6 ms·

A couple of things stand right out here: a) There's almost no time horizon here, so it's basically day trading. It's not enough to know what happens, everyone e

A couple of things stand right out here: a) There's almost no time horizon here, so it's basically day trading. It's not enough to know what happens, everyone else needs to "do the right thing" in a short time horizon. As the article says, that doesn't happen a lot. b) WSJ isn't as good a new source as it used to be, which makes the game harder. Though occasionally, something big will happen and a one day lookback would be helpful. (Say, shorting airline stocks before 9/11...)

petesergeant 2y ago
> it's basically day trading In my understanding day trading is mostly about looking for pretty patterns in lines and ascribing meaning to them, as well as attempting to trade on public information that HFTs and 3,000 well-paid and professional traders at GS have already priced in
- filoleg 2y ago
  What you are talking about (looking for pretty patterns in lines and ascribing meaning to them) isn’t day trading, it is called technical analysis. I mostly agree with your take on its usefulness though, there is a reason it is getting memed as astrology for finance bros.
  - petesergeant 2y ago
    I understand they're different topics, but my understanding is also that amateur day traders go wild for technical analysis over most other signals, so in practice much day trading consists of it...
ac29 2y ago
9/11 is a great example here because even if you knew it would happen a day in advance, trading on that would gain you nothing in this experiment - the markets didnt open that day.
- dheera 2y ago
  Markets in other parts of the world did, and you could have traded those airline stocks. That said, COVID would have been a better example for airline and hotel stocks. [Also IMO, PUT options are safer than shorts. Do your own research.]
  - ghaff 2y ago
    One of the problems is that you mostly need money to make money. There's an SF book called Replay that imagines a guy reliving his life from his fifties or so to his teens or so. One of the clever plot points (not really a spoiler) is that he just happens to remember a highly improbable sports result around the time of his replay which, with the help of some friends and family loans, makes him very wealthy. The book would have been a lot less interesting if the main characterer was just starting out as a poor college student over and over again. Sure, I could have made money in dot-com/bomb (and maybe bitcoin) but I still would have needed a fair bit of capital to have made life-altering gains. To the point of the article, predicting how most specific company or economic outcomes will affect stock prices is pretty much a sucker bet. Perhaps excluding some specific events like 9/11--but even that effect wasn't that great.
    - didgetmaster 2y ago
      Bitcoin is not a good example for your argument. The first bitcoin transaction for a physical good was in 2010 where 10K of bitcoin was used to buy a couple pizzas. Even the poorest American could probably have scraped enough together to own that much bitcoin back then. If they held onto it until now, they would be a billionaire!
      - ghaff 2y ago
        If you really knew Bitcoin was going to be big with certainty/high probability, you could have mined it or done a dark alley transaction early on. Assuming you didn't get scammed out of or simply lost your wallet at some point, I agree, it's an exception where you could have made a fortune from basically nothing. (Though not in a day which is the actual experiment in question.) The more interesting experiment to me would be if you gave me a week's worth (say) of the WSJ from ten years hence, what could I do with it?
Reubend 2y ago
> WSJ isn't as good a new source as it used to be, which makes the game harder. You don't think the front page of the WSJ would contain a decent summary of the day's business news?
- pclmulqdq 2y ago
  It does not. It's a sort of ideological rag at this point for the right wing of Wall Street (see the NYT for the left wing rag). Give them Bloomberg or the AP a day ahead and they will make much better bets. I think they will still trade with too much leverage and poor asset allocation, though.
  - dekhn 2y ago
    The NY Times isn't a left wing rag (I'm excluding the Opinion section, which is a very mixed bag). It's the mouthpiece of the establishment.
    - pclmulqdq 2y ago
      It's the ideological rag for the left wing of Wall Street which is very different than the left wing in general. Both papers are "establishment."
- Etheryte 2y ago
  I wouldn't really say the WSJ is a good summary. If I look at the front page today, most of it is topical, but generic and not really actionable information in any way: - The Drugs Young Bankers Use to Get Through the Day—and Night - CEOs Want Trump to Change Course on Tariffs. He Isn’t Budging. - Untangling America’s Love-Hate Relationship With Corporate Power Etc, it doesn't really tell you much anything about what's going on in the market. Yahoo Finance on the other hand is a great overview, the upcoming fed meeting is front and center, there's a market overview on the right, highlights of specific big movers, etc.
  - alwa 2y ago
ghaff 2y ago
I'm unconvinced that, as you say, reacting to specific short-term events is especially interesting. There may be specific cases where OMG, the stock market is going to fall, is fairly predictable. But the more interesting trends are decade-long+.
- eschneider 2y ago
  This. And even knowing the long-term trend isn't a guarantee that you'll make money. Look at the late 90s: knowing the internet was going to be a huge thing was obvious, but it didn't necessarily translate to knowing who the winners and losers would be long term. You have to both understand the long-term trends AND be smart about the here-and-now.
  - ghaff 2y ago
    You could know the market was going to go silly with stocks like Yahoo in the late 90s. But you also had to pick the right time to dump everything.
    - danenania 2y ago
      You could have also “known” that bitcoin was a massive bubble and going to zero back when that was a popular sentiment. It’s not just timing the bubble that is tricky. Sometimes the “obvious” bubble never ends up popping.
      - ghaff 2y ago
        Markets can stay irrational longer can you can stay solvent and all that. But you could have done pretty well with bitcoin if you got in early so long as you didn't jump out too early (or get scammed).
  - pedalpete 2y ago
    You may be ignoring that at the time of the bubble, many investors, and even regular people, though the internet wasn't that big a deal. Sure email meant we wouldn't send snail mail anymore, but the majority of people were still in the "it isn't safe to use my credit card online" world.
bee_rider 2y ago
I think in the blog post they say they have pulled from Washington Post articles randomly over the last 15 years. When do you think the paper went bad? You can also skip days. If I understand the blog post correctly, maybe you could check your hunch (b) by playing the game and skipping any days after your “WaPo went bad” date.

petesergeant 2y ago

> it's basically day trading In my understanding day trading is mostly about looking for pretty patterns in lines and ascribing meaning to them, as well as attempting to trade on public information that HFTs and 3,000 well-paid and professional traders at GS have already priced in

filoleg 2y ago
What you are talking about (looking for pretty patterns in lines and ascribing meaning to them) isn’t day trading, it is called technical analysis. I mostly agree with your take on its usefulness though, there is a reason it is getting memed as astrology for finance bros.
- petesergeant 2y ago
  I understand they're different topics, but my understanding is also that amateur day traders go wild for technical analysis over most other signals, so in practice much day trading consists of it...

ac29 2y ago

9/11 is a great example here because even if you knew it would happen a day in advance, trading on that would gain you nothing in this experiment - the markets didnt open that day.

dheera 2y ago
Markets in other parts of the world did, and you could have traded those airline stocks. That said, COVID would have been a better example for airline and hotel stocks. [Also IMO, PUT options are safer than shorts. Do your own research.]
- ghaff 2y ago
  One of the problems is that you mostly need money to make money. There's an SF book called Replay that imagines a guy reliving his life from his fifties or so to his teens or so. One of the clever plot points (not really a spoiler) is that he just happens to remember a highly improbable sports result around the time of his replay which, with the help of some friends and family loans, makes him very wealthy. The book would have been a lot less interesting if the main characterer was just starting out as a poor college student over and over again. Sure, I could have made money in dot-com/bomb (and maybe bitcoin) but I still would have needed a fair bit of capital to have made life-altering gains. To the point of the article, predicting how most specific company or economic outcomes will affect stock prices is pretty much a sucker bet. Perhaps excluding some specific events like 9/11--but even that effect wasn't that great.
  - didgetmaster 2y ago
    Bitcoin is not a good example for your argument. The first bitcoin transaction for a physical good was in 2010 where 10K of bitcoin was used to buy a couple pizzas. Even the poorest American could probably have scraped enough together to own that much bitcoin back then. If they held onto it until now, they would be a billionaire!
    - ghaff 2y ago
      If you really knew Bitcoin was going to be big with certainty/high probability, you could have mined it or done a dark alley transaction early on. Assuming you didn't get scammed out of or simply lost your wallet at some point, I agree, it's an exception where you could have made a fortune from basically nothing. (Though not in a day which is the actual experiment in question.) The more interesting experiment to me would be if you gave me a week's worth (say) of the WSJ from ten years hence, what could I do with it?

Reubend 2y ago

> WSJ isn't as good a new source as it used to be, which makes the game harder. You don't think the front page of the WSJ would contain a decent summary of the day's business news?

pclmulqdq 2y ago
It does not. It's a sort of ideological rag at this point for the right wing of Wall Street (see the NYT for the left wing rag). Give them Bloomberg or the AP a day ahead and they will make much better bets. I think they will still trade with too much leverage and poor asset allocation, though.
- dekhn 2y ago
  The NY Times isn't a left wing rag (I'm excluding the Opinion section, which is a very mixed bag). It's the mouthpiece of the establishment.
  - pclmulqdq 2y ago
    It's the ideological rag for the left wing of Wall Street which is very different than the left wing in general. Both papers are "establishment."
Etheryte 2y ago
I wouldn't really say the WSJ is a good summary. If I look at the front page today, most of it is topical, but generic and not really actionable information in any way: - The Drugs Young Bankers Use to Get Through the Day—and Night - CEOs Want Trump to Change Course on Tariffs. He Isn’t Budging. - Untangling America’s Love-Hate Relationship With Corporate Power Etc, it doesn't really tell you much anything about what's going on in the market. Yahoo Finance on the other hand is a great overview, the upcoming fed meeting is front and center, there's a market overview on the right, highlights of specific big movers, etc.
- alwa 2y ago

ghaff 2y ago

I'm unconvinced that, as you say, reacting to specific short-term events is especially interesting. There may be specific cases where OMG, the stock market is going to fall, is fairly predictable. But the more interesting trends are decade-long+.

eschneider 2y ago
This. And even knowing the long-term trend isn't a guarantee that you'll make money. Look at the late 90s: knowing the internet was going to be a huge thing was obvious, but it didn't necessarily translate to knowing who the winners and losers would be long term. You have to both understand the long-term trends AND be smart about the here-and-now.
- ghaff 2y ago
  You could know the market was going to go silly with stocks like Yahoo in the late 90s. But you also had to pick the right time to dump everything.
  - danenania 2y ago
    You could have also “known” that bitcoin was a massive bubble and going to zero back when that was a popular sentiment. It’s not just timing the bubble that is tricky. Sometimes the “obvious” bubble never ends up popping.
    - ghaff 2y ago
      Markets can stay irrational longer can you can stay solvent and all that. But you could have done pretty well with bitcoin if you got in early so long as you didn't jump out too early (or get scammed).
- pedalpete 2y ago
  You may be ignoring that at the time of the bubble, many investors, and even regular people, though the internet wasn't that big a deal. Sure email meant we wouldn't send snail mail anymore, but the majority of people were still in the "it isn't safe to use my credit card online" world.

bee_rider 2y ago

I think in the blog post they say they have pulled from Washington Post articles randomly over the last 15 years. When do you think the paper went bad? You can also skip days. If I understand the blog post correctly, maybe you could check your hunch (b) by playing the game and skipping any days after your “WaPo went bad” date.