Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Investment return statistics

  1. Oct 17, 2014 #1
    It seems to me that I've come up with a way to almost guarantee that one is a profitable trader. Obviously I'm assuming there is some flaw or it would be widely known as its not that complicated.

    Imagine thahe only knowledge you can ascertain about publicly-traded, liquid asset with constant price corrections due to market forces, is the price. For simplicity, it is infinitely divisible and there are no fees or spreads that you must overcome. With no other knowledge about the asset, there is a 50% chance it will go up and a 50% chance it will go down. you have $100.

    Knowing that the asset has a volatile price, you put $25 in hopes to scalp profit from a small upward movement in price.

    The particular volatility in this market means you will either gain or lose 1% in the time frame you will be invested. Because of the 50/50 odds, this strategy will consistently give you a return/loss of $0 as losses and gains will cancel each other out.

    When you lose 1%, you put the $50 you have left into the asset realizing that the odds of a 1% increase are now 2/3 in order to bring the probability that the asset will increase back to 1/2.

    The returns from this second trade over 3 instances:

    and the returns of the whole $75:

    So the returns of on the assets under both instances would be as follows:
    +1% -> 25+(25*.01)=25.25
    -1% -> 25-(25*.01)=24.75+.75=25.50

    The reason that -1% would yield higher results is because more than double is invested on the upswing.

    Obviously, this theory would not actually work under all market conditions because of how prices tend to continue to go in the direction that they are already going. But in a market in which prices bounce up and down, one can probably come up with even better than a 50% chance of guessing which direction the price will go by using technical analysis (chart-reading) to buy at the established low.

    I apologize if this doesn't make sense as I'm pretty exhausted right now. but I wanted to get some feedback on this. If need be, I'll come back after some sleep and try to fix it.
  2. jcsd
  3. Oct 17, 2014 #2
    You give no evidence for this and I see no reason why there should be increased chance for the stock to go up after it went down.
    It should be easy enough to do some research about this.
  4. Oct 17, 2014 #3


    User Avatar
    Homework Helper

    "When you lose 1%, you put the $50 you have left into the asset realizing that the odds of a 1% increase are now 2/3 in order to bring the probability that the asset will increase back to 1/2.

    Not at all. The long term fluctuations you've assumed have no bearing at all on what will happen in the short term.
  5. Oct 17, 2014 #4


    User Avatar
    Science Advisor

    Just to retierate what others have said: If past performance was a reliable indicator of future performance we would all be filthy rich!
  6. Dec 8, 2014 #5
    Just a real world note on this, without going too deeply down the rabbit hole of the forces that affect public securities prices:

    There certainly are trading strategies that use the techniques of _technical analysis_ (I can explain what that is if anyone is interested.... or just have a look at Wikipedia), that involve contrarian trades in highly volatile stocks. Needless to say, none are magic bullets, and opinions widely vary on their specific implementations and efficacy. The above outlined strategy has several underlying problems, particularly with assumptions about how prices tend to react, but again, there are _technical analysis_ methods that attempt to capitalize on volatility without great concern for the specific direction of price movement...... but that is a whole another discussion.

    Perhaps the most important point is that in real world trading, like real world physics, little or nothing is free or perfectly efficient. There is _always_ a spread, commission or other 'cost of trading.' Cannot be avoided. It is also usually comparatively expensive. How expensive is comparatively? There are many variables, and ways to minimize trading costs, but they cannot be eliminated, and for comparison, trading costs tend to outpace (sometimes by a wide margin) the 'house edge' that one would be in thrall to at a Las Vegas casino.

    This was a sad lesson learned my many who flocked to those (now mostly defunct) 'day trading academies' and the like in the 1990s thinking that with a bit of deft cleverness, watching CNBC, and the application of a few formulas or techniques they could sit in a cubicle and follow the 'Royal Road to Riches.' Uh-uh! This proved to be a losing proposition for just about all involved, with the exception of the promoters who encouraged and facilitated these fiascoes.

    Put another way, if you make a bunch of trades, and otherwise 'break even' on the trades themselves, you still stand to lose a bunch of money just getting back to where you started. There are a number of practical factors that invariably exacerbate this tendency, and that doesn't even take into account the built in disadvantages of the small trader, poor or uninformed choices of investment, mutation factors, etc. Real world investing and trading (which are two different things) involve an extremely complex set of variables and underlying mechanics.

    On the other hand, if one does want a comparatively simple, historically consistent (mostly........ certainly over time), method of outperforming broader market averages, once could check out Dow Dog Theory (and some of its variations). Probably wont make you independently wealthy overnight, but you can expect a meaningful and (again, over time) performance spiff over the broader index performances of its components.


    BTW, some commentators cite a 1991 article for its popularization if not its origin. This is somewhere between a stretch and nonsense. Dow Dog theory (by one name or another) has been around and well known and discussed for many decades.


Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook