Strategy CEO Explains Selling Bitcoin Low Then Buying Higher

Have you ever watched a company sell an asset, then turn around and pay a higher price for more of the same thing, and felt that little jolt of “wait, that cannot be right”? That is the reaction a lot of people had when Strategy’s chief executive defended sales of Bitcoin near $60,000 and later purchases closer to $80,000. On paper it looks like buying high after selling low. In the room where those decisions get made, the story is messier, and frankly more interesting, than a simple price chart.

Why Strategy Says Price Was Never The Real Trigger

Phong Le’s argument is blunt. Strategy does not treat Bitcoin like a day-trading toy. The firm looks at cost of capital first, then decides whether buying, selling, issuing stock, or sitting still improves the balance sheet. If that sounds cold, it is supposed to. A treasury that size cannot afford romantic timing.

Last week’s disclosed purchase was not small in dollar terms, even if it was modest next to the pile Strategy already holds. The company bought 4,603 Bitcoin for about $369.7 million between August 24 and August 30, at an average of $80,318. That lifted total holdings to 845,050 Bitcoin, acquired for roughly $63.73 billion, or about $75,412 per coin. Those numbers matter because they show the firm is still a giant holder, not a trader flipping a few blocks for sport.

We don’t buy or sell Bitcoin based on the price of Bitcoin. We buy or sell based on our cost of capital.

– Phong Le

I’ve found that line lands differently once you stop treating it as a slogan. It is an operating rule. Cheap equity can fund more coins. Tight liquidity can force sales. Preferred dividends do not wait for a friendly candle. In my experience covering corporate treasuries, that mix of flexibility and obligation is where most public commentary goes sideways.

The Purchase That Restarted Accumulation

After a two-month pause, Strategy came back as a buyer. Management sold about $602.8 million of common shares in the week ended August 30 and used part of that cash for Bitcoin. It also added roughly $29 million to its general dollar cash pool and spent about $152 million buying back STRC preferred shares below the $100 stated amount. That is not a single trade. That is a week of capital allocation.

The point Le kept circling is simple. Issuing common stock looked attractive again because the shares were trading at a premium to the value of the assets behind them. If you can raise dollars above the implied value of what you already own, buying more Bitcoin can raise Bitcoin per share. If you issue stock when the premium is gone, you can do the opposite. Price of Bitcoin is an input. It is not the whole equation.

Perhaps the most interesting aspect is how ordinary this logic would sound in any other asset-heavy company. A miner selling production to cover costs does not get accused of “giving up on gold.” A REIT recycling capital does not get treated as a traitor to buildings. Bitcoin still carries a purity test that other treasuries do not. Strategy is trying to step off that test without dropping the long-term bid.

What The Earlier Sales Actually Paid For

Le said Strategy sold roughly 7,000 Bitcoin during a balance-sheet cleanup and called that amount “minuscule” next to total holdings. Public ledger figures are a bit tighter. Disclosed sales include 2,225 Bitcoin in early July, 1,638 in early August, and 1,690 the following week. Those three prints add up to 5,553 coins. Earlier in 2026 the firm also reported selling about $218.4 million of Bitcoin to help cover preferred dividend needs. The interview number looks like a rounded view of a longer stretch, not a new secret dump.

Why sell at all? Because preferred dividends are a cash promise. You can issue new paper, dip into reserves, or sell a slice of the reserve asset. Le said selling Bitcoin to meet those payments was “the right trade at the time.” That sentence will annoy people who wanted a forever-hold slogan. It also matches a company that already told investors it would keep paying those coupons.

In June the board authorized a monetization program. Management can sell Bitcoin to fund a dollar reserve, pay dividends and interest, repurchase securities, or meet other approved needs. The policy allows up to $1.25 billion of sales to build a designated USD reserve. It does not order the firm to hit that ceiling. That distinction gets lost in social-media shorthand, where “can sell” becomes “is dumping.”

• Sales were framed as funding tools, not a thesis change.
• Disclosed reductions stayed small versus more than 840,000 coins held.
• Dividend cash needs, not a price target, sat at the center of the July and August activity.
• The later buyback of preferred stock used dollars raised when equity looked expensive in a good way.

A Two-Way Treasury Is Still A Net Buyer Story

Le now calls the approach a two-way strategy. Strategy may sell when a sale improves the capital stack, and it still intends to be a net accumulator over time. Bitcoin becomes one lever among several: common equity, preferred stock, cash, and buybacks. That is less poetic than “never sell.” It is also closer to how a public company actually lives.

The June framework is pretty explicit. Common stock issuance can be accretive when shares trade above adjusted net asset value. Issuing below that line can dilute Bitcoin per share. So the same $80,000 coin can be a smart buy one month and a poor use of cash the next, depending on where the stock sits and what the coupon calendar looks like. I know that feels slippery. Markets are slippery.

Le even floated examples, not forecasts. The company could keep buying at $90,000, $100,000, or $130,000 if capital is cheap enough. It could sell again if proceeds do more work somewhere else. Price alone would not settle it. That is the part critics skip when they paste two prints on a chart and declare incompetence.

The “Fortress” Claim And What Net Leverage Of Zero Really Means

During the purchase pause, Strategy bulked up dollar assets and reduced net exposure to convertible debt, at least by the company’s own yardstick. As of August 30 it reported $6.71 billion in USD assets, cash and other dollar holdings split across the designated reserve and general liquidity. Convertible debt sat near $6.75 billion. Subtract one from the other and management reports net leverage of 0.0%.

That metric is easy to overread. Zero net leverage does not mean the legal debt vanished. Convertible notes are still there. Preferred dividends still come due. The calculation subtracts dollar assets from convertible principal, then compares the remainder with the Bitcoin reserve. It does not subtract every preferred claim. So the “fortress” language is a balance-sheet story, not a magic shield.

Le’s stronger claim is that there is no meaningful forced-sale price baked into the current structure. The debt is not secured against Bitcoin in a way that automatically dumps coins at a stated level. Fair enough. A long slump would still hurt. Asset value would fall. The stock premium could shrink. Raising capital could get uglier. Preferred payments would feel heavier. A fortress with a leaky roof is still a building. It is not invincible.

How Cost Of Capital Turns A “Bad Trade” Into A Funding Trade

Walk through the sequence without the outrage filter. In one window, Bitcoin sat near $60,000 to $65,000 and the firm needed dollars for coupons and restructuring. Selling a thin slice created liquidity without waiting on a perfect equity print. Later, the stock traded at a premium. Issuing shares brought in cash on terms management liked. Buying coins at $80,000 could still raise Bitcoin per share if the equity raise was rich enough. Same asset. Different financing weather.

People hate this because it refuses the folk rule of “never sell your winners.” Corporate finance does not run on folk rules. It runs on coupons, dilution math, and whether the next dollar is cheap. I’ve sat with enough treasury slides to know the ugly truth: the best-looking chart trade can be the worst capital trade, and the reverse is also true.

Is there room to argue Strategy over-complicated a simple stack of coins? Sure. A smaller holder with no preferred stack would have stayed idle and slept better. Strategy is not that holder. It built a machine that uses Bitcoin as both reserve and financing collateral in spirit, if not always in the legal sense. Once you build that machine, you have to service it.

The Reputation Risk Nobody On The Earnings Call Can Shrug Off

For years the public story was simpler. Accumulate. Talk about scarcity. Treat sales as heresy. The first disclosed sale under the newer policy changed the costume. Strategy is still a net buyer. It now admits sales are allowed. That is a branding problem as much as a finance problem. Fans wanted a monastery. They got a treasury desk.

That does not make the sales a scandal. It does mean every future print will be read as a signal. Buy 4,603 coins and the tape cheers. Sell 1,600 and the same tape asks if the dream is over. Communication has to stay boring on purpose. Cost of capital. Premium to asset value. Dollar reserve targets. Repeat until the temperature drops.

The apparent contradiction between selling near $60,000 and buying near $80,000 fades when both prints are read as financing decisions, not price predictions.

I keep coming back to that. If you insist on judging every lot against the last print, Strategy will look sloppy forever. If you judge lots against the cost of the cash that funded them, the tape starts to look like a company managing a very large, very public experiment in corporate Bitcoin.

What Investors Should Watch Next, Without Pretending They Can Time It

The next chapter is not a speech. It is the weekly capital-markets activity. More common stock sold at a premium can fund more coins. Weak demand for equity, or a jump in financing costs, can slow buys or make another sale the cheaper option. That is the dashboard. Not a guru target. Not a slogan.

1. Track whether common shares still clear at a premium to adjusted net asset value.
2. Watch the dollar reserve and general cash, not just the headline coin count.
3. Note preferred dividend coverage and any further STRC repurchase prints.
4. Separate net-leverage marketing from legal debt that still must be serviced.
5. Treat each sale or buy as a capital decision first and a market call second.

Strategy now sits on a bit more than 4% of the maximum 21 million supply. That scale is the whole plot. A firm that large cannot hide. Every lot hits headlines. Every pause becomes a theory. The honest read is less dramatic. Management used sales to get dollars when it needed dollars. It used equity when equity was willing. It bought again when the math looked accretive. You can dislike the optics. You can still follow the cash.

A Plain-Language Walk Through Dilution, Premiums, And Per-Share Bitcoin

Let’s slow this down, because this is where smart people talk past each other. Suppose the stock trades above the company’s adjusted net asset value. Selling new shares brings in more dollars than the slice of assets those shares represent. Deploy those dollars into Bitcoin and Bitcoin per share can rise even if you paid a higher spot price than last quarter’s sale. The coin is “more expensive.” The share of the pile attached to each share can still improve.

Flip it. If the stock trades cheap versus assets, issuing equity can shrink Bitcoin per share. In that window, selling a little Bitcoin to cover a coupon might damage the story less than printing stock into a discount. Neither path is morally pure. Both can be rational. The market then argues about which pain it prefers: a visible coin sale or a quieter dilution.

That is why Le keeps saying they do not buy or sell “based on the price of Bitcoin.” He is not claiming price is irrelevant. He is saying price is incomplete. Incomplete models make loud commentators. They also make bad treasury work.

Simple treasury filter Strategy is describing:
  1. What does the next dollar of capital cost?
  2. Does deploying that dollar raise Bitcoin per share?
  3. Do cash obligations force a sale regardless of mood?
  4. Does a sale strengthen the dollar reserve enough to buy optionality later?

Why The $60K Versus $80K Headline Will Keep Coming Back

Headlines love a contradiction. Sell low, buy high, gotcha. It fits in a single line. The fuller version needs room for preferred coupons, a June policy memo, a two-month pause, a $6.71 billion dollar sleeve, and a stock that sometimes trades like a leveraged wrapper around the coin. That version does not travel as fast. It is still the version that matches the filings.

Will Strategy sell again? Le left the door open. Will it buy at much higher prints if capital is cheap? He left that door open too. The only closed door is the old idea that sales are forbidden. For a company that spent years marketing itself as a permanent accumulator, that is the real shift. Not the $20,000 gap between two average prices.

I’ll say this as plainly as I can. If you own the stock because you wanted a closed-end Bitcoin fund that never touches the pile, the two-way policy should make you sit up. If you own it because you wanted a public vehicle that can raise capital and keep adding coins when the wrapper trades rich, the latest week looks like the model working as designed. Same facts. Different contract with yourself.

The Human Bit: Optics, Trust, And A Very Public Pile Of Coins

There is a human layer here that finance tables miss. People attached a moral story to this treasury. HODL became identity. Then the identity met a dividend calendar. That collision was always coming. Once you wrap Bitcoin in preferred stock and convertibles, you imported the old world into the new one. The old world sends invoices.

Le’s job now is not to win a purity contest. It is to keep the machine funded without looking like the firm lost the plot. Calling the balance sheet a fortress is part of that job. So is reminding listeners that 7,000 coins, even if the rounded figure is a little high versus the three detailed prints, is still a sliver of 845,050. Scale is the defense. Communication is the risk.

In my view, the useful question is not “how could they buy higher after selling lower?” The useful question is “what did each lot of cash cost, and what did it buy in flexibility?” That question is harder. It also keeps you from turning a capital-markets week into a morality play.

A Longer Look At Preferred Stock, Coupons, And Why Cash Still Rules

Preferred securities are the unglamorous character in this drama. They helped fund the pile. They also created a clock. Cash has to show up on schedule. You can romance the long-term thesis all afternoon. The paying agent still wants dollars. That is why a designated USD reserve showed up in the policy. That is why a sale can be “right at the time” even if the spot price later prints higher.

Buying back STRC below the $100 stated amount is the other side of the same brain. If the preferred trades cheap, retiring it can be a better use of cash than stretching for one more lot of coins. If common equity is rich, issuing it and buying coins can be the better use. The asset is constant. The relative prices of the wrappers change. Strategy is trading the wrappers as much as it is trading Bitcoin.

Some readers will call that financial engineering. Fine. Most large public balance sheets are financial engineering with better stationery. The only special thing here is the reserve asset. Everything else — premiums, coupons, buybacks, convertible overhang — is standard corporate life wearing an orange-tinted jacket.

Risks That Survive The Polished Interview

A two-way policy cuts both ways. Flexibility can protect the stack. It can also tempt management into too many small sales that slowly change the story. A 0.0% net leverage print can soothe. It can also hide the fact that preferred claims still sit outside that neat subtraction. A premium to net asset value can vanish in a risk-off week. Then the cheap-equity button turns off and the firm is back staring at coins or credit markets.

• Equity premium compression would make accretive buys harder.
• A long Bitcoin drawdown would pressure asset value and sentiment together.
• Preferred coupons remain a recurring cash need, policy or no policy.
• Market narrative risk is now permanent because sales are on the menu.
• Scale helps, but it does not erase execution risk in weekly issuance.

None of that makes the latest purchase irrational. It does mean the “we might buy at $130,000” line is a conditional statement, not a dare. Conditions change. That is the whole thesis Le is selling, whether or not you like the packaging.

Putting The Week In Perspective Without The Fan Fiction

Zoom out. A company with hundreds of thousands of Bitcoin sold a few thousand to tidy dollars and service paper. It later issued common stock into a premium and added 4,603 coins at a higher average. It reported dollar assets almost equal to convertible principal. It bought back some preferred stock below par. It said it remains a long-term net buyer. That is the week. The fan fiction writes itself in both directions — collapse, triumph, hypocrisy, genius. The filings are quieter than the fiction.

I do not need Strategy to be a saint. I need it to be consistent with the rules it published. On that narrower test, selling when cash obligations pinched and buying when equity was willing is at least coherent. Coherent is not the same as destined to work. Coherent is still a higher bar than most viral takes clear.

So yes, the firm sold nearer $60,000 and bought nearer $80,000. Yes, that looks backwards if the only variable you allow is the coin’s last price. No, that is not the variable management says it uses. Cost of capital is a drier headline. It is also the one that explains the prints without inventing a secret change of heart.

The Takeaway If You Actually Have To Live With This Stock

If you hold the equity, stop scoring each lot like a weekend trader. Score the spread between financing cost and the per-share Bitcoin outcome. Watch the dollar sleeve. Read preferred obligations as real, not decorative. Accept that two-way means two-way. The company said the quiet part out loud. Sales are allowed. Accumulation is still the intent. Both can be true in the same year.

If you do not hold the equity and you just wanted a clean parable about never selling, this episode will irritate you. That is okay. Parables are for campfires. Balance sheets are for Tuesday mornings. Strategy chose Tuesday.

And if you came for a prediction, Le did not give you one. He gave you a filter. Cheap capital, add coins. Costly capital or cash bills, consider the other tools. The $60,000 sales and the $80,000 buys are the same filter in two seasons. Whether that filter keeps working is the live question. The contradiction, at least the cartoon version of it, was never the point.