Why should we reduce trade size when losing?

i am referring to the inital post of this thread. i think you should run some testing on your equity curve to find out whether you have nonrandom streaks or not.

IMHO if you do not have them, each day is independent from the previous. thus it makes no sense to leverage up or down. it is like coin flipping. yes, streaks occur, but they are purely random, thus it does not make sense to bet on them. you would simply be trading on something that does not exist, which means that you are wasting trading cost if you increase size or you pay for missed opportunity if you downsize.

for traders whose track record shows streaks it is suboptimal not to play on them. they are simply letting opportunity go.

i am sure it is impossible to make rules of thumb for this. lndq indicated earlier that his falling knife stratgey shows streaks. whenever he has a bad period, odds shift towards huge gains. this spells "streaks". if he makes money in 50% of trades and makes 1.5 gain for each dollar lost and after two bad days this changes to 55% and 1.8 payoff => streaks. if after two bad days the odds are the same, there is nothing to bet on.

you can only tell from your own equity curve. and only from your own, no matter who says what.


peace
 
Quote from mind:

i am referring to the inital post of this thread. i think you should run some testing on your equity curve to find out whether you have nonrandom streaks or not.

IMHO if you do not have them, each day is independent from the previous. thus it makes no sense to leverage up or down. it is like coin flipping. yes, streaks occur, but they are purely random, thus it does not make sense to bet on them. you would simply be trading on something that does not exist, which means that you are wasting trading cost if you increase size or you pay for missed opportunity if you downsize.

for traders whose track record shows streaks it is suboptimal not to play on them. they are simply letting opportunity go.

i am sure it is impossible to make rules of thumb for this. lndq indicated earlier that his falling knife strategy shows streaks. whenever he has a bad period, odds shift towards huge gains. this spells "streaks". if he makes money in 50% of trades and makes 1.5 gain for each dollar lost and after two bad days this changes to 55% and 1.8 payoff => streaks. if after two bad days the odds are the same, there is nothing to bet on.

you can only tell from your own equity curve. and only from your own, no matter who says what.


peace
More...
Thanks mind for your answer :)

As I see it from all the answers so far, it seems that if you are a disciplined system trader, then the "streaks" that occur are "random" and are inevitable part of any trading system, so it makes no sense to change your position size when such "streaks" occur because they do not indicate "poor trading".

However emotions always play a major role in your trading and once they get in your way these losing "streaks" appear to be not so "random". Simply because the more you lose, the more the regret and the wishful thinking will influence your trading. You will become more intolerant to the pain of losing but your wishes will make you take larger positions instead of smaller ones. And we all know what happens then... You just can't help losing :mad:

I am a discretionary trader but when I go back over my trades I notice that only a small part of my losses is due to poor discipline. The major part is indeed "random" and if I had the chance I would take these trades again. Half of my gross loss however is incurred by day trades (I am a swing trader) and as I analyzed these trades I realized what I did wrong (improper stops and position size).

So in the future I will only reduce my position size if I notice that I get too emotional about my trading. As long as I keep my emotions in check, there will be no need to reduce my position size. :cool:
 
you are making very valid points. if your trading depends on how you feel and you feel better sizing down, you must size down. analyze your equity curve with a simple spreadsheet. see the percentage of updays over all. now look only at days following a loss day. and now do it for two preceeding loosing days.

good luck
 
The most dangerous time to trade is when you have been losing, after previously experiencing winning.

You must reduce size to protect yourself from your own emotional state, which during this disappointed period will put you in some very bad trades.
 
Just stay busy and trade minimally.....

Michael B.



Quote from esc_trader:

The most dangerous time to trade is when you have been losing, after previously experiencing winning.

You must reduce size to protect yourself from your own emotional state, which during this disappointed period will put you in some very bad trades.
More...
 
Quote from axeman:

You should only cut back on size, or increase size,
if your analysis shows non-random "streaks" in your
winners and losers.

If the distribution is random, then you should never
increase or decrease size.
More...

Well, obviously you have to decrease size to maintain the same position size in % terms, relative to your account equity.

There is also merit in reducing % position size once losses get beyond a certain point, because there is now a chance that your edge has deteriorated significantly or even disappeared entirely. I don't think PTJ is talking about cutting back because you have 2 or 3 losing days, more like if you are down 10%+ you should probably reduce your % position sizes.

Regarding the casino analogy, the probabilities at roulette or slot machines are controlled by the casino, so they can be much more confident that their edge is still intact. Even then, if they see that one particular roulette wheel is losing consistently, or one blackjack player is consistently making money off them, they will check the wheel and/or player to make sure their edge has not gone negative (e.g. a "bent" wheel or a card-counting blackjack player).
 
Quote from abogdan:

I disagree. The trade size should be only determined by your money management system. It has nothing to do with the fact that you lost or won the last trade.
More...

This begs the question - PTJ's whole point is that your money management system should be designed so as to increase position size when you have a winning streak and reduce it when you have a losing streak.
 
Try an experiment with a ratio of lightengin up when losing and adding when its in the money . When a stock finally trends you end up making a profit over the small losses you had before. 6 days of churning and small losses but on the day when the stock finally closes at the high and trended all day you make back all and some... i geuss all easier said than done... but you always see people get wiped out as they double down on their losses.. true they usually win 7 out of 8 days but on the day the stock keeps dropping and never comes up they get wiped out. Just do the opposite. You lose 7 out of 8 days but on the time it does end at the high.. you make a good profit!
 
Quote from neutrino:

"Decrease your trading volume when you are trading poorly; increase your volume when you are trading well."
- Paul Tudor Jones

I am sure that this rule is required to achieve a long term success in trading. However I am having a little difficulty in following it strictly because I don't understand very clearly what is the reason behind it... Why would I make more money (or lose less) if I follow it?

Such rule can be justified only if we accept the existence of trends in a trader's equity curve. If you believe that such trends do exist (WHY?!?), then do you "trade" your equity curve?
More...

Physics, Momentum theory and reality.....

those are only a few reasons
 
You must log in or register to reply here.
Back
Top