here is proof that it can be done

Discussion in 'Trading' started by jonnysharp, Aug 4, 2006.

  1. 50% of whats wagered in the markets go to brokers as commissions and "service charges". The rest to the occasional lucky winners who lucked out at getting in on the trend before it turned counter trend on them.

    Efficient markets = THERE is MONEY TO BE MADE as a BROKER. Do you know how many futures brokers there are, there are like 100s of them listed on those exchange websites. Smart ones wouldn't prefer to trade, but will let you be the trader.. they would prefer to sell books, dvds, services, market access etc to you. yeah.. markets are really EFFECTIVE & EFFICIENT for these folks.
     
    #31     Aug 4, 2006
  2. Hey, look what I found out? Historically Mondays tend to be negative for stocks 65% of the time since 1988. Those are not edges, those are extremely coincidental anectdotes that can be used to sell books by those scientist types who like to BACKTEST and research but fail completely in knowing how to trade.
     
    #32     Aug 4, 2006
  3. These guys dont even play the markets, because they cant.
    The fact that they sit back and make commish adds nothing to the argument. They sell books and dvd's because they cant trade.

    If I were god, and mid-day monday, I snapped my fingers and GOOG was suddenly priced at $10, what do you think would happen?

    That gap would be closed by massive buying, probably the same day.

    Now who made money? All the traders who went long instead of short because the smart ones realized how horribly mispriced GOOG was at $10. They rushed through the doors with market orders and grabbed every bit they could until it rocketed back up to $300.

    Mispricings like this occur every single day on a smaller scale, and traders rush in, and scoop up the shares. This is exactly what makes the market so efficient. Therefore, efficient markets prove that someone is making money in them.

    Your problem is, you either havent figured out how to identify mispriced stocks/futures, or you are too slow, and better traders snag all the juicy shares before you do.


    Now back to my question, how many backtests on trading strategies have you done in the last 3 years????????????




     
    #33     Aug 4, 2006
  4. I understand not all BACKTESTS are in vane, there are some legitimate ones where you can come up with some real edges.

    My trading is too short-term for backtesting, and I just don't believe in it. I only belive in forward testing. Yeah, If I was a swing trader, I know I would be trading a system that backtests well.
     
    #34     Aug 4, 2006
  5. Ok, that is a tiny crumb, of nearly worthless info. What has the monday bias been in the last year? 2 years? 3 years? 5 years?
    How about hour by hour for the last year?

    Can you think of any way to use this bias? No, that isnt a trading system, and barely an edge of any kind, but it could be. MAYBE.

    Discovering something like that is like getting the tip of your pinky toe in the water. Now dig deeper. A lot deeper and dig up something you can actually trade.


    "those are extremely coincidental anectdotes "

    How do you know this? Do you have statistical proof this is a mere coincidence? Do you know how to test if this is purely a coincidence or a statistical fact of market behavior????



     
    #35     Aug 4, 2006
  6. Now I see your problem. You dont understand backtesting, and dont know how to do it right.

    95% of my backtesting is done in short timeframes. There goes that theory.




     
    #36     Aug 4, 2006
  7. Ok, my last post was a bit too harsh.

    Let me rephrase, it seems to me you are still in the very early stages of becoming a master backtester.
     
    #37     Aug 4, 2006
  8. Correction, when I said I did over 1 million backtests in my first 3 years, I mean, 1 million portofolio backtests. In other words, each "backtest" includes nearly every symbol in the NYSE or NASDAQ. I dont test NYSE and NASDAQ together, they behave too differently.

    If I count each individual symbol as a seperate backtest, then the number is closer to 5 billion backtests in 3 years.
     
    #38     Aug 4, 2006
  9. Na.. I got enough programming skills to do good enough backtests. And I have done some, but not to the extent that I would've had I believed in backtesting. lol..


     
    #39     Aug 4, 2006
  10. How are you ever going to reach the point where you "believe" in backtesting, if you dont backtest enough to find a kick ass trading system?

    I really dont like the word "backtest" myself. I run "simulations" of trading strategies.
     
    #40     Aug 4, 2006