10 ms·

Ask HN: Stocks with a Pocketful of Money

3 points by fuminides 6y ago

Hello everyone, I am a PhD student with CS background. Recently, I have been interested in how some stocks perform, and how some traders perform their decision making. I have seen some papers regarding this question, but I found that a surprising amount of the authors never bet themselves anything. So, as to really understand their thinking process, I decided to start investing a pocketful amount of money (let's say 100$). How should I do it? Where should I start? Of course it would be nice to earn a little bit of cash in the process, but it is not a realist goal. Thank you all.

llampx 6y ago
www.reddit.com/r/wallstreetbets
- superbcarrot 6y ago
  I still can't tell how much of that place is for real and how much is memes and trolling. I suppose that's a compliment to the quality of their trolling.
  - llampx 6y ago
    I think most of it is real, but I take the extreme gains/losses posts with a pinch of salt. Reddit attracts many people who have nothing better to do than make up stuff and its not only wallstreetbets that suffers from it.
- fuminides 6y ago
  I find the subreddit funny but I think that the level of knowledge/memeing is too high for me right now
  - llampx 6y ago
    It was meant as a joke of course, but the subreddit actually helped me understand quite a few things about the stock market. I first got into options in February 2020 and made good gains with the help of WSB since as a beginner I really had no clue where the market was going next. Also it helps unearth some good tickers sometimes. You just have to go through every post with a critical eye and do your own research. To this I would add that Fundamental analysis is only a part of stock trading, and being able to read the various technical analysis indicators is also important to get a good entry (and exit). Some stocks go to the moon and some burrow into the earth but most just go up and down in a range. On WSB I remember someone in March 2020 who bought Amazon $3000 01/15/2021 call options very cheap, when Amazon was around $2000. This person was sitting on an incredible gain in July/August 2020, Amazon was up around $3600. However if they hadn't exited they would have seen the option lose money until last Friday, when Amazon ended up around $3100.
superbcarrot 6y ago
> but I found that a surprising amount of the authors never bet themselves anything. This should be a giveaway. Trading individual stocks or derivatives is gambling. If you have some spare cash and want to gamble, that's fine but it's helpful to be explicit about what you're doing.
- fuminides 6y ago
  Yes, I mean of course randomness beats any algorithm or applied theory, but it is not uncommon to find statements like "this algorithm gain x% in this stock simulator", whilst no real world validation was actually performed. I think that putting a little bit of "skin in the game" might help me better understand how the stock buying/selling/predicting actually works.
  - superbcarrot 6y ago
    In 2020 Bridgewater Associates which is the largest hedge fund in the world lost double digits on their main fund. They spend billions on research, have a lot of experience and many talented people working for them. At the same time people who yoloed on Bitcoin and Tesla from their phone app while sitting on the toilet have made multiples of their original investments. The gap between theory and reality is so big that you're wasting your time reading papers about this. Unless you want to do it for a living - in that case read up as much as you can, your technical background and PhD will help get a job in finance, and then you can be one of the people who pretend to know what they're doing.
    - fuminides 6y ago
      Used* to! I stopped some time ago, but the idea of putting something "real" (but small) into real practice still bugs my mind. I am no more interested in getting a job in finance as I am in getting a well payed job in general, and I still have 1.5 years of full time PhD thesis, so it is not a very short-time goal either. Thanks for the insights, by the way.
- gus_massa 6y ago
kqr 6y ago
First of all: if you just want to have relatively good returns at relatively high risk, put your money into an index fund with low or zero fees. These will get you market returns with no effort from your part. You're unlikely to do any better than this by holding individual stocks yourself. Evidence: there are professionals who try to actively manage portfolios. By definition, half of them get worse than market returns. Half! And these are professionals! With teams and research and 40+ hour work weeks put into this. ---- As someone else pointed out, the stock market is gambling. This doesn't mean you should avoid it, it just calls for utmost responsibility. There are at least three fundamental rules of responsible gambling that you must follow. - Rule zero is to never bet unless you have an edge. If the tables are tilted against you, you will lose. This is probably going to be more obvious after the next rule. - Rule one is to gamble a lot. Once you have decided you want to earn money through taking risk, you must take risks over and over again. That is the only way to make sure you approach the expected value. You can't control individual outcomes, but you can expose yourself many times and let the law of large numbers control the end result. (If you do this when you don't have an edge, you guarantee loss.) - Rule two is to make different bets, not the same bet over and over. This is because you might not have the edge you think you do, so by making the same bet over and over you might, accidentally, work hard toward a negative expected value. - Rule three is to not over-bet, and to try to avoid under-betting too. There is such a thing as a perfectly-sized bet to maximise the end result. Find out about the Kelly criterion. It tells you how much of your capital you should put into a bet based on what your edge is. (But again, it only works with repeated exposure, and when you re-invest your winnings.) ---- Translated to speculating in stocks, this means, roughly, stash most of your capital away somewhere very safe. Bonds, treasury bills, savings accounts, you know. I'm not going to tell you to hunt for under-valued stocks. We can just pretend they don't exist. If there's anything humanity is really fucking good at, it's determining the value of stocks. There are many things you can be one of the best in the world at, but evaluating publically-traded equity is not one of them. So what to do then? With small amounts of capital, buy stocks you think are in the beginning of an exponential growth phase. These are hard to find, because the beginning of exponential growth looks anything but exponential. Early exponential growth looks constant. Stagnated. Underwhelming. Most of the stocks you find will at best break even, or more probably decrease in value. This is why it's important that they're on an exponential trajectory, because decreasing value on an exponential curve is actually not that dangerous, because the decrease is slower and slower as it loses value. If you're lucky, a few stocks will grow instead. Growth on an exponential curve will also go slow at first. Maybe for months, or years. But if it takes off, then it goes wheeeeee—SPLAT, once it hits some natural limit. The stock will keep going up, because now the momentum traders and index funds are buying it, but this is hollow value that will disappear once people realise the underlying growth is no longer there. You step out before the SPLAT but not before the wheeeee. If you can do that, just a few of these will pay off all your other loser stocks. It is important that you follow rule two here: don't buy several similar stocks, because they will rise and fall together. The market is more correlated than most people think.

3 points by fuminides 6y ago

llampx 6y ago

www.reddit.com/r/wallstreetbets

superbcarrot 6y ago
I still can't tell how much of that place is for real and how much is memes and trolling. I suppose that's a compliment to the quality of their trolling.
- llampx 6y ago
  I think most of it is real, but I take the extreme gains/losses posts with a pinch of salt. Reddit attracts many people who have nothing better to do than make up stuff and its not only wallstreetbets that suffers from it.
fuminides 6y ago
I find the subreddit funny but I think that the level of knowledge/memeing is too high for me right now
- llampx 6y ago
  It was meant as a joke of course, but the subreddit actually helped me understand quite a few things about the stock market. I first got into options in February 2020 and made good gains with the help of WSB since as a beginner I really had no clue where the market was going next. Also it helps unearth some good tickers sometimes. You just have to go through every post with a critical eye and do your own research. To this I would add that Fundamental analysis is only a part of stock trading, and being able to read the various technical analysis indicators is also important to get a good entry (and exit). Some stocks go to the moon and some burrow into the earth but most just go up and down in a range. On WSB I remember someone in March 2020 who bought Amazon $3000 01/15/2021 call options very cheap, when Amazon was around $2000. This person was sitting on an incredible gain in July/August 2020, Amazon was up around $3600. However if they hadn't exited they would have seen the option lose money until last Friday, when Amazon ended up around $3100.

superbcarrot 6y ago

> but I found that a surprising amount of the authors never bet themselves anything. This should be a giveaway. Trading individual stocks or derivatives is gambling. If you have some spare cash and want to gamble, that's fine but it's helpful to be explicit about what you're doing.

fuminides 6y ago
Yes, I mean of course randomness beats any algorithm or applied theory, but it is not uncommon to find statements like "this algorithm gain x% in this stock simulator", whilst no real world validation was actually performed. I think that putting a little bit of "skin in the game" might help me better understand how the stock buying/selling/predicting actually works.
- superbcarrot 6y ago
  In 2020 Bridgewater Associates which is the largest hedge fund in the world lost double digits on their main fund. They spend billions on research, have a lot of experience and many talented people working for them. At the same time people who yoloed on Bitcoin and Tesla from their phone app while sitting on the toilet have made multiples of their original investments. The gap between theory and reality is so big that you're wasting your time reading papers about this. Unless you want to do it for a living - in that case read up as much as you can, your technical background and PhD will help get a job in finance, and then you can be one of the people who pretend to know what they're doing.
  - fuminides 6y ago
    Used* to! I stopped some time ago, but the idea of putting something "real" (but small) into real practice still bugs my mind. I am no more interested in getting a job in finance as I am in getting a well payed job in general, and I still have 1.5 years of full time PhD thesis, so it is not a very short-time goal either. Thanks for the insights, by the way.
gus_massa 6y ago

kqr 6y ago

First of all: if you just want to have relatively good returns at relatively high risk, put your money into an index fund with low or zero fees. These will get you market returns with no effort from your part. You're unlikely to do any better than this by holding individual stocks yourself. Evidence: there are professionals who try to actively manage portfolios. By definition, half of them get worse than market returns. Half! And these are professionals! With teams and research and 40+ hour work weeks put into this. ---- As someone else pointed out, the stock market is gambling. This doesn't mean you should avoid it, it just calls for utmost responsibility. There are at least three fundamental rules of responsible gambling that you must follow. - Rule zero is to never bet unless you have an edge. If the tables are tilted against you, you will lose. This is probably going to be more obvious after the next rule. - Rule one is to gamble a lot. Once you have decided you want to earn money through taking risk, you must take risks over and over again. That is the only way to make sure you approach the expected value. You can't control individual outcomes, but you can expose yourself many times and let the law of large numbers control the end result. (If you do this when you don't have an edge, you guarantee loss.) - Rule two is to make different bets, not the same bet over and over. This is because you might not have the edge you think you do, so by making the same bet over and over you might, accidentally, work hard toward a negative expected value. - Rule three is to not over-bet, and to try to avoid under-betting too. There is such a thing as a perfectly-sized bet to maximise the end result. Find out about the Kelly criterion. It tells you how much of your capital you should put into a bet based on what your edge is. (But again, it only works with repeated exposure, and when you re-invest your winnings.) ---- Translated to speculating in stocks, this means, roughly, stash most of your capital away somewhere very safe. Bonds, treasury bills, savings accounts, you know. I'm not going to tell you to hunt for under-valued stocks. We can just pretend they don't exist. If there's anything humanity is really fucking good at, it's determining the value of stocks. There are many things you can be one of the best in the world at, but evaluating publically-traded equity is not one of them. So what to do then? With small amounts of capital, buy stocks you think are in the beginning of an exponential growth phase. These are hard to find, because the beginning of exponential growth looks anything but exponential. Early exponential growth looks constant. Stagnated. Underwhelming. Most of the stocks you find will at best break even, or more probably decrease in value. This is why it's important that they're on an exponential trajectory, because decreasing value on an exponential curve is actually not that dangerous, because the decrease is slower and slower as it loses value. If you're lucky, a few stocks will grow instead. Growth on an exponential curve will also go slow at first. Maybe for months, or years. But if it takes off, then it goes wheeeeee—SPLAT, once it hits some natural limit. The stock will keep going up, because now the momentum traders and index funds are buying it, but this is hollow value that will disappear once people realise the underlying growth is no longer there. You step out before the SPLAT but not before the wheeeee. If you can do that, just a few of these will pay off all your other loser stocks. It is important that you follow rule two here: don't buy several similar stocks, because they will rise and fall together. The market is more correlated than most people think.