Comment by dismalaf
2 months ago
The most difficult part about the stock market is that stocks don't move for the reasons people think.
People think fundamentals, news, financial performance, or chart patterns matter. But none of it does.
Markets are driven purely by supply and demand. And in this case, supply and demand for the stock itself.
Fundamentals are already priced in. Financial performance is already priced in. Expectations are priced in, and others probably have more information than you anyway. Other people also watch chart patterns and place their bets based on that.
What you're betting on is whether the market is right, or wrong. Or whether other investors will think the market is right or wrong.
Edit - the other part of it is that Wall Street firms hire tons of people who are incredibly smart, have math, economics and finance degrees or even PhDs, hire software engineers and spend hundreds of millions on infrastructure just to know the market better than you. You're not smarter than all those people put together.
In the long term fundamentals will matter. But the market can remain irrational longer than you can remain solvent.
Ordinarily I would say that an investment in a broad index for the long term (a decade or more) is always wise, and there's no point in trying to time it more specifically than that. However right now it's been irrationally exuberant for a long time. it's going to have to correct itself... Either tomorrow or another ten years.
Depends. If the fundamentals are already good, they're priced in, you probably have little upside, even long term. Maybe you'll just match the market. You have to think the fundamentals will improve over time more than the market expects to beat the market.
Long term, I think economics has more to do with market performance. Stocks rise as the money supply expands and capital has no where else to go.
You can't beat the market but you don't need to. The ordinary returns of GDP growth are pretty good.
The free money has definitely been screwing with GDP as a base. I'm still unclear on why the Fed would lower interest rates when the market is fine, the GDP is fine, and inflation is still slightly above baseline 2%. I know everyone is yelling for the punch bowl to come back but now would be a good time to pay down some debts.
2 replies →
The market is not efficent. The fundamentals are only partialy priced in. However finding the times when the difference is significant is not easy.
The market is definitely efficient. Try to find arbitrage opportunities (which is what actual market inefficiency is); they disappear in milliseconds.
What you're saying is that it's not infallible as far as predictive power is concerned, which is true. But every current bit of knowledge is definitely priced in.
Markets are efficient over long time spans but in short term are definitely inefficient. You can just look at volatility during earnings and lack of liquidity in small caps.
I don't think it's as easy as "supply / demand" and everything is priced in.
There's many forces at play here, and many different ways to play the game.
Markets are not that efficient, not everything is priced in. Earnings would have 0 effect on price if that was the case for example.
Whether you play the market on higher timeframes or on lower timeframes, different factors come in to play.
I think it's obvious that chart patterns are a thing, (shameless plug here: i'm building https://edgefound.xyz, an app to create and backtest complex trading setups). It's actually interesting how many patterns (or setups) repeat across markets and timeframes. Mostly because prices are set by traders and traders react in the same way to the same events / price movements.
> Expectations are priced in, and others probably have more information than you anyway.
I have always made money on the stock market. I bought Nvidia in December 2022 and then again in April 2023. Yes the information was out there - anyone reading this board back then could see it. Or who saw Emad Mostaque's tweets that Stable Diffusion was trained on Nvidia, or who knew the semi-public information that OpenAI trained on Nvidia. I bought and it quadrupled in two years when averaging out. The expectations on this board were not priced in the market then.
When FAANG and more b2b tech companies (Salesforce, Oracle) tanked in 2022 I poured money into tech indexes like IYW and IGV. Not as much as into Nvidia, but they were good companies and finally and unusually cheap. They went way up too.
I was making money in the 1990s buying SGI, Sun Microsystems, Netscape, Cisco etc. No brainers for me.
Only two losses I recall - Cascade, a router company whose stock dipped and was bought ny US Robotics on the dip (1990s?). Also some ad company which was crushed by competition with Google.
I completely disagree with your premise. I don't even know how to make a neural network in Pytorch, nor can I do matrix math with NumPy. But in December 2022 Nvidia seemed worth the risk, and by April of last year a lot of that risk evaporated with a huge potential upside. I knew the stock was cheap and all the PhDs on Wall Street did not. Also, they are stuck in a bureaucracy - I used to work in IT for a major investment bank in Manhattan. Watch the Big Short on the pressure the people who were right got from their management and investors. I don't have that pressure, it is my money.
We, the people here, do get insights which are not priced in. It's not that there is no risk, but the coin flip sometimes goes 51% in our favor. Sometimes even 52% (or 53%...or 60%). This is how Warren Buffett, Peter Lynch etc. made money.
Read the Wall Street analyses from the middle of last year on AMD being a big Nvidia competitor, then read the posts here this week about how AMD engineers don't have boards to fix bugs. "We the customer had to mail the AMD engineer a spare board to fix the bug". The cobbler's children wear no shoes at AMD. Then go read some Wall Street report about the stiff competition Nvidia faces from AMD. The information is here, not in the Wall Street analysis report by people who know less than me about how transformers work.
> Read the Wall Street analyses from the middle of last year on AMD being a big Nvidia competitor, then read the posts here this week about how AMD engineers don't have boards to fix bugs. "We the customer had to mail the AMD engineer a spare board to fix the bug". The cobbler's children wear no shoes at AMD. Then go read some Wall Street report about the stiff competition Nvidia faces from AMD. The information is here, not in the Wall Street analysis report by people who know less than me about how transformers work.
Are you saying that AMD will not be a serious competitor to NVIDIA?
I am saying several different things. For your question, the farther out in the future, the more uncertain it is. The stock analyses from the middle of last year which talked about the threat of AMD in a more immediate tone were obviously wrong. Some of the analysis now that talks about the short-term threat of AMD does not jibe with what programmers here working with AMD have said in threads this week. Or what George Hotz or others are saying about buggy AMD chips.
One point is we here know certain things here that stock analysts looking at the last quarterly report do not. The short-term threat of AMD was certainly overstated by Wall Street last year, and in my opinion still is.
If we stop talking in terms of months and start talking in terms of years, it is harder to tell. Maybe AMD will get its act together and give Nvidia a run for its money. Maybe not. For 2025, I don't think AMD will have much effect on Nvidia though.
The second company benefiting from AI Broadcom, not AMD.
For every story like this, there’s another story where everything moves in the opposite direction.
https://xkcd.com/1827/
This is known as the Efficient Market Hypothesis, and having an edge against this requires specialization and focus in specific industries in which you have significant knowledge. That, or learning to recognize when the market is surprised by an outcome, and acting fast.