what about instead of a coin flip mkt is up go long/mkt is down go short(regardless of how up or down)intraday or some sort of short term MA for swing trading....tight stops and a 2-1 min win/loss ratio....would be curious to see a monte carlo sim run on this
Fine. I agree. Now what? And how will flipping coins provide any remedy? That is my point. We're just chewing gum here.
if the market trends, and you are on the right side, you will make money if you haven't screwed things up by poor money management. How do you get on the right side? well, with a toss at least you have a 50/50 chance. But I doubt anything will work on a crap table, so you need the market to perform less random than dice. Now, if you could make money at craps betting on how people will bet....
You can't trend follow a truly random 50/50 occurance. You would need to imply some level of non-random behavior in order to trend follow (or as some have said, cut your losses quickly and let your profits run-essentially trend following). Can one of you geniuses explain to me how you think that you can make money on a truly random 50/50 bet even without transaction costs. By definition, you will break-even over time.
LOL... I'm actually getting an interesting response. Most of the guys in ET talk about how money management is very important but they don't seem to see the basis of it. Well... let me expand it a little bit. Under a series of flips there will be runs and after each run the probability of the same side coming out decreases.... 1 / 2^10 or 1 / ( 2*2*2*2*2*2*2*2*2*2) or 1 / 1024 ========== Next... each flip is really is independent of each other. Though there is going to be series of runs, each flip individually will be 50/50 each time. The coin does not have a memory of the previous flip and so forth. Let's move on considering the above. How do you win with a 50/50 Coin Flip system?
Trading 101 The frequency of a win/loss can is a normally distributed random process. i.e. coin flip The payoff or size of price movement is a non-normal distribution and in some instances non-random. i.e. fat tail (pareto-levy distribution) Simply put the chance of the coin flip may be 50:50 but the payoff using trendfollowing may amount to say .50 loss average on each loss and a average win of 1.50. Over time this makes lots of money.
I don't think so. There is absolutely no way to make money on a series of 50-50 events over time with equal payoff/loss per event. All the fancy progressions you try will fail. That's the mathematical bottom line. Wally
Here's a quote from this site: http://internetcasinolist.com/articles/101102.html First of all, let's distinguish betting systems from money management. While money management is also unable to guarantee a win, and has no mathematical impact on the game, it's often not a bad idea for a player to use. Money management is a process of limiting or controlling the amount that you bet. This could take the form of having a stop loss or a stop win number, which means that when you win or loose a certain amount of money, you stop playing. This isn't a bad idea, especially for people who only have a limited amount of money to gamble. Likewise, a stop number can help a person who has a bad habit, like the inability to leave the table. Money management is simply the act of creating a gambling budget and sticking to it. However, like I said, this budget can help you control your finances, but it can not help you win. Actual betting systems are betting plans. What usually happens is the betting system will talk about "varying your unit bet", which means varying the amount that you bet. They will have various rules, depending on the system. Some will tell you to keep increasing your bets is you win. Others will say keep increasing your bets if you loose. You will also see the exact opposite; decrease when loosing, decrease when winning. Some will want you to hopscotch your bets. Bet five units, followed by three units, followed by 1 unit and then five units again. Just take a moment to look at these plans. How could any of them make you a winner, or influence the outcome of the game? They can't. How could varying your bet size matter, unless you are counting cards in blackjack? (Or unless you have a budget concern, but remember that is money management). Increasing or decreasing systems play off the fallacy of the "streak", riding it one way or the other. Imagine this scenario. You are going to bet on a coin flip. You bet tails. The coin toss results in heads. You lost one unit. Should you now double your bet on the next toss? What if heads came out five times in a row? On the sixth toss, should your bet be six times the original? Absolutely not. On that sixth toss, is tails six times as likely to come out? Once again, absolutely not. Every toss is 50-50. There is no mathematical reason to increase the bet, except in some unwise attempt to "get it all back at once". Now transfer that concept to a casino where you are making bets that are way below 50-50. In fact, the odds are against you instead of even. Does it sound like a good idea to increase your bets? Remember that gambling, except for card counting at blackjack, is a series of independent trials. There isn't any short cut to gambling success. The few games that are beatable, like blackjack and poker, require lots of knowledge and experience to beat. And no betting system is going to eliminate the house edge built into the other casino games.
If payoff's were equal that would be a true statement. What I said was that the payoff's were not equal "non-normal distribution". That's pretty much the heart of all trendfollowing methods.