Magic Eye
In the late 1900s, a book series called "Magic Eye" materialized on the bookshelves of Millennial kids. The gimmick was to hide "3D" pictures inside an image that looked just like noise. If you look at the page, you see the noise. If you focus beyond the image, you see the hidden 3D picture. The whole thing is called an autostereogram.
How you look at it reveals the object. The process is the only way to see what's there. If you fixate on the page, the image turns back into noise. All you can do is focus on the technique.
Not everyone can see autostereograms. People without depth perception can't. People with strabismus can’t. Plenty of others. This isn't to pick on people who struggle with autostereograms — I struggle with them — just to point out that you can't separate the process from people.
Looking past the noise
Investors like to figure out what their "edge" is — what is it that gives them the ability to generate persistent favorable risk-adjusted returns? They also like to put that in a bottle separate from the people doing the investing. "Our edge is in our process. Our process is purely institutional," they'll insist.
Investor Gavin Baker talked about investing "process" in a recent interview. He recalled how Chicago Bulls owner Jerry Krause preached that it wasn't just Michael Jordan winning all those NBA championships back in the 1990s. It was the organization — the Chicago Bulls — it was Pippen and Kerr — it was Phil Jackson, the training staff — it was the facilities, Krause insisted.
Since Jordan left, the Bulls haven't won a championship. Gavin's point was that it was mostly Jordan. If you take away the most important people from an investment shop or any business, you won't get the same results. People matter.
Jordan, though, also didn't win another championship. The process mattered to him too. For autostereograms, basketball and investing, it's about matching the right people with the right process.
A new way of looking at the world
Investors like to think of their edge boiling down to some repeatable and external anchor. "I can call Dominos franchises to see if they've sold more pizzas this month than last before the market figures it out." This would give an investor informational edge. This type of work though tends to attract a lot of scrappy investors and end up incorporated into share prices so that if you did get an edge from it, it would fade. The market already knows. Prices already reflect it.
There are two other more rugged ways to gain an edge:
Behave — keep your head when others are losing theirs
Perception — finding an investing angle others miss because of how you see it.
Most of the time, in most names, there's no edge. The market does a good job of pricing stocks and bonds. Most stocks don't outperform treasuries in the long run. All that makes it crucially important — as an investor — how you act and how you see.
Here's a quick case of good behavior and good perception to make this a little more concrete:
A little over a year ago, SK Hynix was just getting its footing coming out of a painful, money losing era. This had scarred the company — it tightened capital discipline — and investors — they were hesitant to jump back into shares.
The AI boom already hit its two-year birthday. Demand for high-bandwidth memory, HBM, soared. SK Hynix was the leader here. The market for HBM was Nvidia and Hyperscalers - Google, Amazon, Microsoft and Facebook. This changed the risk profile of the memory business. It transformed from a cutthroat commodity market keyed on price into a multi-year structured market. This is a much more favorable operating environment.
Frontier labs — OpenAI and Anthropic and DeepMind — meanwhile charged ahead with accelerating capabilities. That meant they needed more compute and more of SK Hynix's higher value memory products at a scale the industry's never seen before. That capacity is now sold out for the next three years.
Most of this was known a year ago. At that point, SK was cheap on a trailing basis. That's fair for a risky commodity business with fickle customers. But looking at a leery investor base and changing market dynamics, SK seemed like a screaming deal.
Abracadabra
Since then? It's up. So far so good.
Did Lamplighter count on commodity memory prices going vertical? No. Did it count on the Cowork moment this past winter throwing even more gas on the AI explosion? No. What it saw was a company knocking down the risks that caused it trouble back in 2022 and previous memory cycles and a memory market rapidly transforming into something that supported much better conditions for much longer.
Lamplighter's outlined a few of the frameworks it uses to see investments. SK Falls squarely into the "Nick Sleep/Economies of Scale" window. These techniques work for Lamplighter to see through the noise. Milage may vary. They won't work for others. But whatever the investment -- whether its successful like SK Hynix has been or not -- they're a crucial foundation to an investment process.
Disclaimer: None of this is investment advice. It's meant to illustrate ways LCM thinks about investing. Things that LCM decides are good investments for LCM and its clients are based on many criteria, not all of which are covered here. Some or all of LCM's ideas may not be suitable for other investors. LCM does not recommend investing either long or short any position mentioned. LCM may own positions in some of the companies mentioned. Some of its ideas will lose money — investing entails risk. See full disclaimer here.