Gooaye EP712: The Day the Market Broke, He Went to Disneyland

Contents

1. He woke up and assumed the app was broken
2. Along the moat, everyone was passing him
3. When the parade came around
4. He compressed the politics into one switch
5. Which layer gets cut first
6. Everything fell, so why did mine fall harder
7. With news this frightening, which item should change my mind
8. Sources worth your time
9. One thing to take with you

The wind rose and the waves surged. Sun, Wang and the others turned pale and called for the boat to put back. The Grand Tutor’s spirits were at their height; he hummed and whistled and said nothing. … But when the wind turned fierce and the waves grew violent, and no one could sit still for the clamour, he said slowly: “At this rate, shall we never get home?”

—— Liu Yiqing, A New Account of Tales of the World, “Cultivated Tolerance” (Southern Song dynasty, c. 440 CE; my own translation)

In Gooaye EP712, released 7 November 2026, host Hsieh Meng-kung was in Japan while Taiwan’s market opened sharply lower the day before. He says he had no time to feel bad about it: he ran seven kilometres around the Imperial Palace that morning and took his family to Disneyland afterwards. He had bought the Wednesday dip and bought again at Friday’s open, then day-traded part of that dip position out mid-session because the fall was deeper than he expected — and after the close, when the National Stabilization Fund stepped in, he thought he had cut too early. The most portable judgement in the episode is how he compresses post-midterm AI policy risk into one observable switch: whether new data center permits get blocked. Orders through 2027 are unaffected, he says; a permit freeze would hit 2028–2029 revenue, and the market prices that early. Here is what I took from it.

He woke up and assumed the app was broken

He woke up in a Ginza hotel. He had gone to bed early and knew nothing about the open, so when he unlocked his phone his first thought was that the app was displaying an error.

That reaction had a history. On Wednesday he had bought the dip, the market bounced right after, US stocks bounced that night, and he felt he had timed it well — his only regret was not putting more money in. He joked that someone upstairs heard him and offered a second, cheaper chance, which is how Friday opened.

That same evening in Ginza he found a bottle he had been hunting for since a trip to Hokkaido, and paid the premium because he wanted to buy it in Japan rather than from a reseller at home. He picked out several, and told the clerk one of them didn’t need duty-free processing — he was going to drink it that night and put the whole day’s pressure into that bottle. I laughed, and then recognised the pattern: when the number on the screen is beyond doing anything about, you go handle one small thing you can.

Along the moat, everyone was passing him

Before heading out he settled what he was going to do. Wednesday’s dip-buy he still considered correct, and pushing in at Friday’s open was correct too, but fear was starting to show up — the market shouldn’t have fallen that far. His term for it is muscle memory: a name that slides into support and reopens higher can keep its uptrend; one that drops too far at once and doesn’t lift in the short run has broken the habit of grinding upward. So the plan to keep adding got cut back to one more tranche next week, and if a bounce comes he wants to raise cash into it instead.

Then he went running. The plan a friend wrote for him doesn’t ask for twenty-kilometre days: seven to ten kilometres, three or four sessions a week, rest in between, and a hundred kilometres a month is enough to finish a half marathon. While running, he says, he wasn’t thinking about the market — he was thinking about whether he’d finish that race at the end of the month.

The moat path was busy and at roughly six minutes per kilometre he kept getting passed. Somewhere in there he noticed something: only the faster runners can overtake him, and everyone slower never enters his field of view at all. He carried it over to ordinary life — you watch the neighbour buy a new car, the other neighbour buy a bigger house, the pressure pushes you into things past your capacity, and one liquidation later you pay for it. Meanwhile your own life also moved forward, and you never credited yourself for it.

When the parade came around

He placed a few orders on the way back, and then the family went into Disneyland. He had no time to watch the market all day, and no opening for either relief or grief. His friends’ group chat kept wailing; someone said he was stuck and had to sell a position to free up cash, and the others told him to go ahead and dump it.

He has been through several of these massacres, and says this was the one he handled with the most room to spare: positions were still being shot on the field, but his mood held, he could ride rides with his kid, watch the parade, walk around, and he didn’t lose his temper at anyone. He believes Peter Lynch’s line that the most important organ in investing is the stomach — whether you can absorb what the market hands you. He gave his own example: a holding that slid from the two hundreds through a hundred while everything he could see said otherwise, to the point where he wondered whether he or the market had lost its mind. It later traded above four hundred.

Mid-session that Friday he day-traded out part of the dip position, on the grounds that he had bought too early and shouldn’t keep pouring in. After the close, with the stabilization fund buying, he wished he had kept all of it for next week’s bounce. His own diagnosis: the brain overwrites itself easily — you write a script, it feels wrong, you revise, an event lands, you revise again. When noise and volatility get this high, slow down. He added a line I wrote down: a correct judgement doesn’t have to show up in the result.

He compressed the politics into one switch

The other half of the episode is the US midterms. After the Democratic sweep, one Democratic legislator posted that he wanted AI stopped altogether, and the replies were savaged by people from both parties: halting the technology hands the lead to China, and AI is already accelerating drug discovery. His read is that the progressive wing isn’t the dominant force inside the party, so the worst case is unlikely — but the market gets scared by that kind of post anyway.

So he compresses the whole mess into one observable signal: whether new permits are getting blocked. Politicians can say whatever they need to say, because going after corporations and the rich is the cheapest way to hold a voter base. Gavin Baker had shared a view that the Texas governor will impose some AI regulation but is unlikely to touch projects already permitted or under construction — which he finds reasonable, since halting a build invites a pile of litigation.

The chain is the part worth keeping. Projects under construction keep pulling material, so orders from now through 2027 are fine. If new permits start getting blocked and new data centers can’t break ground, what suffers is 2028–2029 revenue. That sounds distant, except the market prices it early, which is how the script changes shape.

He also spent time on why Americans voted this way. Campaign language lands on daily costs — fuel is expensive, groceries are expensive, and if you feel the inflation you vote against whoever is in charge. Taiwan’s index looks strong because so many of its components sit on the AI line, and his old classmates and colleagues who aren’t on that line never felt the good economy. His warning to the winners: when people who made money get loud, a backlash follows. And a Democratic administration isn’t redistribution either — everyone has a 401(k), so a drawdown makes everyone poorer together.

Which layer gets cut first

He brought up Musk’s post alleging a SoftBank default on power-generation turbines, meaning some orders may be cancelled. He says he doesn’t know what Musk is basing it on, but the man’s sources are far better than anyone else’s, so it’s worth watching.

Cancellations travel in a sequence. The cheap layer goes first: brackets, stampings, the commodity parts anyone can make, where the contract carries no real penalty and the supplier can’t do much about it. Cancelling TSMC-class orders is expensive — prepayments may not come back, and next time you need capacity they won’t give you an allocation, so you lose the right to order at all. Hence his test: a single customer with a problem whose volume gets picked up by someone else isn’t a big deal; a run of customers cutting the parts that “look unimportant and anyone can make” is an early warning. The 2018 and 2022 supply chain corrections had this shape — cut the edges first, then transmit up the chain.

Everything fell, so why did mine fall harder

I ask this in every drawdown. His answer that day was margin debt.

The selloff wasn’t aimed at margin accounts, he says, but names with heavy margin do swing harder, because a group of holders inside them has no choice about selling. He told listeners to go back and look at which names saw margin balances jump on Wednesday, and whether those were the worst performers over the following days. The reverse case: some names barely move early, because the register is settled — mostly insiders, or holders who agree on what the story is and aren’t trading a headline, so nobody is forced to adjust. In a full crash everyone goes down eventually.

The usable version for me: when a holding falls harder than the index, look first at how its margin balance moved this week and who actually owns it, then go find the news. My own book is concentrated in semiconductor equipment and software, with several names more than 50% off their highs, and my reflex has always been to hunt for deteriorating fundamentals first. After this episode I added a step before that one — separate “the fundamentals moved” from “someone is being forced to sell.” He quotes a friend’s observation in the same direction: once a stock goes vertical or breaks hard, fundamentals matter less inside the move, and what’s left is positioning and sentiment.

With news this frightening, which item should change my mind

This is where I get lost in the noise. The episode’s template is that permit switch: posts and speeches from politicians go in one bucket, permits actually issued or withheld go in another, and only the second one moves my view of a holding. The difference is whether it can be written as “if X happens, I change judgement Y.”

The cancellation signal has the same shape. One customer cancelling and the volume getting absorbed changes nothing; a run of customers cutting edge components means I reopen every company on that chain. The thing I’ve tried is writing one invalidation condition per holding — and the ones where I can’t write it are the ones where I never worked out why I’m holding them.

On timing, his reminder is more useful to me than any level: a VIX spike is usually a good short-term entry, and the real question is how far out you’re looking once you’ve bought — whether you’re trading this bounce or betting it opens higher ground. He doesn’t go to cash; he raises the cash weight above where it was, and waits to see whether the bounce can reclaim the quarterly moving average and hold there before going on. Reclaim it and fall straight back, and you’ve built a large top.

Sources worth your time

• Gooaye EP712 (7 November 2026)
• Peter Lynch, One Up on Wall Street — context for the line about the stomach
• Liu Yiqing, A New Account of Tales of the World, “Cultivated Tolerance”

One thing to take with you

Only the people passing you are visible. Everyone slower than you is structurally absent from your field of view, so your sense of where you stand is computed from a sample that tilts by construction — and the tilt doesn’t shrink as you try harder. That’s what pushes people past their own capacity, and what gets paid for in the next liquidation.

Here’s something I’ve tried that fits in today: pick one thing you feel behind on, dig up the record from the same week last year, and write last year’s number and today’s number on the same line. Savings, weight, running pace, job title, the words your kid can say — any of them count. You’re not after reassurance; you’re after the slope of your own line, which is the only information that still belongs to you while someone is passing you.

Raw Markdown

This article is an educational discussion of investment method. It is not advice to buy or sell any individual security, offers no target prices, and does not analyze any current holding. Investing carries risk; make your own decisions or consult a qualified professional.