Would you trade a coin toss?

[Splitlink]

Random entry in a strong trend should be very profitable providing that it is based on a good risk reward ratio for the stops and profit targets. Using the example you gave, imaging that we are going to trade a random entry in a stock (or whatever) in a strong downtrend.

If heads means go long and tails means go short and we place a stop at 10 points and take profits at 40 points, then we have a random entry with a good r/r ratio. We have already assumed the price is going down and because of the random entry we can assume that we go long 50% of the time and go short 50% of the time, but when we are long we lose 10 points, but if short we win 40 pints, we should end up 30 in profit * the number of trades minus spread and commission. Clearly in practice it wouldn't always work like that, so I wouldn't recomend trading on a random entry. The theory behind it is very encouraging, there are just too many assumptions and not enough control over the system.

Wouldn't it be better to "sell" and "don't sell"? That way you are eliminating trading long in a downtrend.

Split

 

[danfreek]

if you knew you were in a downtrend then yes. But if you know you are in a downtrend then you are better off just selling and waiting for the end of the trend. The random entry does not require you to know that you are in a downtrend, or even that you are in a trend at all. If the trend suddenly changed to up then the rules would still work with the same performance, whereas a system tailored purely for a single direction would simply stop working and simply produce losses.

I do not think that random entry is a good strategy, but it is one way to trade that completely automates the decision process and removes human error.

 

[Rhody Trader]

Don't think that can be done. Do you believe that a toss of the coin should give a 50/50 chance of winning? If you do, then enter the spread into the equation and the odds are against you.

The question being addressed here (unless I am mistaken) is not whether you would bet on a 50/50 profitability system, but if you would flip a coin to pick your trade direction. The two are not the same. In the latter, you enter randomly with no notion of whether the trade will end up profitably or not, regardless of the outcome of the coin toss.

Would you toss a coin on which direction a bus will run away on a hill? The same with a share in medium or long term trend. I'd try to find a better system than that because the fact that you know the trend is a factor in your favour. I suppose you could toss a coin on when you will sell. Will I get a better price today or tomorrow, for instance, but I think that it is imprudent to go long when the share has been dropping for days.

The bus example is completely irrelevant to the question. You know with 100% certainty which way the bus is going to going. No matter what how strong a market trend may be, you cannot say with 100% certainty that it will persist.

I'm not saying that a coin toss is your best system. I too would like to think one could do a better job of entry selection. The real discussion point here is that if one follows the principle of riding the winners and cutting the losers, and uses a system which maximizes that, one should be profitable over the long run with a random entry system. If you can then add entry rules which put the directional odds more in your favor, you would have a very good system.

 

[Splitlink]

I'm not saying that a coin toss is your best system. I too would like to think one could do a better job of entry selection. The real discussion point here is that if one follows the principle of riding the winners and cutting the losers, and uses a system which maximizes that, one should be profitable over the long run with a random entry system. If you can then add entry rules which put the directional odds more in your favor, you would have a very good system.

I don't know how to answer this. I'm not, by nature, given to tossing coins because 50/50 ,less spread , simply tells me that I am going to be a loser over an "x" number of times. I don't think that adding entry rules are going to be much use if the final result depends on the toss of a coin. On the other hand, I can think of any number of failures I've had that I might just have well tossed a coin. Trading is only a numbers game, anyway, but it is the persistance of doing the same kind of trade over and over until you are convinced that it is wrong before you try another system that coin tossing or random entry cannot give.

Split