To The Moon

$MARA Just Sold 15,000 BTC. Let Me Break Down The Math.

Hey everyone. Long time no post — been watching, accumulating, and waiting for something worth writing about. This is it.

For those of you holding MARA or watching it from the sidelines, yesterday was a big day. News dropped premarket, stock flew 10%, then gave back most of it by close. A lot of noise, a lot of headlines. Let me cut through it and give you the actual numbers.

What Happened

Between March 4 and March 25, MARA quietly sold 15,133 BTC for approximately $1.1 billion. They used the proceeds to repurchase $367.5 million of their 2030 convertible notes and $633.4 million of their 2031 convertible notes — both at roughly 9% below face value. CoinDesk

That 9% discount translates to over $88 million in immediate cash savings — a billion dollars in debt wiped off the books, no new shares issued. FinancialContent

Total convertible debt drops from $3.3 billion to $2.3 billion. MARA now holds 38,689 BTC. CoinDesk

Simple version: they sold Bitcoin to kill debt. And they killed it cheap.

The P&L — Let's Be Honest About It

Average sale price was ~$65,300 per coin. Estimated blended cost basis sits near $80,900 per BTC. Crypto Times

That's a $15,600 loss per coin. Across 15,133 BTC — roughly $236 million in realized losses on the Bitcoin itself.

Now the mining economics on top of that:

Electricity alone costs MARA about $38,956 per Bitcoin. Add hosting and operations and you're at $70,027 per BTC. Factor in SG&A and the all-in cost runs $110,000–$113,000 per BTC. Crypto News

So they sold at $65,300 against an all-in cost of $110,000+. On the surface that looks terrible.

But here's where the math gets interesting. They didn't sell BTC to make money on BTC. They sold BTC to buy back debt at a 9% discount to face value. That $88 million gain from the debt repurchase is real, risk-free, and immediate. You're essentially converting a depreciating asset position into a guaranteed book gain on the liability side.

Net-net, the balance sheet came out ahead. That's the trade.

Why They Did This

A few things converging at once.

Mining margins are getting murdered. The weighted average cash cost to produce one bitcoin among publicly listed miners hit ~$79,995 in Q4 2025. Hashprice fell to post-halving lows near record levels, with a roughly 31% drop in Bitcoin from the October 2025 ATH of ~$125,000 down to ~$86,000 by late December. The Block By March, BTC was sitting at $65K-$69K. The math just doesn't work at those levels.

The convertible note overhang was a ticking time bomb. This is the one most people don't talk about enough. Convertible bonds can convert into equity. If those notes converted during a period of share price weakness, MARA shareholders would have gotten massively diluted. By selling Bitcoin near $65,300, MARA secures non-dilutive liquidity to retire longer-dated convertible notes, reducing future dilution risk from potential conversions. Crypto Times That threat is now dead for $1 billion worth of notes.

And here's the part most people completely missed on why the stock spiked so hard premarket:

When institutions buy convertible bonds, a lot of them don't just sit on them for yield. They run a delta-hedging strategy — essentially writing covered calls or shorting the underlying stock against their bond position to extract yield from the volatility. Think of it like a covered call on MARA stock, funded by the bond. When MARA announced they were buying back those notes, those bondholders had to unwind their short hedges — meaning they had to buy back MARA shares they were short. Forced short covering on top of the fundamental good news. That's why you saw a violent +10% move premarket. It wasn't just people buying the headline. It was bond desks scrambling to cover. When you understand the mechanics, you trade it very differently from the guy just reading the news.

The AI pivot needs capital. MARA has a 2.5-gigawatt AI data center joint venture with Starwood Capital Group, converting legacy mining sites into hyperscale facilities. FinancialContent That doesn't fund itself. Selling BTC from treasury is the cleanest way to do it without going back to shareholders with another equity raise.

How The Rest Of The Miners Are Playing It

Because context matters.

Riot (RIOT) — Still sitting on 18,005 BTC. Bitcoin Treasuries Mostly HODLing, playing the lowest-cost pure-play miner angle. Their AMD data center lease went operational in January 2026, with a nearly 2-gigawatt power portfolio as the crown jewel. 24/7 Wall St. Respectable strategy but they're feeling the same margin squeeze.

CleanSpark (CLSK) — Mined 568 BTC in February, sold 553 of them. BitcoinEthereumNews.com Almost everything they mine goes straight out the door to fund operations. That tells you everything about how tight margins are for the smaller players.

Core Scientific (CORZ) — Expects to monetize substantially all of its Bitcoin holdings in 2026 as part of its transition to AI and high-density colocation. Bitcoin Magazine Already arguably the furthest along with $10B+ in AI contracts. MARA is chasing them.

The big picture: since October 2025, publicly listed miners have collectively sold more than 15,000 BTC. Glassnode data shows miners collectively selling more coins than they produce on a 30-day basis. BitcoinEthereumNews.com

The "mine and HODL forever" era is done. Every major miner is adapting or dying. MARA just made the most decisive single move to date.

Why It Went Up, Why It Faded

Premarket: +10%. Close: +3.6%. What happened in between?

The spike I already explained — fundamental good news plus forced short cover unwind from the convertible bond delta hedges. Double-barrel catalyst.

The fade: pure macro sitting on top of everything.

The US and Israel launched joint air strikes on Iran on February 28. The Strait of Hormuz has been effectively closed to most shipping traffic since. IEA We're nearly a month into an active war that is directly spiking energy prices.

On March 26 — the exact same day MARA's news dropped — Brent crude jumped more than 5% to $107 per barrel. WTI hit $94. FinancialContent

Think about what that means for MARA specifically. Their all-in mining cost is already $110K+ per BTC. Energy is the biggest component of that. $107 Brent is a direct hit to their cost structure. The market sees that and hedges accordingly.

S&P 500 dropped 1.74%, Nasdaq fell 2.38%, Bitcoin itself down 3.58% on the day. Yahoo Finance

You're not holding a +10% mining stock move when the broad market is selling off, BTC is down 3.5%, and oil just hit $107. The good news absorbed the macro headwind and you ended up with a net +3.6%. Honestly, in that environment, that's not nothing.

Should it have gone up more in a normal market? Yes. We're not in a normal market. War risk premium on oil is front and center right now and it directly hits the thesis.

Is This Good For Shareholders?

Short answer: yes, but it depends on what kind of shareholder you are.

If you owned MARA as a Bitcoin proxy — you now own less of that. MARA dropped to the third-largest corporate holder of Bitcoin, behind Twenty One Capital and MicroStrategy. FinancialContent The pure BTC leverage play just got diluted. For that use case, Strategy (MSTR) is now the cleaner vehicle.

If you own MARA as a digital infrastructure company pivoting to AI — this move just massively improved your risk profile. Dilution threat eliminated. Balance sheet cleaner. Capital freed up for the pivot. By retiring over $1 billion in face value debt at a 9% discount, MARA generated an immediate risk-free return on that capital. Invezz That's just good capital allocation.

Long term, everything rides on execution. The first 100 megawatts of energized AI capacity is expected by Q3 2026. FinancialContent If that hits on schedule with real revenue attached, the stock re-rates from "struggling miner" to "digital infrastructure company." The valuation difference between those two labels is enormous. If it doesn't — well, they still have $2.3B in convertible debt and the clock keeps ticking.

What I'm Watching

The Iran war and Strait of Hormuz. This is the number one macro variable for MARA right now. Any de-escalation = energy costs ease = mining margins improve = MARA re-rates. Until something concrete is signed, the war continues and so does the oil risk premium. BNN Bloomberg

BTC holding $68K. Lose that and we're looking at $62K which takes MARA down with it. Hold it and we have a base.

May 7 earnings. First chance for management to show actual AI revenue. Market will be unforgiving if there's nothing concrete to point to.

Any further debt buybacks. They've still got remaining proceeds from the sale — watch for another move on the remaining $2.3B in convertible notes.

Whether RIOT follows. If RIOT does something similar, it validates the playbook sector-wide and re-rates the whole group.

Options Angle

The debt restructuring just meaningfully reduced the left-tail risk. The catastrophic dilution scenario from convertible note conversion is largely off the table for $1B worth. That changes how you price longer-dated optionality.

IV is elevated right now — short-dated options are expensive off the back of this news. If you believe in the AI pivot story, longer-dated calls June or September expiry in the $10–$12 strike range offer interesting risk/reward if you think Q3 AI revenue materialises on schedule.

If macro stays heavy and BTC pulls toward $62K-$65K on continued war sentiment, I'd be watching for a better entry there before adding. Don't chase it here at elevated IV.

The company still carries $2.3B in convertible debt and hasn't proven the AI thesis yet. That's the risk. Size accordingly.

Bottom Line

MARA sold 15,133 BTC at a realized loss of ~$236 million and used the proceeds to buy back $1 billion in debt at a 9% discount, capturing $88 million in immediate value and eliminating a massive dilution overhang — all without issuing a single new share.

On paper that's selling low. In practice it's smart balance sheet surgery.

The stock should have held more of that +10%. The Iran war, $107 Brent crude and a broad market selloff had other plans. That's the game.

Near term it's choppy — war risk, oil prices, BTC trying to find its footing after giving back 45% from the October ATH. But the thesis hasn't changed. There's still a massive infrastructure story being built here, and the balance sheet just got a lot healthier to build it from.

If you've been following since the start of the cycle you know we've seen uglier than this. Stay disciplined. Don't oversize into the noise. And don't die on these drawdowns — live to fight on the next leg.

Let's keep going. 🤙