Are the following statements true and why?
(1) All trading strategies with no cummulative profit stop or quit time limit have a 100% risk of ruin regardles of parameters and path.
(2) Trading strategies with a cummulative profit stop or quite time limit and win rate less than 50% have risk of ruin 100% for any risk-reward ratio, depending on path.
Any reading material you can suggest?
Risk of ruin
-
aaron
- Posts: 0
- Joined: Thu Jan 01, 2004 12:00 am
Risk of ruin
1 is a superstrong version of the gambler's ruin strategy, much too strong to be true. There are many versions, one of the simplest is that if you play any gambling game, regardless of how good your edge is, if there is any risk of loss at all, and if you increase your bet sizes proportionately with your wealth but never decrease it, you will eventually go broke.
So if we play a game where we shuffle a deck of cards and you pay X if the cards end up in perfect Bridge order and win X otherwise, and you set X equal to 10^-100 times your maximum wealth at any point in the past, you will eventually go broke. You might as well save some time and shuffling, just give me all your money now.
2 I don't understand at all, but if it makes sense it's probably false.
So if we play a game where we shuffle a deck of cards and you pay X if the cards end up in perfect Bridge order and win X otherwise, and you set X equal to 10^-100 times your maximum wealth at any point in the past, you will eventually go broke. You might as well save some time and shuffling, just give me all your money now.
2 I don't understand at all, but if it makes sense it's probably false.
-
Steve Castle
- Posts: 0
- Joined: Thu Jan 01, 2004 12:00 am
Risk of ruin
my take:
2. Probability of getting stopped out at a loss is larger than probability of stopping out with a gain, since position is a net loser.
you might win some and take profits on a stop, but you will be a loser in the long run if you continually repeat the strategy even with stops since you still get stopped out with a loss more often.
2. Probability of getting stopped out at a loss is larger than probability of stopping out with a gain, since position is a net loser.
you might win some and take profits on a stop, but you will be a loser in the long run if you continually repeat the strategy even with stops since you still get stopped out with a loss more often.
in the words of one such quant ‘were on the whole either less quanted or not quanted at all’.
-
Nonius
- Posts: 0
- Joined: Thu Jan 01, 2004 12:00 am
Risk of ruin
1 contradicts Kelly, which does not entail a cumulative profit stop or quit time.
2.. Too vaguely framed of a question.
in any event, both statements are way too general to say anything, other than the 1st is clearly not true for every strategy and every market.
2.. Too vaguely framed of a question.
in any event, both statements are way too general to say anything, other than the 1st is clearly not true for every strategy and every market.
Chiral is Tyler Durden
-
Aceoface
- Posts: 0
- Joined: Thu Jan 01, 2004 12:00 am
Risk of ruin
2 is false. It depends on the average win vs the average loss. If probability of winning tradesXaverage win>probability of lossXaverage loss then the system has positive expectancy over the long term (given proper money management as always)