To The Moon

SpaceX broke below its $135 IPO price for the first time this week. On Thursday, July 23, SPCX hit an all-time low of $110.85. It closed Friday at $113.45.

That's a roughly 50% drawdown from the post-IPO high of $225.60. Just six weeks ago the stock was trading at nearly double where it is today. Below the IPO price is a specific psychological level for a new listing — it's the first point at which every single public shareholder is underwater on their entry.

I want to talk about what I did this week. I doubled the short side of the ECHO/SPCX pair I wrote about last time. Went from 225 shares short to 450. Added 300 more shares on the long ECHO side too. Position sizing on this specific trade is now materially bigger than it was seven days ago.

I want to walk through the reasoning, because there's a distinction here that most retail traders miss.

Adding To A Working Trade Is Not The Same As Adding To A Winner

If I were long a stock in a normal directional trade and it dropped 20%, adding to that position is a controversial decision. You're compounding your bet that you were right about direction, and the market is telling you (via price) that you were wrong so far. Sometimes you're actually right and the market comes around. Often you're just averaging into a mistake.

That's the psychology every retail trader has been warned about with the phrase "don't add to losers, don't add to winners." It's decent advice for directional trading.

But this trade isn't directional.

The ECHO/SPCX pair isolates a spread — the discount between EchoStar's implied SpaceX exposure and SpaceX's actual trading price. That spread is what I'm exposed to. The individual price of SpaceX doesn't matter to my P&L in isolation, because the two legs are structured to hedge each other's directional moves.

When SpaceX drops 20% and EchoStar drops 5%, one of two things happened:

- The spread compressed — good, my trade is working

- The spread widened — I got a better entry into the same catalyst

Either way, the direction of SpaceX isn't the thing that pays me.

This week SpaceX dropped further and the spread widened. The EchoStar-implied embedded value per share (the 261.8 million SpaceX shares landing on their balance sheet in November 2027, priced at the current market mark) is still enormous compared to EchoStar's market cap. And now the short leg of the pair is available at meaningfully lower prices than my original entry — meaning if I add more short exposure at $159 average and buy the offsetting ECHO long, my new tranche is a better spread trade than my original tranche.

The catalyst hasn't changed. The FCC approval from May stands. The deal closing conditions and target date haven't moved. What changed is the market's mark on SPCX, and that changed in a direction that made the pair a better entry, not a worse one.

That's the distinction. Adding to a merger arb when the spread widens is entering a better version of the same trade. Adding to a directional loser is compounding a bet against the market. Different frameworks, different rules.

Where The Pair Stands Now

- Long 750 ECHO at an average of about $102.50, currently around $87. Long leg unrealized: roughly negative $11,600.

- Short 450 SPCX at an average of about $159.50, currently around $113. Short leg unrealized: roughly positive $20,700.

- Net pair: approximately positive $9,100.

Last time I wrote about this, the pair was up about $5,300. In a week where SpaceX dropped through its IPO price and EchoStar dropped alongside it, the pair produced another $3,800 of net gain. That's exactly what a well-structured hedge should do — the short leg outperformed the long leg's drawdown.

I want to be clear about scale though. The trade is now bigger. Which means if the deal breaks between now and November 2027 for any reason — regulatory, contractual, corporate action, anything — the downside is proportionally bigger too. Sizing discipline matters more when you add, not less.

The risks I named in the last writeup all still apply. Deal risk. Lockup risk on the SpaceX shares EchoStar receives. Borrow cost over 16 months. Divergence risk. Management risk on how EchoStar deploys the proceeds. All of those cut against a larger position more than a smaller one.

I'm going into this add with those risks priced in mentally. Doesn't make me right. Just means I'm not surprised if the pair widens against me next week or next month, and I don't have to make a fresh decision when it happens.

Oracle Got Delivered To Me This Week

Different topic, real news for the book.

Three short puts I'd sold on ORCL at the $135 strike came due on July 24 expiry. Oracle closed the week at $114.85 — well below strike. So the puts got exercised and I now hold 300 shares of Oracle at a $135 cost basis, currently marked at about negative $6,000.

Oracle dropped hard this month on the AI capex guidance. They laid out plans to spend up to $95 billion in cloud capex by FY27 and raise about $40 billion in debt and equity to fund it. Free cash flow turned negative. The market is now working through whether that spend is a strategic build-out or a margin-compression accident. Fair debate.

I'm not going to pretend the assignment happened at a great mark. It didn't. If I could pick the day Oracle got delivered to me, it wouldn't be this week.

What the wheel structure gives me from here is optionality. I've already sold an August 21 $135 covered call at $2.91 against the position. If Oracle recovers to $135, I get called out at breakeven on the shares plus keep the premium. If it stays here or drops further, I keep the shares, keep the premium, and can sell the next round of calls at whatever strike makes sense at the time.

The wheel doesn't magic away the entry mark. It just gives me a working income stream on a position that would otherwise just sit red on the P&L blotter. That's what it's for.

Housekeeping

The MARA put wheel decayed cleanly through Friday's expiry. Puts I'd sold at $11.5 and $12 for July 24 came off at close to zero, so the full premium was retained. Nothing dramatic — just the design working on a week when the underlying happened to cooperate.

MSFT LEAPS partial recovery this week. Still all underwater, still all held. New short calls refreshed at the August 14 $450 strike. Earnings coming July 29 — I'll write about whichever direction that goes.

IBIT wheel scaled up modestly. COIN wheel had a roll cost this week on a short call that ran, offset by fresh premium at higher strikes further out. Nothing catastrophic anywhere.

The Honest Read

The pair trade is now a bigger position with the same catalyst pending. That's a real risk-taking decision, not a passive hold. If SpaceX rips higher on a Starlink revenue beat or a big defense contract announcement between now and closing, the pair widens the wrong way and I'll write that post too with the same level of specificity as this one.

The Oracle assignment is a real cost. It's now part of the book. The wheel handles it from here, but the mark is what it is.

The rest of the book is doing what it's designed to do — grinding premium, absorbing weekly noise, maintaining structure without needing my emotional input.

That's the week.

If you want the follow-up detail on how I sized the pair add, or how borrow cost affects the arithmetic over a 16-month hold, or how the ORCL wheel plan changes if Oracle stays below $130 through August — drop a comment below. Faster channels are TikTok and YouTube DMs (Mathematical Money on both) or through trueknot.sg.

Stay disciplined. Size your positions properly. See you next week. 🤙