Is increasing risk/size after successive trading losses good risk management or bad?

[BlacknBlue]

Wow people, pull up martingale in the dictionary before laying down advice...

Myself and other traders I know have similar strategies concerning when to increase risk but the answer is more than just yes or no. In some breakout trading I use a strategies I may use a riskier system than Martingale since the stocks I trade rarely stay in narrow priced channels; there's usually a clear direction upon confirmation as long as there isn't any surprise news. In some strategies I'll use something similar to D'alembert however I confine this method to long only trades.

A lot of naysayers giving advice presume the market is random walk or zero sum which is not correct. You can literally pull up the daily candlestick charts of an index and see that most days are positive. I don't see professional stock traders strictly trade ~1 unit in every single trade. When traders see something very obvious they usually increase their risk.

Just over 50% isn't good enough though especially over these past few months when all you needed to do is trail buy and sell around close. In fact this can be done over the past 10+ years, mind that you can't be completely blind to what's happening in the news.

I'll ask some questions that are basic things that my analytics team looks for when improving a trader's strategy.

Long win ratio on positions? Short win ratio on positions? Long win ratios on positive days? Short win ratios on negative days?

 

[danielc1]

Wow people, pull up martingale in the dictionary before laying down advice...

Myself and other traders I know have similar strategies concerning when to increase risk but the answer is more than just yes or no. In some breakout trading I use a strategies I may use a riskier system than Martingale since the stocks I trade rarely stay in narrow priced channels; there's usually a clear direction upon confirmation as long as there isn't any surprise news. In some strategies I'll use something similar to D'alembert however I confine this method to long only trades.

A lot of naysayers giving advice presume the market is random walk or zero sum which is not correct. You can literally pull up the daily candlestick charts of an index and see that most days are positive. I don't see professional stock traders strictly trade ~1 unit in every single trade. When traders see something very obvious they usually increase their risk.

Just over 50% isn't good enough though especially over these past few months when all you needed to do is trail buy and sell around close. In fact this can be done over the past 10+ years, mind that you can't be completely blind to what's happening in the news.

I'll ask some questions that are basic things that my analytics team looks for when improving a trader's strategy.

Long win ratio on positions? Short win ratio on positions? Long win ratios on positive days? Short win ratios on negative days?

You are confusing martingale strategy with simple position size strategy. If you risk more on some trades because you 'know' it will be a big winner, has nothing to do with Martingale. Martingale on its own is also not a way to calculate what risk you are taking on in trading. The pure definition of Martingale is adding to your risk, when you have had a loser, to come out above your losses when you win. This does not work in the longterm.

 

[BlacknBlue]

You are confusing martingale strategy with simple position size strategy. If you risk more on some trades because you 'know' it will be a big winner, has nothing to do with Martingale. Martingale on its own is also not a way to calculate what risk you are taking on in trading. The pure definition of Martingale is adding to your risk, when you have had a loser, to come out above your losses when you win. This does not work in the longterm.

No I'm not confused at all. Martingale is a probability system when you double your risk everytime you "lose". Martingale isn't defined as whatever you want it to, sorry. She clearly said she waits 2-3 losses until she increases risk. Her increase of risk isn't even stated however you are assuming she is doubling her risk each time thereafter.

 

[danielc1]

No I'm not confused at all. Martingale is a probability system when you double your risk everytime you "lose". Martingale isn't defined as whatever you want it to, sorry. She clearly said she waits 2-3 losses until she increases risk. Her increase of risk isn't even stated however you are assuming she is doubling her risk each time thereafter.

Okay, no problem, let's agree to disagree on this. But OP should not increase risk after a loss. She stated it was between 25 to 100% more risk. The only reason OP do this, is to make up her losses of previous trade with increasing her risk to double what she normal does. Hence a martingale strategy. If I would ask a question like that on a traders forum, I hope every trader answer to not do that. Even if you are an experienced trader, and even if you use some kind of martingale in some degree, that makes part of a bigger position size system that is anti martingale, you do not want to confuse the novice with going in to a direction of ruin for the OP because you stated as a trader that she can do this.

 

[BlacknBlue]

Okay, no problem, let's agree to disagree on this. But OP should not increase risk after a loss. She stated it was between 25 to 100% more risk. The only reason OP do this, is to make up her losses of previous trade with increasing her risk to double what she normal does. Hence a martingale strategy. If I would ask a question like that on a traders forum, I hope every trader answer to not do that. Even if you are an experienced trader, and even if you use some kind of martingale in some degree, that makes part of a bigger position size system that is anti martingale, you do not want to confuse the novice with going in to a direction of ruin for the OP because you stated as a trader that she can do this.

Agree to disagree on what martingale is defined as? LOL. Also your mention anti-martingale causes additional confusion, and it seems that you don't know what that is even defined as. It's the progression of doubling your bet when you win.

Anyways, let's do some simple math based on the facts she's given us with some averaged presumptions. She says she doesn't increase size until after 2-3rd loss. Let's say 3 losses she doesn't increase risk but lets say she increases position by 65% each time instead of using 50% average.

First three losses = 3 units. Loss 4 = 1.65u. Loss 5 = 2.7u. Loss 6 = 4.5u. Total loss = 11.85u. Compared to martingale 6 loss in a row = 1,2,4,8,16,32 totaling 63u which is over 5.3x greater than what is assumed. Regardless, losing 12u from 6 trades isn't bad as the same result can occur from 12 losing trades at 1 unit - far from catastrophic.

I highly doubt that my previous statements confused her into thinking martingale is OK and that she should go with it.

 

[NinaB]

thanks, and no, i don't think what you said was confusing at all black.

I'm not really following daniel, martingale, anti martingale. i don't think any of those applies to what i'm doing and certainly not my cup of tea. what black mentioned above is pretty close to my trades.

anyway i look over my trades since early march.
total long wins: 67%
total short wins: 42%
total long wins on positive days: 73%
total short wins on negative days: 56%

Give or take a % or 2...

I guess I should stop shorting?!

 

[KCalhoun]

If I'm daytrading and having a bad run I decrease size over a sequence of trades. If I'm winning I scale up.

For swings I always start small, and if the trend is strong my goal is to add, every few days. I think it's smart to only add to winners.

 

[NinaB]

If I'm daytrading and having a bad run I decrease size over a sequence of trades. If I'm winning I scale up.

For swings I always start small, and if the trend is strong my goal is to add, every few days. I think it's smart to only add to winners.

i dont think i'm at the point where i want to add to risk after i win once or a few to potentially wipe out my gains but it is something to think about for the future.

 

[danielc1]

Omg, I'm for sure now that we are at top of a market when I see this on the forum appear.

Cost averaging down, increasing your bet size when you have a loss are all called martingale strategy's in the trader and investor world. BlacknBlue is taking a black and white approach to a context matter in trading. It has nothing to do with only 'doubling' after losing. Any size increasement will do to refer as a martingale strategy. Ask around.

First three losses = 3 units. Loss 4 = 1.65u. Loss 5 = 2.7u. Loss 6 = 4.5u. Total loss = 11.85u. Compared to martingale 6 loss in a row = 1,2,4,8,16,32 totaling 63u which is over 5.3x greater than what is assumed. Regardless, losing 12u from 6 trades isn't bad as the same result can occur from 12 losing trades at 1 unit - far from catastrophic.
.

This is still refered to as a Martingale strategy because the same results are trying to be achieved as a martingale strategy, because you want to recup your losses quickly by adding to a losing positions or by increasing your size after a loss. THAT DOES NOT WORK. Long term capital, numorous other hedge funds that are out of business, traders that believe that this approach work, do not last in the market. Period.

Increasing you bet's when you have more capital and/or adding to winning trades are called anti martingale strategy's:

Anti Martingale strategy:

If I'm daytrading and having a bad run I decrease size over a sequence of trades. If I'm winning I scale up.

For swings I always start small, and if the trend is strong my goal is to add, every few days. I think it's smart to only add to winners.

The choice between the two is yours to make. But any trader that has seen any market, will tell you that martingale will not work.

i dont think i'm at the point where i want to add to risk after i win once or a few to potentially wipe out my gains but it is something to think about for the future.

This comment should make you worried like hell about trading because you are fine and even prefer the opposite, NinaB.

Do your home work.

 

[KCalhoun]

Omg, I'm for sure now that we are at top of a market when I see this on the forum appear.

Cost averaging down, increasing your bet size when you have a loss are all called martingale strategy's in the trader and investor world. BlacknBlue is taking a black and white approach to a context matter in trading. It has nothing to do with only 'doubling' after losing. Any size increasement will do to refer as a martingale strategy. Ask around.



This is still refered to as a Martingale strategy because the same results are trying to be achieved as a martingale strategy, because you want to recup your losses quickly by adding to a losing positions or by increasing your size after a loss. THAT DOES NOT WORK. Long term capital, numorous other hedge funds that are out of business, traders that believe that this approach work, do not last in the market. Period.

Increasing you bet's when you have more capital and/or adding to winning trades are called anti martingale strategy's:

Anti Martingale strategy:



The choice between the two is yours to make. But any trader that has seen any market, will tell you that martingale will not work.



This comment should make you worried like hell about trading because you are fine and even prefer the opposite, NinaB.

Do your home work.

True. Add to winners only, or as I say "feed your winners, starve your losers"

Trades are independent events.

To not give back unrealized profit from first entry use slightly better than breakeven trailing stop.

For example I buy at $22/share and double at $30 I trail stop on all at 27