Risk reward ratio and winning odds

Discussion in 'Trading' started by traderzhangSan, Feb 28, 2010.

  1. My thread of how not to blow up draw some attention. Now I want to discuss risk reward ratio.

    My assumption is that you have no edge in trading and every trade you enter has 50% chance of win.

    The intuition tells me by setting up wrong r:r would ruin your chance of win. obviously if your r:r is 1:2 and you still get 50% chance , u will make money for sure. it is very obvious your winning odds is less than 50% if your r:r is 1:2.

    there is something deeper here, how to set up r:r according to price volatility is something can be done here.


    So even you have no edge at all, but doing some math (analyzing volatility), you probably slightly increases the winning odds. I just don't know how to do it yet.
     
  2. As I stated previously your assumption on 50/50 is flawed. You think it's 50/50 before Non Farm Payrolls number is released? You are kidding yourself.

    Put all your money in a money market account and forget about trading. Save yourself the heart attack and the money.

    By the way, good luck analyzing volatility like that. Take a look at the hourly ES Candles from late 2008 and you tell me if you can transform that into any model you analyze.

    Best bet for people like us is to stay out of the markets until volatility is at or under 25 IMO. We are in a high volatility market according to historical standards.
     
  3. lescor

    lescor

    Poor assumption. The outcome of every trade is random, but that is not the same as 50/50 odds. If you flip a loaded coin, the odds are not 50% it will land heads, but the outcome of each flip is still unknown and random.
     
  4. And that is IT in a nutshell, but my money is on the OP to try to drag this out for at least 10 pages:p


    NiN
     
  5. In The Master Swing Trader by Alan Farley, he said beginners would need a 1:3 or even 1:4 risk/reward ratio.

    He did not state the reasons why however.

    I'm getting to a point though where I'm starting to believe a contrarian strategy might work in the short term (intraday) and on a daily chart, trend following might work better. I'll see in the following months.
     
  6. We traders thrive on volatility. 2008 was a great year for trading, especially in Q4. It was common to see Dow swinged 400-500 points a day. As 2009 went on, volatility grinded down quite a bit. Dow doesn't usually swing over 200. I don't believe we are in a high volatility right now. Perhaps depends on how far back you want to look.

    We don't trade the way our grandparents' generation traded.
     
  7. If we look at each coin flip independently, then the outcome for the next landing (assuming a fair coin) is 50/50. But I think if we examine the history of the recent coin flips, then the odds of the next outcome changes.

    For example: if the casino lets me only bet occassionally and not at every toss, then I would wait until there has been a series of head-head-head toss, or better yet a series of head-head-head-head toss, then I will bet the next one to be tail. I think the odds of having a 5-in-a-row head-head-head-head-head is 1/2 x 1/2 x 1/2 x 1/2 x 1/2 = 1/32 and I will have a higher chance of winning. (Assuming a fair coin.) If it is head again, I woud double my bet for the following toss. If it is head again, I would quadruple my bet. Etc. (But unfortunately the casino has table limits and I may have limited funds.)
     
  8. I really doubt everyone shares your opinion about volatility and 2008.

    History is continually repeating itself over and over in the market, I am guessing not too much has changed except for the technology supporting us currently.
     
  9. no offense. But from what I read from text book and what I was taught in university, you have no basic training in probability and statistics whatsoever.

    but I could be wrong, what I was taught and what I learned could be total BS and waste of time.




     
  10. tgtrader

    tgtrader

    I'm sorry but this is wrong. Each coin flip is an independent trial, heck, its commonly THE definition for independent trials in elementary statistics. You can make absolutely no assumptions of what your next coin flip will be based on the prior 1 or 1,000 flips.

    To go off your example, lets say you flip the coin 4 times and you get heads-heads-heads-heads. What are the chances of you getting heads 5 flips in a row? As you stated... (1/2) * (1/2) * (1/2) * (1/2) * (1/2) = 1/32.

    ...now, another question. What are the odds that you get heads 4 times in a row and a tail on the 5th flip? (1/2) * (1/2) * (1/2) * (1/2) * (1/2) = 1/32, exactly the same. What about heads-tails-tails-heads-tails? Still (1/2) * (1/2) * (1/2) * (1/2) * (1/2) = 1/32.
     
    #10     Mar 1, 2010