Topic: Past Performance Not Indicative of Future Results? (Read 1796 times)

The beauty of this adage is that, in poker, a professional player wins in the long term via a novice player drawing to an inside straight, and hitting (it worked the first time, it'll work again), and by knowing that you only enter the pot with your very good hands, mucking the bad ones. Of course, I'm only speaking about the mathematical aspects of the game and not the psychological ones, for we all know, or you better know, that poker is a people game played with cards, and not a card game played with people. Big difference!

I think this is what happens in the stock market by the brokers. They'll take a certain percentage of their clients money and invest in a losing proposition, then come in at its down cycle and buy with their own money, fully knowing that it'll go up, and soon. The losers, thinking that the stock was going to go up, are re-informed of the adage, during that down cycle. When the smoke all clears, those same losers are happy with their 4-8 percent return on their investments, as a whole, while the brokers enjoy a return rate in double figures.

On the other hand, if somebody came up to me and said that I would receive a 20% return on a 10K loan, and he proved to me that it's entirely risk free, legal, and without any hocus-pocus, I would have to tell him to lick my balls, for I, personally, would lose money with his proposition. In fact, I would lose no less than 80%. He would be shocked and ask me how. I would tell him, whereupon he'll want no part of my deal, although he would enjoy a higher return. But what I do is not his cup of tea. What he does is not mine. Yet both of us do well. Very well!

"Indicative" is the important word here. It implies prediction. Outside of psychics, prophets, and oracles, modern prediction is the exclusive domain of the scientific method. So where in the peer reviewed journals of market economics does it reject this disclaimer?

You're wrong.

You can still lose money on one (or many) deals and outperform the market.  And you don't need to hit every peak and valley exactly.  For example, if you bought 99 bitcoins at 1, sold 1/3 at 10, 1/3 at 20, and 1/3 at 30, you end up with $1980 dollars, a 1900% return on your investment, which is better than if you just held the coins until today and got 200% return.

These measurements need to be medium to long term though.  I'm not saying that if on a single day, you buy when the price drops to 1.0 but the bottom ends up being 0.98, and you sell at 3.0 but the price peaks at 3.02, that you didn't outperform the market.  Well you didn't that day, but that in-and-of-itself is a pretty meaningless number.  If you start comparing your results for the last month, the last quarter, the last year, things start averaging out.  Then you can demonstrate if you do consistently better than everyone else.

This is the measurement you need to use to determine if you're doing better than a hypothetical coin flip.

Define outperform?

Correct me if I'm wrong, but it seems your definition of beating the market involves never making a losing trade ever while also simultaneously hitting every peak and valley exactly..  Either that, or you buy once and the price continues to increase infinitesimally.  

That's a meaningless measure.  The question you need to ask is if you consistently out-perform the market.  If I bought at 2, sold at 2.50, rebought at 2.75 and sold at 3, I've got too profitable deals, but I would have made more by buying at 2 and selling at 3.  I did not outperform the market.

To outperform the market for the past year, you would have to have got more than a 600% return on your investment.  (I'm basing this on the bitcoin price hitting 50 cents in November 2010, I don't have the exact price for Dec 2, 2010.)

After getting some reading and thinking done, I am still not sure what to make of it. Right now I am leaning towards "past performance indeed is NOT indicative of future results."

Here is some food for thought, with the caveat that I am completely uneducated and non-indoctrinated in the field of economics:
(1) price/value of a stock or commodity is mainly influenced by factors other than past performance. This is, I believe, called "fundamentals." This is why insider trading is considered unethical/illegal - obviously knowledge of things other than past performance can help you predict the price trend and profit. Ergo, it's not about past performance, but mostly about plans and business decisions. It's also about completely external factors, such us competitors' moves and inadvertent timing with related products or technologies (the time was/wasn't ripe, etc.).

(2) Bitcoin is presently an oddball because fundamentals and external factors are essentially non-existent.  The economy, apart from SR and maybe some laundering operations is simply not here yet. Therefore, the price is mostly formed by speculative trading, and many traders believe in technical analysis and don't buy the subject line of this thread. Ergo, bitcoin price indeed is to some extent shaped by trends and fashion. Having said that, it is still a complex system (different and changing strategies, especially recently with trading robots; relatively small market cap that enables significant move of the market by individual players).

Fair point.  With respect to the topic, I assume the general interpretation and definition of "past performance" is equatable with price.  So, if the current value of BTC is $3.50, then the market is performing better than if the value were at $3 independent of which direction the market is going or other factors.  That is, I understand the general argument to be something like, "Well, we know on day/time x the price was $3, and on day/time y the price was $4...but this does absolutely nothing to tell us what the market will be like on day/time z." 

I would agree with this logic, but I am suggesting that people who argue that "past performance is not indicative of future results" should expand their definition of past performance to include lots of other variables.  The reason I mention this is because this argument is often coupled with ideas of "traders can't beat the market" or something about chance.  Now I don't know about you, but the number of profitable trades I've conducted far exceeds the number of non-profitable ones, and I don't think it was due to chance.  There's more than a coin flip going on.

You wrote a lot, but didn't explain what you meant by "past performance." Once you define this clearly we can proceed with the discussion.

Oh no?

What genius determined the economy is a one way function?

More importantly, who are the idiots that don't understand that sometimes a system can be so complex that it's much more truthful (and, less arrogant) to simply say, "Well, past performance probably, almost certainly, IS indicative of future results, but I'm too dumb."

The economy isn't the only system in the world and it is by no means closed.  

Oh yeah, another thing...

If you're equating "past performance" with price (cross-sectional data), then this is a whole new issue.  Sometimes a market will go to $5 because it is at $2, whereas the same market would go to $1 if it is at $3.  Sometimes a price is too high and this can have a negative impact on the market for various reasons.  Conversely, sometimes a price is low, but this can have a positive impact on the market because of the context in which the price is embedded.

What does this mean?  The price you see is not a completely reliable indicator of "performance," nor is price simply an arbitrary number -- the price is established because of a multitude of factors, so the price is a reflection of these factors.  Maybe once we realize that 1) Some things are just far too complex for a human to deconstruct, or 2) that the ego doesn't like to be wrong and it likes to project ignorance onto it's environment, we will look back on the phrase in the subject heading and say, "so THAT'S why I always hated school..."