ~$252.6M of PP&E commitments within 12 months, plus ~$500M of remaining Tydal capex [2,13].
    • Funding since: ATM net $58.0M (Jul–Sep), and ~$100M spent on Milam land [1,2]. No Tydal project debt has closed [2].
    • The company calls liquidity "sufficient… for at least the next 12 months" [2], but that relies on the ATM, borrowings and bitcoin sales.
  • Why it matters: Every quarter without project debt means more ATM selling near $9, or more related-party borrowing at 13%.
  • 🛡️ Counter-argument:
    • Much of the H1 burn was one-time wafer buying for the SEALMINER fleet. Q2 operating cash flow improved to −$158.5M from −$346.9M in Q1 [12,13].
    • Potter says Q2 wafer spend was funded by mining cash and wafer-backed debt, not equity [13].
    • The ATM is large and liquid (10–30M shares trade a day).
  • Severity: High Even after the wafer spike, Tydal capex alone exceeds cash.

🐻 Bear #2Dilution is relentless and the overhang is huge

  • The point: Each share's claim on the business keeps shrinking, and much more dilution is lined up.
  • Evidence:
    • Class A shares rose from 167.4M (2025-06-30) to 227.4M (2026-06-30), +35.8%. Class A + V rose from 211.8M to 271.8M, +28.3% [1].
    • Further dilution since: ~231.75M Class A by 2026-09-29 (derived: +4.37M ATM shares) [1].
    • Convertible overhang ≈ 88.0M shares, about 32% of A+V (derived from conversion rates) [1]:
      • Feb-2026 notes: 37.8M shares at $9.93, 8% above the $9.19 price.
      • Jun-2025 notes: 23.6M at $15.88.
      • Nov-2025 notes: 22.5M at $17.77.
      • Nov-2024 notes: 4.1M.
    • ATM: the $1B supplement (2026-08-10) is almost untouched (~$8.4M used by 09-29) [2,18]. At $9.19 that is ~108M shares, about 40% of A+V (derived).
    • Options/RSUs: 8.2M [18].
  • Why it matters: Consensus FY2027 EPS is −$0.37 [30]. Even if earnings turn positive, a 30–40% larger share count dilutes per-share value by the same amount.
  • 🛡️ Counter-argument:
    • Shares sold in H1 averaged ~$16, so past raises were done at good prices.
    • The $100M buyback (~10.9M shares at $9.19) can offset some dilution [4].
    • The convertibles don't mature before December 2029 (first holder put 2027-12-06 on $65M) [19].
  • Severity: High This is the factor that most directly destroys per-share value.

🐻 Bear #3The founder's other company is lender, custodian and counterparty

  • The point: BIT Group, which Jihan Wu also chairs, holds Bitdeer's crypto, lends it ~$0.6–0.7B at high rates, and sits on the other side of its bitcoin trades.
  • Evidence:
    • Related-party debt at 2026-06-30 was $633.1M, of which $491.0M is current [1].
    • BIT Assets loan: 9% plus a reference rate, 13.0% effective in H1. It amortises over 24 months per drawdown, with a loan-to-value covenant. $423.3M was outstanding at 09-29 [1,2]. A $45M one-week facility was added in August [2].
    • BIT Bitcoin Loan: 1,953.84 BTC at 06-30, rising to ~3,669 BTC at 09-29. It is revolving since August and backed by stablecoins (162.4M USDT pledged) [1,2]. At ~$83.4k, 3,669 BTC ≈ $306M (derived).
    • Related-party interest was $35.2M in H1, against $3.0M a year earlier [1].
    • BIT holds "substantially all" of Bitdeer's digital assets in custody. The 20-F warns Bitdeer could be a general unsecured creditor if BIT failed [11].
    • Wu's voting power was 69.5% at 2026-04-21 [11], ~66% on current counts (derived).
  • Why it matters:
    • About $490M of related-party debt falls due within 12 months.
    • The bitcoin loan means a bitcoin rally raises a debt (Interpretation: Bitdeer is effectively short ~3,669 BTC to BIT).
    • Minority shareholders cannot change the arrangement.
  • 🛡️ Counter-argument:
    • BIT was a lender of last resort when banks would not lend to miners. The terms are disclosed.
    • The BTC loan costs only ~3% and partly hedges the mined bitcoin.
    • Class V has not been sold, and the March-2025 fall from 48.4M to 44.4M was a conversion, not a sale [1]. Wu remains locked in.
  • Severity: Medium High. It has not caused a problem so far, but it concentrates refinancing and custody risk in one related party.

🐻 Bear #4Net debt is about 12.6 times trailing EBITDA

  • The point: Debt is large against what the business actually earns today.
  • Evidence:
    • Debt at 06-30 ≈ $1.87B: converts $1.215B principal, BIT $633M, bank loan $26M (10.31%, due Dec 2026) [1].
    • Net debt ≈ $1.42B (derived).
    • LTM adjusted EBITDA ≈ $113M: Q3-25 $43.0M + Q4-25 $24.3M (GAAP recast) + Q1-26 $14.4M + Q2-26 $31.1M [3,12,14,16].
    • H1 interest expense was $62.8M [1].
  • Why it matters: Leverage limits how much project debt lenders will add on Tydal. It also means equity absorbs any shortfall.
  • 🛡️ Counter-argument:
    • The converts are cheap (4–5.25%) and long-dated (2029–2032).
    • Needham forecasts FY2027 adjusted EBITDA of $585M [35]. On that figure leverage falls to ~2.4x, if Tydal and the bitcoin price cooperate.
  • Severity: Medium High. It is manageable only if 2027 EBITDA arrives on time.

🐻 Bear #5Mining lost money at the gross line in H1

  • The point: The core business, ~75% of revenue, did not cover its own costs including depreciation in H1.
  • Evidence:

    • Self-mining gross profit by quarter [1,12,16]:

      Quarter Gross profit Margin
      Q3-25 +$36.3M 27.7%
      Q4-25 +$5.8M 3.4%
      Q1-26 −$34.4M −23.4%
      Q2-26 −$2.7M −1.6%

    • Company gross margin −3.7% and operating margin −44.4% in Q2. H1 net loss $251.8M [1,13].

    • Fees are 0.70% of block rewards [36], so they won't offset the halving (~April 2028 [40]).

  • Why it matters: The cash engine that is supposed to fund AI was a cash user in H1, after counting the wafers needed to keep the fleet current.
  • 🛡️ Counter-argument:
    • H1 included the worst hashprice on record ($27.74 in June) [37].
    • At ~$40 hashprice and Q2 costs, mining should be clearly gross-positive in Q3 (interpretation).
    • Depreciation over 2–3 years is conservative.
  • Severity: High in a downturn, Medium at today's prices.

🐻 Bear #6Missed dates and a retracted press release hurt credibility

  • The point: Bitdeer's own timelines slip often, and its disclosure has become harder to follow.
  • Evidence:
    • Targets from October 2025 against later filings [3,17]:
      • Massillon full energisation was due Q1 2026. It is at 174 of 221 MW, with fire-damaged buildings rebuilt "by end Q3'26".
      • Fox Creek slipped one year to Q4 2027.
      • Wenatchee and Clarington are now "to be updated".
      • SEAL04 was ~one quarter late. SEAL04-2 has no date [12].
    • On 2026-10-08 [22,23,24]:
      • Bitdeer issued a release on a 60 MW "A103" at Cyberjaya, then told readers to disregard it.
      • About 50 minutes later it issued a different release: "A901", 67 MW, no city named.
      • "Secured" capacity fell from the 333.5 MW stated in the retracted release to 273.5 MW. No explanation has been given.
    • Monthly production updates ended with the August data. Infrastructure updates are now quarterly [7].
    • The auditor was changed one week before Q2 results [6].
  • Why it matters: Tydal (2026-12-31 / 2027-03-31) and Malaysia (Q1 2027) carry hard dates. Fewer and less reliable updates make slips harder to spot early.
  • 🛡️ Counter-argument:
    • The 2025 hashrate target was met (41.2 EH/s vs 40) [17].
    • The Tydal lease came about three months after "won't be too long" [12].
    • Weathersfield was pulled forward to Q4 2028 [3].
  • Severity: High Every bull argument depends on on-time delivery.

🐻 Bear #7Tydal's tenant is borrowing at 11% and has a year-10 exit

  • The point: The flagship lease depends on a months-old company paying for 16 years, while it holds a fee-free exit at year 10.
  • Evidence:
    • Volta was founded in 2026 and raised $300M at $2.4B [27].
    • The Anthropic link is unconfirmed by any party. Bitdeer says only "a leading AI lab" [9,25].
    • Bloomberg reports the end-customer deal runs ~6 years [60]. Volta can exit Tydal fee-free at year 10 [9].
    • Committed rent before the exit ≈ $2.95B, not $4.7B (derived: 10/16 of the total).
    • Volta's $5B loan: ~11% yield, unrated, fully amortising, commitments due 2026-10-14. It has not closed [25].
    • KBW: cover falls over 6 years, after which Bitdeer is "effectively exposed to Volta's standalone credit" [28].
  • Why it matters: If the loan fails to syndicate, the letter of credit and GPU funding are in doubt three months before Phase 1.
  • 🛡️ Counter-argument: JPMorgan structured the loan so that $1.3B of cash sits behind the letter of credit [25]. Hydro-powered, grid-connected Nordic capacity could be re-let.
  • Severity: High until the loan closes. It falls to Medium if it closes on schedule.

🐻 Bear #8Revenue rests on one pool, one country and one lease

  • The point: Several single points of failure sit under the revenue line.
  • Evidence:
    • "Customer B" was 51.6% of H1-26 revenue and "Customer C" 10.4% [1]. Interpretation: Customer B is probably the mining pool that pays out Bitdeer's rewards, so it is a counterparty risk more than a demand risk.
    • Bhutan provided $171.4M of H1 revenue, 41% of the total [1].
    • The Gedu lease with Druk Holding [48]:
      • Gives DHI no express obligation to supply power.
      • Ends with the power purchase agreement, whose expiry date was not found.
      • Can be terminated on a change of more than 50% of ownership.
    • Bhutan has no commitment to supply power in January–March [49].
    • Tydal is ~85% of Bitdeer's disclosed contracted revenue ($4.7B of ~$5.5B with A102; derived).
  • Why it matters: A dry winter in Bhutan or a pool failure hits a large share of revenue at once.
  • 🛡️ Counter-argument: Mining pools are interchangeable. Bhutan's hydro deal has run since 2023, and the government is a co-beneficiary.
  • Severity: Medium

🐻 Bear #9"Secured" Malaysia capacity has no customers yet

  • The point: Bitdeer signs long site agreements before it has tenants, which creates fixed obligations ahead of revenue.
  • Evidence:
    • A901 (67 MW): a 10-year data-centre services agreement. "No offtake commitments" as of 2026-10-08 [22].
    • A202 (65.1 MW): a 10-year agreement, with no cloud contract shown. A201 (21.7 MW) is still "in late negotiations" [7].
    • "Secured" means owned or covered by a signed services agreement. "Pipeline" (>$10B) is "not backlog… or a binding commitment" [22].
    • Interpretation: ~154 MW of the 273.5 MW "secured" figure is capacity Bitdeer committed to pay for without a customer.
  • Why it matters: At ~$25–30M of GPU capex per MW (estimate), filling 154 MW would cost ~$4B, far beyond current funding. Empty, it is a 10-year cost.
  • 🛡️ Counter-argument: A102 sold out quickly, and the GB300-ready design suits current demand. Prepayments of ~50% would reduce the equity needed.
  • Severity: Medium High. The 10-year commitments' payment terms are not disclosed (not found).

🐻 Bear #10Power sits in jurisdictions that are tightening rules

  • The point: Rules are moving against miners and large loads in several of Bitdeer's locations at once.
  • Evidence:
    • Ethiopia: EEP cut miners to 23% of contracted power on 2026-09-15 and will review again in October [50]. Bitdeer has 50 MW there, about $4.8M of H1 revenue [1].
    • Texas: approvals to energise new loads of 75 MW or more are paused. ERCOT reports to the PUCT are due 2026-12-10 [51].
    • Norway:
      • The Storting asked for a proposed ban on new crypto data centres [47].
      • No bill was found to have passed.
      • The committee says AI centres "must be clearly distinguished", which protects Tydal.
    • Ohio: the Clarington 570 MW site is tied up in the AHP suit, which is in discovery after a motion to dismiss was denied [13].
    • Securities class action: Sakar v. Bitdeer, SDNY, on SEAL04 statements. Its status was not found [54].
  • Why it matters: Each rule hits growth sites more than producing ones, which slows the AI conversion.
  • 🛡️ Counter-argument: Rockdale's 563 MW is already energised, so the Texas pause targets new loads [52]. Ethiopia is ~1% of revenue.
  • Severity: Medium

🐻 Bear #11The chip lead over Bitmain is within measurement error

  • The point: The "own chip" edge is too small to protect margins.
  • Evidence:
    • A4 Ultra Hydro: 9.45 J/TH, with ±5% tolerance on efficiency [46]. Bitmain's S23 Hydro: 9.50 J/TH [45].
    • SEAL04-2 has "no new timing" [12].
    • Rig sales fell to $0.4M in Q2 from $69.5M a year earlier [3].
  • Why it matters: If Bitdeer stops leading on efficiency, building its own rigs becomes a costly wafer-buying operation, not an advantage.
  • 🛡️ Counter-argument: Owning the design still avoids Bitmain's mark-up and supply allocation. US scrutiny of Bitmain could favour a Singapore-based supplier.
  • Severity: Low Medium.

Tailwinds NOT in the forecasts

Consensus FY2027 is revenue $1.56B and EPS −$0.37 (S&P via StockAnalysis, ~12–14 analysts) [30]. EPS is negative, so impacts are given in EBITDA per share. The reference is Needham's FY2027 adjusted EBITDA of $585M [35], ≈ $2.15 per share on 271.8M A+V shares. No consensus EBITDA was found, so Needham is used as a proxy. No FY2028 consensus was found, so FY2028 effects are compared with FY2027 consensus and labelled.

  • Malaysia A202/A901: offtake signed for half the capacity (~66 MW) — 20% chance within 36 months

    • Impact on revenue: +$1.11B a year in the first full year (FY2028), +71% vs FY2027 consensus ($1.56B)
    • Impact on EPS: +$1.80 EBITDA per share after funding dilution, +84% vs $2.15
    • How we got the number (estimate): A102 implies $800M ÷ 5 yrs ÷ 9.5 MW ≈ $16.8M/MW/yr.
      • Revenue: 66 MW × $16.8M ≈ $1.11B.
      • EBITDA: × 60% neocloud-style margin ≈ $665M.
      • Funding: GPU capex 66 × $27M ≈ $1.78B, half prepaid, so $0.89B of equity at $9.19 ≈ 97M new shares.
      • Per share: $665M ÷ 369M shares ≈ $1.80.
    • Expected value: 0.20 × $1.80 ≈ +$0.36 EBITDA per share
    • What it is: customers signed for capacity Bitdeer already committed to [7,22].
    • Why it's not priced in: the sites have no offtake. The >$10B pipeline is explicitly not backlog [22].
    • Trigger to watch: a release naming a contract value for A202 or A901, or A201 moving out of "late negotiations".
    • Why this probability: A102 filled fast and demand for GB300 capacity is strong. But two large contracts plus ~$1.8B of GPU funding within 36 months is a high bar for a company with ~$457M of cash.

  • Rockdale: a ~200 IT MW AI lease with a strong tenant — 20% chance of signing within 36 months

    • Impact on revenue: +$384M a year once delivered (first full year likely FY2029), +25% vs FY2027 consensus
    • Impact on EPS: +$1.20 EBITDA per share before any funding dilution, +56% vs $2.15
    • How we got the number (estimate):
      • Revenue: 200 MW × $160/kW/month × 12 ≈ $384M (US peer range $139–197).
      • EBITDA: × 85% ≈ $326M.
      • Per share: $326M ÷ 271.8M ≈ $1.20.
    • Expected value: 0.20 × $1.20 ≈ +$0.24 EBITDA per share
    • What it is: converting part of the 563 MW energised Rockdale site, plus the 200 new acres, to AI colocation [2,13].
    • Why it's not priced in: management has made no decision and given no timing. Analysts value the AI side on Tydal and Malaysia [13,28].
    • Trigger to watch: an ERCOT/PUCT ruling after 2026-12-10, a Rockdale design or tenant announcement.
    • Why this probability: energised Texas power is scarce and valuable. But the large-load pause, the lack of any decision, and US tenants' preference for investment-grade landlords hold it down.

  • Bitcoin averages about $100k through FY2027 — 25% chance

    • Impact on revenue: +$227M, +15% vs FY2027 consensus
    • Impact on EPS: +$0.83 EBITDA per share, +39% vs $2.15
    • How we got the number (estimate):
      • Coins: ~1,140 self-mined BTC a month (87% of August's 1,310) × 12 ≈ 13,700 BTC, with fleet growth assumed to offset network growth.
      • Revenue: × $16.6k uplift (from $83.4k) ≈ $227M.
      • Margin: ~100% incremental, because power cost is fixed.
      • Per share: ÷ 271.8M ≈ $0.83.
    • Expected value: 0.25 × $0.83 ≈ +$0.21 EBITDA per share
    • What it is: a market move back toward the October 2025 high ($126k) [38].
    • Why it's not priced in: consensus was cut after the June low. The bitcoin-price assumption is not published, so this rests on the assumption that it is near spot.
    • Trigger to watch: hashprice above $50/PH/day.
    • Why this probability: a +20% move in a year is common for bitcoin, but it needs a sustained average, not a peak. That case is weaker with bitcoin ~28% below a year ago.

  • Tydal's retained ~43 IT MW leased on the same terms — 30% chance by end-2027

    • Impact on revenue: +$104M a year (first full year FY2028), +6.6% vs FY2027 consensus
    • Impact on EPS: +$0.34 EBITDA per share, +16% vs $2.15
    • How we got the number (estimate):
      • Revenue: 42.7 MW × $202 × 12 ≈ $104M.
      • EBITDA: × 90% ≈ $93M.
      • Per share: ÷ 271.8M ≈ $0.34.
    • Expected value: 0.30 × $0.34 ≈ +$0.10 EBITDA per share
    • What it is: the 47 gross MW (A401) Bitdeer kept for its own AI cloud, leased to Volta or another tenant instead [7,9].
    • Why it's not priced in: it is not in the $4.7B lease. Management said only that Volta's demand "can be much larger" [13].
    • Trigger to watch: a lease amendment or a second Tydal release.
    • Why this probability: the tenant already sits on site and wants more. But Bitdeer has said it plans to use the space itself, and Volta's financing is not yet closed.

Headwinds NOT in the forecasts

  • Bitcoin averages about $60k through FY2027 — 20% chance

    • Impact on revenue: −$321M, −21% vs FY2027 consensus
    • Impact on EPS: −$1.18 EBITDA per share, −55% vs $2.15
    • How we got the number (estimate): 13,700 BTC × −$23.4k ≈ −$321M; power cost is fixed, so ≈ −$321M EBITDA; ÷ 271.8M ≈ −$1.18.
    • Expected value: 0.20 × −$1.18 ≈ −$0.24 EBITDA per share
    • What it is: a return to the June 2026 level ($59k close) [37].
    • Why it's not priced in: forecasts implicitly use near-spot prices. A fall would also force equity raises at lower share prices.
    • Trigger to watch: hashprice below $32/PH/day, or large miners switching off.
    • Why this probability: bitcoin was at this level just four months ago. But a full-year average that low needs a renewed bear market.

  • More than 60M new shares within 12 months (ATM or new equity-linked debt) — 55% chance

    • Impact on revenue: none directly (0%)
    • Impact on EPS: −$0.39 EBITDA per share, −18% vs $2.15
    • How we got the number (estimate): $585M ÷ 271.8M = $2.15 → $585M ÷ 331.8M = $1.76 → −$0.39. This assumes the proceeds earn nothing extra within FY2027.
    • Expected value: 0.55 × −$0.39 ≈ −$0.21 EBITDA per share
    • What it is: using about $550M of the ~$990M ATM at ~$9, or a new convertible [2,18].
    • Why it's not priced in: per-share forecasts use the current share count. Management guides to "non-dilutive" project debt [13].
    • Trigger to watch: the Q3 6-K share count (~mid-November), a new 424B5, or a convertible launch.
    • Why this probability: H1 free cash flow was −$463M, and Tydal capex is still unfunded. Bitdeer has issued equity every quarter in 2026, but a Tydal project-debt closing would reduce the need.

  • Tydal Phases 1 and 2 slip by one quarter — 35% chance

    • Impact on revenue: −$73M in FY2027, −4.7% vs consensus
    • Impact on EPS: −$0.24 EBITDA per share, −11% vs $2.15
    • How we got the number (estimate): run-rate $293M a year (121 MW × $202 × 12) ÷ 4 ≈ $73M; × 90% ≈ $66M; ÷ 271.8M ≈ −$0.24.
    • Expected value: 0.35 × −$0.24 ≈ −$0.08 EBITDA per share
    • What it is: handover after 2026-12-31 (66.5 MW) and 2027-03-31 (66.5 MW) [9,13].
    • Why it's not priced in: consensus assumes Tydal rent from 2027 [28,30].
    • Trigger to watch: Q3 results commentary, Volta loan closing, the EPC schedule.
    • Why this probability: Bitdeer's record shows slips at most sites. But this EPC is open-book cost-plus, and the tenant is financing its own GPUs to the same dates.

  • BIT Group stops rolling the related-party loans, forcing an equity refinancing — 15% chance within 24 months

    • Impact on revenue: none (0%)
    • Impact on EPS: −$0.31 EBITDA per share, −14% vs $2.15
    • How we got the number (estimate): $423M ÷ $9.19 ≈ 46M shares; $585M ÷ 317.8M = $1.84 → −$0.31.
    • Expected value: 0.15 × −$0.31 ≈ −$0.05 EBITDA per share
    • What it is: the 24-month amortising BIT Assets loan ($423.3M at 09-29) not being replaced by new drawdowns [2].
    • Why it's not priced in: analysts rarely discuss the related-party structure.
    • Trigger to watch: BIT loan balances in the Q3 6-K, or a loan-to-value top-up.
    • Why this probability: BIT has added facilities every quarter, including in August. The founder controls both sides, so a sudden cut-off is unlikely.

  • Bhutan power cut beyond the normal winter pattern (25% of Bhutan revenue for one quarter) — 20% chance in FY2027

    • Impact on revenue: −$21M, −1.4% vs FY2027 consensus
    • Impact on EPS: −$0.06 EBITDA per share, −3% vs $2.15
    • How we got the number (estimate): H1 Bhutan revenue $171.4M ÷ 2 ≈ $86M a quarter × 25% ≈ $21M; × 70% margin ≈ $15M; ÷ 271.8M ≈ −$0.06.
    • Expected value: 0.20 × −$0.06 ≈ −$0.01 EBITDA per share
    • What it is: a dry season or a policy change. DHI has no express obligation to supply power [48,49].
    • Why it's not priced in: seasonal patterns are probably in forecasts, but a deeper cut is not.
    • Trigger to watch: Bhutanese press on winter allocation, and Gelephu progress.
    • Why this probability: Jigmeling already ran at 424 of 500 MW because of a substation limit, and Bhutan imports winter power.

Investment Risk Score: 71/100 (high)

# Factor Weight Measured value Score Why
1 Liquidity & funding runway 15% Cash $456.8M (06-30); H1 FCF ≈ −$463M (op CF −$505.4M + BTC sales $402.4M − capex $359.7M) → ~2 quarters of runway without new money; ~$500M Tydal capex unfunded; ~$990M ATM unused; management says "sufficient" for 12 months; no going-concern language 70 Runway under 12 months on internal cash alone. It stays out of the 81+ band because the ATM is large and liquid and there is no going-concern or covenant language.
2 Leverage 10% Debt ≈ $1.87B (converts $1.215B, BIT $633M, bank $26M); net debt ≈ $1.42B; LTM adj EBITDA ≈ $113M → ~12.6x; H1 interest $62.8M; BIT loan 13% effective 80 Far above 5x. It stays just short of 81+ because EBITDA is positive and the converts don't mature before Dec 2029.
3 Dilution (12 months + overhang) 10% Class A +35.8% (A+V +28.3%) in 12 months; convertible overhang ~88M (32%); ATM ~108M shares at $9.19 (40%); options 8.2M; $100M buyback 90 Over +25% realised, plus an overhang that could add a further third or more.
4 Profitability 10% Q2 gross margin −3.7%, operating margin −44.4%; H1 net loss $251.8M; FCF ≈ −100% of revenue 80 Deep losses, improving from Q1 (gross margin −20.7%). Adjusted EBITDA is positive ($31.1M in Q2), so not the extreme band.
5 Revenue quality & concentration 10% ~84% variable mining revenue; Customer B 51.6% of H1 revenue (probably the mining pool); Bhutan 41% of H1 revenue; Tydal ~85% of disclosed contracted revenue 72 More than 35% from one counterparty and one country. The pool is replaceable, so it scores below 81.
6 Execution & track record 10% Massillon, Fox Creek, Wenatchee and Clarington slipped; SEAL04 late, SEAL04-2 no date; A103 retracted and replaced the same evening; monthly updates stopped; auditor changed a week before Q2; no open material weakness or restatement 72 Frequent delays and a disclosure misstep, but no control failure.
7 Competitive position & disruption 10% A4 9.45 J/TH vs S23 9.50 (±5% tolerance); Tydal rent at the top of peer range; AI cloud subscale with negative gross margin; peers have investment-grade tenants 55 Contested: a power edge and an integrated rig supply against bigger, better-financed AI peers.
8 Cyclicality & commodity/macro sensitivity 10% ~84% of revenue tied to bitcoin; hashprice $27.74 (June low) → $39.87; ~$0.83 EBITDA per share per +$16.6k BTC (estimate); halving ~April 2028 78 Close to a one-commodity business. It survived June at $59k bitcoin, so survival does not hinge on one price level.
9 Regulatory, legal & geopolitical 5% Bhutan 41% of revenue under a lease with no power obligation; Ethiopia 23% deliveries; Texas large-load pause; Norway ban in preparation; AHP suit in discovery; SDNY class action; related-party custody 60 Material across many fronts, but none existential on its own.
10 Valuation & expectations 10% EV ~$3.99B ≈ 4.9x LTM revenue vs peers 17–56x; price $9.19 vs targets $22–23; short interest 22.5–26.9%, 5.9 days to cover; −20% in two weeks 50 Cheap against peers, but short interest is heavy and the story depends on financing.

Overall: weighted average = 0.15×70 + 0.10×80 + 0.10×90 + 0.10×80 + 0.10×72 + 0.10×72 + 0.10×55 + 0.10×78 + 0.05×60 + 0.10×50 = 10.5 + 8.0 + 9.0 + 8.0 + 7.2 + 7.2 + 5.5 + 7.8 + 3.0 + 5.0 = 71.2 → 71.

Change vs last: +4 vs 67 (old 6-area method).

  • Part is the method: v3 gives dilution, leverage and profitability their own weights, and all three sit at the bad end here.
  • Part is new evidence:
    • The H1 cash statement shows a ~−$463M free cash outflow once bitcoin-sale proceeds are counted.
    • The BTC loan grew from ~1,954 to ~3,669 BTC.
    • The A103 retraction.
  • Valuation risk fell. The stock is ~$9.19, and the peer gap is now measured.

Ranking check: BTDR sits between WYFI (70) and GRRR (72).

  • It is riskier than WYFI because:
    • Its burn is ~$230M a quarter against ~$457M of cash.
    • Net debt is ~12.6x EBITDA.
    • ~84% of revenue rides on bitcoin.
  • It is safer than GRRR because it has:
    • 1.75 GW of producing power.
    • A filed $4.7B lease about to gain cash-backed credit support.
    • A deep, liquid market for its shares.
    • No convertible reset.

Conclusion

  • What you are really betting on: that Bitdeer converts power into contracted AI rent (Tydal first, then Malaysia or Rockdale) before cash burn and ATM selling hand most of the value to new shareholders and to BIT Group. Bitcoin staying above ~$75k is the bridge.
  • The 2–3 things that will decide the outcome:
    1. The Volta $5B loan closing (commitments due 2026-10-14), followed by Tydal Phase 1 handover on 2026-12-31.
    2. Whether Tydal project debt closes before the ATM is drawn down heavily. The Q3 share count is the tell.
    3. The bitcoin price into the 2028 halving.
  • Signposts:
    • Bull confirmed by:
      • The Volta loan closing at or near the marketed terms.
      • A Tydal project-debt facility of ~$400M or more.
      • Self-mining gross profit clearly positive in Q3.
      • A named offtake for A201, A202 or A901.
    • Bull broken by:
      • Volta's loan stalling or being re-priced.
      • Q3 Class A shares above ~250M.
      • A new convertible before project debt closes.
      • The BIT loan balance rising again, or a Tydal date moving.
  • New since the last report:
    • The real free-cash-flow burn is ~−$463M in H1 once bitcoin sales are counted. The BTC loan nearly doubled to ~3,669 BTC.
    • JPMorgan's $5B Volta loan includes a $1.3B tranche that cash-backs Bitdeer's letter of credit. Commitments are due 10-14.
    • A103 was retracted and replaced by A901 the same evening. "Secured" Malaysia capacity has ~154 MW without customers.
    • Monthly updates have ended. Deloitte replaced MaloneBailey. Cantor is Overweight, not Neutral. A $100M buyback was launched alongside the ATM.

Open questions & data gaps

  • Q3 numbers: cash and share count at 2026-09-30 are not found. Results are expected ~2026-11-09 (unconfirmed). This matters most for factors 1 and 3.
  • Volta loan outcome after 10-14: not yet known. It decides Bear #7's severity.
  • Self- vs co-mined BTC split: not disclosed. Cost per coin is my estimate.
  • Malaysia payment terms: the 10-year A202 and A901 commitments are not disclosed. They could make Bear #9 worse.
  • Why A103 was retracted: no explanation found.
  • Litigation: status of the class action (lead plaintiff, motion to dismiss) and an AHP trial date. Not found; PACER and the Ohio docket were not accessible.
  • Bhutan: PPA expiry dates and the Jigmeling lease terms. Not found.
  • Rig tariffs: the current rate on Southeast Asian rig imports. Not found from primary sources.
  • Analyst notes: Barclays, Northland and Citizens rationales are paywalled. The Northland $23 raise rests on a single aggregator.
  • Call transcripts: the Q4 2025 call was not available. Q1 and Q2 were read from third-party transcripts (Motley Fool), not company PDFs.
Sources
  1. Unaudited Interim Condensed Consolidated Financial Statements, six months ended 2026-06-30 (6-K Ex. 99.1) — SEC EDGAR — 2026-09-29 — https://www.sec.gov/Archives/edgar/data/1899123/000121390026104638/ea030647301ex99-1.htm
  2. Recent Developments (6-K Ex. 99.2) — SEC EDGAR — 2026-09-29 — https://www.sec.gov/Archives/edgar/data/1899123/000121390026104638/ea030647301ex99-2.htm
  3. Bitdeer Reports Unaudited Financial Results for Q2 2026 (6-K Ex. 99.1) — SEC EDGAR — 2026-08-10 — https://www.sec.gov/Archives/edgar/data/1899123/000121390026086938/ea030131301ex99-1.htm
  4. 6-K: US$100M share repurchase programme — SEC EDGAR — 2026-10-08 — https://www.sec.gov/Archives/edgar/data/1899123/000121390026107955/ea0307821-6k_bitdeer.htm
  5. 6-K: resignation of director Sheldon Trainor-DeGirolamo — SEC EDGAR — 2026-10-02 — https://www.sec.gov/Archives/edgar/data/1899123/000121390026106360/ea0307497-6k_bitdeer.htm
  6. 6-K: change in certifying accountant (MaloneBailey → Deloitte) — SEC EDGAR — 2026-08-03 — https://www.sec.gov/Archives/edgar/data/1899123/000121390026084288/ea0299664-6k_bitdeer.htm
  7. Bitdeer August 2026 Production and Operations Update (6-K Ex. 99.1) — SEC EDGAR — 2026-09-16 — https://www.sec.gov/Archives/edgar/data/1899123/000121390026100537/ea030523501ex99-1.htm
  8. Bitdeer July 2026 Production and Operations Update — SEC EDGAR — 2026-08-26 — https://www.sec.gov/Archives/edgar/data/1899123/000121390026093742/ea030344101ex99-1.htm
  9. Bitdeer Announces $4.7 Billion, 16-Year AI/HPC Data Center Lease for Tydal (6-K Ex. 99.1) — SEC EDGAR — 2026-08-04 — https://www.sec.gov/Archives/edgar/data/1899123/000121390026085655/ea030075201ex99-1.htm
  10. Bitdeer Milam County land acquisition release — SEC EDGAR — 2026-09-01 — https://www.sec.gov/Archives/edgar/data/1899123/000121390026096425/ea030402101ex99-1.htm
  11. Bitdeer Form 20-F for FY2025 (keyword-searched: Items 6, 7, 15, 16F, risk factors) — SEC EDGAR — 2026-04-30 — https://www.sec.gov/Archives/edgar/data/1899123/000121390026049770/ea0282628-20f_bitdeer.htm
  12. Bitdeer Q1 2026 Earnings Call Transcript — Motley Fool — 2026-05-14 — https://www.fool.com/earnings/call-transcripts/2026/05/14/bitdeer-btdr-q1-2026-earnings-transcript/
  13. Bitdeer Q2 2026 Earnings Call Transcript — Motley Fool — 2026-08-10 — https://www.fool.com/earnings/call-transcripts/2026/08/10/bitdeer-btdr-q2-2026-earnings-call-transcript/
  14. Bitdeer Reports Unaudited Financial Results for Q1 2026 (6-K Ex. 99.1) — SEC EDGAR — 2026-05-14 — https://www.sec.gov/Archives/edgar/data/1899123/000121390026056587/ea029029701ex99-1.htm
  15. Bitdeer Reports Unaudited Financial Results for Q4 and FY2025 — SEC EDGAR — 2026-02-12 — https://www.sec.gov/Archives/edgar/data/1899123/000121390026015251/ea027678001ex99-1_bitdeer.htm
  16. Bitdeer Reports Unaudited Financial Results for Q3 2025 — SEC EDGAR — 2025-11-10 — https://www.sec.gov/Archives/edgar/data/1899123/000121390025107865/ea026473801ex99-1_bitdeer.htm
  17. Bitdeer October 2025 Production and Operations Update — SEC EDGAR — 2025-11-10 — https://www.sec.gov/Archives/edgar/data/1899123/000121390025107866/ea026473802ex99-1_bitdeer.htm
  18. Form 424B5 prospectus supplement (US$1B ATM) — SEC EDGAR — 2026-08-10 — https://www.sec.gov/Archives/edgar/data/1899123/000121390026087060/ea030073702-424b5_bitdeer.htm
  19. Form 424B5 prospectus supplement (Feb 2026 notes, convertible table incl. put dates) — SEC EDGAR — 2026-02 — https://www.sec.gov/Archives/edgar/data/1899123/000121390026019332/ea0277924-424b5_bitdeer.htm
  20. Schedule 13D/A No. 13 (Tether) — SEC EDGAR — 2026-06-30 — https://www.sec.gov/Archives/edgar/data/1899123/000110465926079298/xslSCHEDULE_13D_X02/primary_doc.xml
  21. Form 4s: Potter (2026-08-12, 2026-09-01) and Basit (2026-08-14) — SEC EDGAR — 2026-08/09 — https://www.sec.gov/Archives/edgar/data/1899123/000121390026088668/ownership.xml
  22. Bitdeer AI Expands Southeast Asia Footprint with A901, a New 67MW AI Data Center Campus in Malaysia — GlobeNewswire via Finviz — 2026-10-08 — https://finviz.com/news/400718/bitdeer-ai-expands-southeast-asia-footprint-with-a901-a-new-67mw-ai-data-center-campus-in-malaysia-active-ai-cloud-pipeline-exceeds-10-billion
  23. Bitdeer AI Secures 60MW A103 Facility at Its Existing Cyberjaya, Malaysia Campus (retracted) — GlobeNewswire via Finviz — 2026-10-08 — https://finviz.com/news/400675/bitdeer-ai-secures-60mw-a103-facility-at-its-existing-cyberjaya-malaysia-campus-reaches-approximately-95-of-its-up-to-350mw-target-capacity
  24. Notice to Disregard — Bitdeer Technologies Group — GlobeNewswire via Finviz — 2026-10-08 — https://finviz.com/news/400712/notice-to-disregard-bitdeer-technologies-group
  25. JPMorgan Dangles 11% Yield on Volta's $5 Billion AI Loan Sale (Bloomberg) — Yahoo Finance — 2026-10-05 — https://finance.yahoo.com/technology/ai/articles/jpmorgan-dangles-11-yield-volta-200522405.html
  26. Bitdeer stock pops and drops on $4.7B AI data center lease (incl. Benchmark estimates) — The Block — 2026-08-04 — https://theblock.co/post/410578/bitdeer-stock-pops-drops-4-7-billion-ai-data-center-lease-norway
  27. JPMorgan leads a $5B debt package for Volta's AI data centre buildout — The Next Web — 2026-08-27 — https://thenextweb.com/news/jpmorgan-volta-5b-debt-package-norway-tydal-anthropic
  28. KBW cuts Bitdeer target on Tydal credit risk (search summary only) — Blockspace — 2026-08 — https://blockspace.media/insight/kbw-cuts-bitdeer-target-tydal-credit-risk/
  29. BTDR analyst forecast, ratings table and insider summary — MarketBeat — 2026-10-12 — https://www.marketbeat.com/stocks/NASDAQ/BTDR/forecast/
  30. BTDR forecast (consensus FY2026–27, rating changes) — StockAnalysis — 2026-10-12 — https://stockanalysis.com/stocks/BTDR/forecast/
  31. BTDR and peer statistics (EV, EV/sales; IREN, CIFR, WULF, HUT, CORZ, CANG, CAN) — StockAnalysis — 2026-10-12 — https://stockanalysis.com/stocks/btdr/statistics/
  32. BTDR price history — StockAnalysis — 2026-10-12 — https://stockanalysis.com/stocks/btdr/history/
  33. BTDR short interest (FINRA 2026-09-30 settlement) — Equibles — 2026-10-12 — https://equibles.com/stocks/btdr/shortinterest
  34. Barclays initiates Bitdeer at Overweight, $15 target — Crypto Briefing — 2026-08-20 — https://cryptobriefing.com/barclays-initiates-bitdeer-overweight-15-target/
  35. Needham price target and estimates after Tydal lease (search summary only) — Blockspace — 2026-08 — https://blockspace.media/insight/bitdeer-needham-price-target-tydal-lease-2026/
  36. Hashrate Index Roundup (September 28, 2026) — Hashrate Index — 2026-09-28 — https://hashrateindex.com/blog/hashrate-index-roundup-september-28-2026/
  37. Luxor Hashrate Lookback, June 2026 — Hashrate Index — 2026-07 — https://hashrateindex.com/blog/luxor-hashrate-lookback-series-june-2026/
  38. Current price of Bitcoin for October 12, 2026 — Fortune — 2026-10-12 — https://fortune.com/article/price-of-bitcoin-10-12-2026/
  39. Bitcoin difficulty eases 0.03% at block 969,696 — Coinotag — 2026-10-03 — https://en.coinotag.com/bitcoin-btc-difficulty-eases-0-03-block-969696
  40. Bitcoin halving countdown — CoinWarz — 2026 — https://www.coinwarz.com/mining/bitcoin/halving
  41. Bernstein maintains $36 TeraWulf target after $19B, 20-year Anthropic lease — The Block — 2026-07-07 — https://www.theblock.co/post/407429/bernstein-maintains-36-terawulf-target-after-19-billion-20-year-anthropic-lease
  42. Applied Digital inks $7.5B lease for 300 MW at Polaris Forge 3 — TheEnergyMag — 2026 — https://theenergymag.com/news/market-news/applied-digital-inks-7-5-billion-lease-for-300-mw-at-polaris-forge-3-expanding-ai-factory-footprint
  43. Galaxy Digital Q2 2026 / CoreWeave lease — TheEnergyMag — 2026 — https://theenergymag.com/news/market-news/galaxy-digital-reports-85-million-q2-2026-net-loss-and-526-mw-coreweave-lease
  44. Hut 8 signs 245MW capacity deal with Fluidstack (search summary only) — DatacenterDynamics — 2026 — https://www.datacenterdynamics.com/en/news/hut-8-signs-245mw-capacity-deal-with-fluidstack-as-part-of-multi-gigawatt-partnership-with-anthropic
  45. Antminer S23 Hydro specifications — Hashrate Index — undated — https://hashrateindex.com/rigs/antminer-s23-hydro
  46. Bitdeer Launches SEALMINER A4 Series, 9.45 J/TH — GlobeNewswire via AAP — 2026-04-07 — https://www.aap.com.au/aapreleases/globenewswire9685251/
  47. Innst. 140 S (2025–2026), energy and environment committee report on data centres — Stortinget — 2026-02-24 — https://www.stortinget.no/globalassets/pdf/innstillinger/stortinget/2025-2026/inns-202526-140s.pdf
  48. Land Lease Agreement, Bitdeer Gedu and Druk Holding & Investments (6-K Ex. 10.1) — SEC EDGAR — 2023-08-11 — https://www.sec.gov/Archives/edgar/data/1899123/000114036123040877/brhc20057629_ex10-1.htm
  49. The 500 MW Jigmeling Data Center owned by Bitdeer — The Bhutanese — 2026-01-17 — https://thebhutanese.bt/?p=41944
  50. Ethiopia cuts Bitcoin miners' power amid hydro shortage (citing Bloomberg) — Cointelegraph — 2026-09-16 — https://cointelegraph.com/news/ethiopia-bitcoin-mining-power-cut-hydro-shortage
  51. Batch Zero Update — ERCOT — 2026-09-11 — https://www.ercot.com/files/docs/2026/09/11/14-Batch-Zero-Update.pdf
  52. Texas Large Load Interconnection Update: ERCOT Batch Zero Pause — Baker Botts — 2026-08 — https://www.bakerbotts.com/thought-leadership/publications/2026/august/texas-large-load-interconnection-update---ercot-batch-zero-pause-and-verification-process
  53. Texas PUC adopts data center interconnection rules — Utility Dive — 2026-09-21 — https://utilitydive.com/news/texas-puc-adopts-data-center-interconnection-rules/830899
  54. Kessler Topaz announces securities class action against Bitdeer (Sakar v. Bitdeer, SDNY 25-cv-10069) — Business Wire — 2025-12-12 — https://www.businesswire.com/news/home/20251212567812/en/NASDAQ-BTDR-Kessler-Topaz-Meltzer-Check-LLP-Announces-the-Filing-of-a-Securities-Class-Action-Lawsuit-Against-Bitdeer-Technologies-Group
  55. Cantor Fitzgerald upgrades Bitdeer to Overweight, PT $18 (search summaries of TheFly/Benzinga; note not read) — secondary — 2026-08-05 — https://stockanalysis.com/stocks/BTDR/forecast/
  56. Bitdeer Secures $400M Malaysia AI Contract, Targets 350 MW by 2028 — TheEnergyMag — 2026-08-20 — https://theenergymag.com/news/market-news/bitdeer-malaysia-ai-cloud-contract-350-megawatt
  57. Bitcoin and ethereum prices today, Monday, October 12, 2026 — Yahoo Finance — 2026-10-12 — https://finance.yahoo.com/personal-finance/investing/article/bitcoin-and-ethereum-prices-today-monday-october-12-2026-how-bond-yields-threaten-crypto-prices-114018710.html
  58. IREN Q4 FY2026 results summary — KuCoin News — 2026-08 — https://www.kucoin.com/news/flash/iren-q4-fy2026-revenue-meets-expectations-ai-cloud-revenue-doubles
  59. Schedule 13D/A No. 2 (Shinning Stone) — SEC EDGAR — 2026-04-20 — https://www.sec.gov/Archives/edgar/data/1899123/000121390026045581/xslSCHEDULE_13D_X02/primary_doc.xml
  60. Anthropic Deal Underpins Bitdeer's $4.7 Billion Norway AI Lease (relays Bloomberg) — TheEnergyMag — 2026-08 — https://theenergymag.com/news/market-news/bitdeer-announces-4-7-billion-16-year-ai-hpc-data-center-lease-for-tydal-norway-campus
All BTDR Bulls vs Bears reports
Disclaimer. AI-generated research and podcast (Claude by Anthropic, voices by Google Gemini) for educational purposes — not financial advice and not a recommendation to buy or sell. Willem (MR.INVEST) owns this stock. Check the company’s own filings before making any decision.
← All Bulls vs Bears reports