New Ask Claude or ChatGPT about any stock, using Bullpen. Bullpen in Claude or ChatGPT. 7 days free
Bullpen.News From Edwin Dorsey
Get Daily Recap Feedback

Friday, November 13, 2026

Back to the front page

From X

@StableBread

Fajasy on X

Boost Run $BRUN shows $218.6M of new GPU leases in Q3, and its September 30 lease schedule still implies a ~7% interest rate, the same as the June 30 schedule implied. $BRUN had to put $96.0M of its own cash up front on two of the three GPU leases it signed in Q3: → 36-month lease: $59.2M advance, then $2.7M/month. → 24-month lease: $36.8M first payment, then $4.3M/month. For reference, the $59.2M advance alone is more than the $39.4M of advances across all 34 GPU leases $BRUN signed in H1 2026. In Q3, $88.2M of warrant exercises covered most of the $109.8M of lease principal $BRUN paid, and cash still rose $36.9M to $170.0M ($157.0M unrestricted). All 11.4M warrants are now exercised, and GPU lease payments due from January to December 2027 rose to $186.2M as of September 30, from $103.5M as of June 30. The two big Q3 leases alone add $6.9M/month of payments in 2027. The ~$221M of 2027 payments I counted across $BRUN's signed commitments in my October 19 update is now ~$307M, before the five equipment leases $BRUN signed in October, which require $83.1M of payments over the next two years. Four of those five are six-month leases that end with $39.6M of final payments, 59.8% of their $66.2M of total payments. Against those payments, $BRUN says $482M of customer prepayments are due as upcoming projects deploy. So 2027 now runs on (1) those prepayments arriving on schedule and (2) monthly billings, since the warrant cash won't repeat.
@calvinblissett 1 click

Calvin Blissett on X

The Korean preferred (the one with the SpaceX stake $SPCX ) reported this week, and the headline was a loss. I read the filing, the call coverage and the first round of broker notes. Here's my takeaways: The loss came from marking investments to market. SpaceX dropped in the quarter and a few other positions went against them. Take that out and the brokerage still made good money in a quarter when Korean trading volumes fell hard. That was the test I cared about. The boom ended and the core business still earned roughly what I'd assumed for a normal year. Growth: On the call, management said they’re preparing to enter Singapore, the US and Australia as well as Japan. They gave no names, prices or timing for those three. Japan, through the Sanko Securities deal, comes first. The plan, in their words, is to buy local brokers in developed markets and use them to offer online wealth management to local retail investors. That’s the same model they run in Korea, sold through an app rather than branches. They said online is much cheaper to run than offline. The long-term goal is one platform where clients hold both regular investments and digital assets, which is why the Hong Kong unit already offers crypto trading. Separately, they’re folding their Shanghai and Beijing offices into Shenzhen and Hong Kong. They also just agreed to buy a small licensed Japanese broker for a modest sum, aiming to launch an online wealth app there within about two years. Japanese households keep an enormous pile of savings in cash and are only starting to move it into markets. If the playbook that works in Korean pensions carries over, it’s a cheap option. The parts that contribute to baseline earnings held up: Pensions and wealth management barely moved while trading fees fell sharply. Retirement money doesn't leave because the market had a bad month, and the company is still #1 in the part of the pension market it competes in. Pricing held on overseas trades while competitors keep giving trades away for free. The overseas businesses are having their best year ever. The SpaceX stake is still worth several times what they paid. Management said on the call that gains get returned to shareholders once they're actually realized. With the lockup ending next month, that money is close to being real. What management has done for shareholders this year: Its largest buyback ever, a big chunk of it spent on the preferreds, including the first preferred for the first time. The largest share cancellation in its history in July. Over the last two years it handed back more than the policy required. This week it said it will cancel about 60% of the treasury shares left over from an old merger right after the March shareholder meeting. The new three-year policy keeps the same minimum payout, with extra returns once SpaceX gains are realized. They've also said in writing that part of the reason for buying preferred was to narrow the gap between the common and preferred prices. Companies are rarely that explicit although I have seen it before Why the prefs could be the next thing they buy: the new policy dropped the fixed yearly commitment to cancel preferred shares, which I didn't love at all, and nobody on the sell side seems to have noticed. But the wording changed in a way that helps. The cancellation of the old merger shares sits outside the payout policy. So every year, the minimum payout has to be met with dividends and fresh buybacks. That money has to go somewhere. With the prefs at about a 63% discount to the common, each won spent on them retires roughly 2.7x as much of the company as the same won spent on the common, and the prefs get exactly the same dividend. If the goal is shrinking the share count, the prefs are the cheapest place to do it. Management already acted on that in June, at almost twice today's price. The negatives: management plans to sell some treasury stock over time instead of cancelling all of it. One broker puts the dilution at under 2% a year. Next quarter will carry a regulatory fine plus another SpaceX markdown. REGULATORY RISK IS SOMETHING TO WATCH BECAUSE OF AN INVESTIGATION! Most of the brokers kept their buy ratings. One estimated returns on equity in the low teens without giving SpaceX any credit.

Investor Letters

All letters

QXO Inc Earnings Update: Q3 2026

QXO's Q3 sales rose 79% to $4.9 billion after TopBuild and Kodiak acquisitions, while acquisition debt lifted interest expense to $163…

QXO $13B
SixSigmaCapital: Navigating the Financial Markets 2 clicks

Qure Anniversary Update

UniQure’s Huntington’s disease BLA has a May 1, 2027 priority-review date, with an FDA advisory committee expected in late first quarter.

QURE $1.7B
Triple S Special Situations Investing 2 clicks
WULF · long

Terawulf INC

TeraWulf is shifting from bitcoin mining toward data centers, seeking a higher-quality and more defensive business model.

WULF $6.7B
Value Investors Club

The Data Center Bridge

Small- and mid-cap data center stocks trade as one despite sharply different funding bridges: 12 of 15 cover fixed costs on signed…

IREN $8.8BFRMI $1.3B
CV's Substack 16 clicks

Amkor Q3 FY12/26 Earnings Review

Amkor's Q3 sales were flat at $1.991 billion as computing and automotive/industrial growth offset a 27% communications decline; Q4 sales…

AMKR $9.2B
Cestrian Capital Research 4 clicks

Molina & Kinsale

Molina Healthcare and Kinsale Capital’s earnings are reassessed through a long-term lens, emphasizing conviction and detachment during…

MOH $11BKNSL $8.8B
Bearstone 8 clicks

Publications we follow

Every publication on this site, hand-picked.

130 publications

Newsletters 110

YouTube 4

Blogs 15

Seeking Alpha 85 of 289

Nominate a publication

X accounts we follow

Every account on this site, hand-picked.

91 accounts
Nominate an account

Investor letters we follow

376 firms

Sectors

Ordered by how often our editor reaches for them.