Menu
Sister Sites

Options

Bookmark Thread

First Page First Unread Last Page Last Post

Print Thread

Similar Threads

Winners scale winners, losers scale losers 12 replies

What methods do you use to let your winners run? 38 replies

If losers average losers, then I am a loser 10 replies

My trading journal-winners and losers shown 137 replies

Winners And Losers Trading System 138 replies

Attachments: Do you manage losers and let your winners run?
Exit Attachments

Do you manage losers and let your winners run?

  • Opened Dec 12, 2017 | Never Closes | 51 Votes
  • Poll Results
Do you manage the losers and let the winners run?  
Yes 31 Votes
61%
No 20 Votes
39%
  • Post #1
  • Quote
  • First Post: Dec 12, 2017 10:26pm Dec 12, 2017 10:26pm
Think about this.

We take 100 trades. Every trade is a long with a 100 pip stop and 100 pip target.

50 trades hit our target. 50 trades hit our stops.

How many of the price paths that reach our stops were once profitable during it's way to the stop?

How many of the price paths that reach our targets were once losing on its way to the stop?

I think it'd be safe to assume 50%.

So 25 of our winner trades were once losers and 25 of our losing trades were at some point winners.

Do you manage the losers and let the winners run?

Yes or No... and Why?
Quoting Sis.yphus
Think about this. We take 100 trades. Every trade is a long with a 100 pip stop and 100 pip target. 50 trades hit our target. 50 trades hit our stops. How many of the price paths that reach our stops were once profitable during it's way to the stop? How many of the price paths that reach our targets were once losing on its way to the stop? I think it'd be safe to assume 50%. So 25 of our winner trades were once losers and 25 of our losing trades were at some point winners. Do you manage the losers and let the winners run? Yes or No... and Why?
When I first started. I thought the ratio has to be equal. 10 years later. I found a way to lose 5 to 10 pips to make 50 or 100 pips. Just yesterday I made 50 pips but I didn't close because I wanted to let my winner run.

So I always look for 2R to 5R. My time is freed to do what really matters.
You're going to pay a price for every bloody thing you do and don't do
Quoting newbietofx
{quote} When I first started. I thought the ratio has to be equal. 10 years later. I found a way to lose 5 to 10 pips to make 50 or 100 pips. Just yesterday I made 50 pips but I didn't close because I wanted to let my winner run. So I always look for 2R to 5R. My time is freed to do what really matters.
This isn't about the R:R. The 1:1 RR was used to make the example simple.

The question is do you manage the losers and let the winners run. This applies to any R:R.
  • Post #4
  • Quote
  • Edited at 5:24am Dec 13, 2017 1:00am | Edited at 5:24am
Quoting Sis.yphus
Think about this. We take 100 trades. Every trade is a long with a 100 pip stop and 100 pip target. 50 trades hit our target. 50 trades hit our stops. How many of the price paths that reach our stops were once profitable during it's way to the stop? How many of the price paths that reach our targets were once losing on its way to the stop? I think it'd be safe to assume 50%. So 25 of our winner trades were once losers and 25 of our losing trades were at some point winners. Do you manage the losers and let the winners run? Yes or No... and Why?
The most important aspect of the practical trading process is to manage the losers. It starts with the sl placement and the entry that's determined to survive the sl set. Loss management post-entry involve when and where to call off the trade when initial conditions that presented the positive expectancy have changed, when to tighten sl and when to activate sltbe. I have this loss management automated so that I mostly focus on market conditions.
To whatever extent the probability of (price) continuation exceeds the probability of reversal, cutting losses quickly and letting winners run will deliver positive expectancy (ignoring costs). That is the rationale behind trend following systems. And the converse also applies.

The nearer price gets to becoming a completely random walk, the nearer ALL strategies approach zero expectancy (minus costs).

Detailed info here.
Quoting Sis.yphus
Think about this. We take 100 trades. Every trade is a long with a 100 pip stop and 100 pip target. 50 trades hit our target. 50 trades hit our stops. How many of the price paths that reach our stops were once profitable during it's way to the stop? How many of the price paths that reach our targets were once losing on its way to the stop? I think it'd be safe to assume 50%. So 25 of our winner trades were once losers and 25 of our losing trades were at some point winners. Do you manage the losers and let the winners run? Yes or No... and Why?
Why people in these forums are so confused??

1. It's mathematically impossible to get 50% winning trades with equidistant fixed stop and target! The win % will be ALWAYS below 50%. If you place the target closer, the win% will increase. If you place the stop closer, the win% will decrease.

2. There is no way to know which trade is going to hit the target and which the stop. So, by "managing" your losses, you automatically reduce the win% even more.

3. Placing random trades is dumb way to trade!
---------------------------------------------------------------------------------------

There is only a few ways to win with technical systems.

1. You have to rely on repeatable patterns which produce consistent result. And in general there are only 2 patterns. Trend/momentum (including breakouts) and mean reversion patterns. Everything else is just a variation and falls in one of these 2 groups.

2. You have to use variable position sizing. If the win rate of the strategy( the pattern) is high enough, you can even apply aggressive martingale. If the win rate is low, you have to apply anti-martingale (directional pyramiding).

Cutting your losses quickly is applicable only for trend-momentum systems! If you try to cut the losses quickly with mean reversion systems you will destroy the strategy.
Quoting hanover
To whatever extent the probability of (price) continuation exceeds the probability of reversal, cutting losses quickly and letting winners run will deliver positive expectancy (ignoring costs). That is the rationale behind trend following systems. And the converse also applies. The nearer price gets to becoming a completely random walk, the nearer ALL strategies approach zero expectancy (minus costs). Detailed info here.
Hey Hanover.

I was waiting for you to drop by as I know many posts like this drive you crazy suggesting some edge can be gained through management. That isn't exactly what I am trying to do, but I guess it's kind of similar.
Quoting alphaomega
{quote} Why people in these forums are so confused?? 1. It's mathematically impossible to get 50% winning trades with equidistant fixed stop and target! The win % will be ALWAYS below 50%. If you place the target closer, the win% will increase. If you place the stop closer, the win% will decrease. 2. There is no way to know which trade is going to hit the target and which the stop. So, by "managing" your losses, you automatically reduce the win% even more. 3. Placing random trades is dumb way to trade! ---------------------------------------------------------------------------------------...
wise talks here!

Would you please explain or direct me to study and learn more about anti-martingale (directional pyramiding) ? and
what is the mean reversion system ?
  • Post #9
  • Quote
  • Edited at 7:41am Dec 13, 2017 7:28am | Edited at 7:41am
Excellent conversation. This has been my focus for a while now. Below is the outline of how I choose to approach this theory.

Everything below are simply my opinions. In no way do I claim to KNOW anything.
• Price is basically unpredictable.
• Mathematics may be a way of tinkering with results.
• I agree that using a basic R:R is not that helpful, according to methods I've explored.
• The best results I have had, involve using a fixed stop, and beginning with no take profit level.
• In my opinion, H4 gives me a very good idea of the current trend. By current, I mean what is happening today, which can extend into several days SOMETIMES.

1. First I pick something that looks good, like a moving average or two, on an H4 chart. I believe this will work with any method of choosing a level.
2. Next I look at pairs with certain characteristics. For instance, I prefer generally "Tame" pairs, like EU, UCad. These are more simple for me to choose a very conservative stop loss. On H4 I use a 20-22 pip SL, including the spread.
3. Entries are always a number of pips Beyond my chosen level (MA, Fib, etc.) thereby keeping me out of consolidation traps, to some degree. These traps can still create havoc no matter what I've tried. I had one run of 12 trades with only one win.
4. Then I test the crap out of how this looks over at least a one year period of time. HOWEVER! every year is not like the last, so never bet the farm.

That's the basics of my thoughts. In testing I generally get about a 32 to 44% win rate. Wins average about 70-75 pips. The methods I'm working with so far, average about two trades weekly, per pair. So losing months are rare, but can happen.
For clarification purposes, when I say "manage" a trade, I mean take it off manually at some level before your Stop or Target would take it out.

Quoting alphaomega
{quote} 2. There is no way to know which trade is going to hit the target and which the stop. So, by "managing" your losses, you automatically reduce the win% even more.
BINGO BONGO!

Let's take it back.

So 25 of our winner trades were once losers and 25 of our losing trades were at some point winners.

----------------------------------------------------------------------------------------------------------------
If we managed our losers, 25 of the losing trades that became winners would have closed as losers.

If we let our winners run, 25 of the winning trades would have become losers.

Now let's imagine we managed our winners and let the losers run...

We could stop the 25 trades that were at one point winning from becoming losers.

We would let the 25 trades that were losers run till they became winners.

One could almost argue that it is next to impossible to stay profitable if we don't manage winners...

Nothing mind-blowing... Just a different take on the conventional wisdom that comes so naturally to us.

Inserted Video


"When you're up it's never as good as it seems (you always want to let it run for more profits) and when your down you never think your gonna be up again (cut losers that will become winners because you are scared of losing).
I would think that everyone who makes money has to manage their losers. If they don't the losers will eat up the winners, or, the losers will cut deep into your profit.
If you are good at something, never do it for free--Joker
the most common mistake is to let the loosers run, hoping the market will reverse and come back your direction.
Quoting jmn5611
I would think that everyone who makes money has to manage their losers. If they don't the losers will eat up the winners, or, the losers will cut deep into your profit.
I should have clarified. In this scenario, there is a hard stop in place. When I say manage losers, I mean cutting the trade before it hits your stop. So the risk is always defined.
From my experience and observation, sometimes when the market is turning against you, before hitting your SL, you can cut or BE during the last retracement
  • Post #15
  • Quote
  • Edited Dec 14, 2017 3:19am Dec 13, 2017 3:12pm | Edited Dec 14, 2017 3:19am
Quoting Sis.yphus
I know many posts like this drive you crazy suggesting some edge can be gained through management. That isn't exactly what I am trying to do, but I guess it's kind of similar.
It depends on how you define MM. It's possible mathematically to gain an edge from exits/trade management, but not from position sizing, unless you can get larger positions to coincide with the winning trades. Otherwise it becomes a game of hit or miss, see the example below.
______________________________

Quoting alphaomega
Why people in these forums are so confused??
Perhaps it's because (to paraphrase Jack Schwager) there are a thousand different ways to profit, and a million different ways to lose?

FWIW:

1. It's mathematically impossible to get 50% winning trades with equidistant fixed stop and target! The win % will be ALWAYS below 50%.
I know at least one institutional trader who averages something exceeding 1:1 R AND also has WR > 50%.

If you place the target closer, the win% will increase. If you place the stop closer, the win% will decrease.
Agree. Everything else being equal, WR and RR tend to operate in inverse proportion. Mathematically, the nearer price comes to a completely random walk, the more applicable this relationship becomes.

2. There is no way to know which trade is going to hit the target and which the stop.
Agree.

So, by "managing" your losses, you automatically reduce the win% even more.
Not necessarily, this is a non sequitur from your last point.

3. Placing random trades is dumb way to trade!
Agree, there are much better ways to enter.

There is only a few ways to win with technical systems. 1. You have to rely on repeatable patterns which produce consistent result.
Agree, assuming the system is mechanical/quantitative.

And in general there are only 2 patterns. Trend/momentum (including breakouts) and mean reversion patterns. Everything else is just a variation and falls in one of these 2 groups.
Agree.

2. You have to use variable position sizing.
Disagree. Assuming no serial correlation between trades, variable position sizing is beneficial only if you can somehow get larger positions to coincide with the better outcomes.
(Simple example: Suppose I have a 50 pip win and a 40 pip loss. If the first trade was 1 lot and second 2 lots, I'm ahead on pips but down in overall $$ (1x50-2x40 = -30). Conversely, if the first trade was 2 lots and the second 1 lot, my overall $$ improves (2x50-1x40 = +60). So unless I know beforehand which trade setup/outcome has higher probability/expectancy, variable pos sizing is merely a lucky game of hit or miss).

If the win rate of the strategy( the pattern) is high enough, you can even apply aggressive martingale.
Potentially yes, but why increase risk of ruin exponentially, and without improving overall expectancy, if you already have an edge?

If the win rate is low, you have to apply anti-martingale (directional pyramiding).
Not necessarily, it depends on the nature of your edge and market conditions. In general, pyramiding will work only in 'trending' situations, you will lose money pyramiding in sideways markets, as you add positions and then price reverses back toward the mean.

Cutting your losses quickly is applicable only for trend-momentum systems!
Agree.

If you try to cut the losses quickly with mean reversion systems you will destroy the strategy.
Agree, it would be counter productive.

Don't want to get into a protracted debate, so I'll simply state my views and move on. Sounds like you might be a quant trader, whereas there are institutional traders whose analysis includes other techniques, which could account for some of the differences between our conclusions.
Quoting UlliC
the most common mistake is to let the loosers run, hoping the market will reverse and come back your direction.
This statement is not quiet true,it is true only for traders who take trades very light heartedly,if we do the top down trend following trading then losers trades comes back to it level or very near in near future.I do trade without stop loss and I close losing trades with minimum losses,so this is my way of managing the loss.

one must manage it by break even on retracement or calculated safe hedging.
Quoting Sis.yphus
{quote} I should have clarified. In this scenario, there is a hard stop in place. When I say manage losers, I mean cutting the trade before it hits your stop. So the risk is always defined.
I see. Personally I will lean on my stop if I use them, which I don't very often.
If you are good at something, never do it for free--Joker
Quoting Sis.yphus
. Do you manage the losers and let the winners run? Yes or No... and Why?
Hi Sis.yphus,
Not sure there are too many other ways to trade profitably!

But that is a bit pat. So much depends on the interplay between your long-run trade success rate and the relative size of average winners to average losers. It means there can be many quite different trade management techniques depending on how the method works.

As a practical example, I trade essentially off a 1:1 risk:reward structure and a long-run success rate nearer 60% than 50%. Is this managing losers and letting winners run? I wouldn't really say so but it's been profitable for me for several successive years now and I'm loathe to give it up.
Quoting alphaomega
{quote} 1. It's mathematically impossible to get 50% winning trades with equidistant fixed stop and target! The win % will be ALWAYS below 50%.
Hi alphaomega,
Could you explain why this is mathematically so?

Unless I am badly misinterpreting my trading records for the last several years I've traded profitably using equidistant stop & target with an achieved success rate in the high 50s%.
Quoting marcmarc
{quote} Hi alphaomega, Could you explain why this is mathematically so? Unless I am badly misinterpreting my trading records for the last several years I've traded profitably using equidistant stop & target with an achieved success rate in the high 50s%.
Perhaps I had to say it differently.

If you place trades in a random way and without any structure/framework the expectancy per trade is always negative. It doesn't matter where you place the stop and where the target. It also doesn't matter if cut your losses or if let the "winner" run. The result is always negative. And the expectancy per trade doesn't change.

However, if you find some specific (proven) pattern where you can extract consistent result, then the story is completely different.
Of course you can have positive result with equidistant stop and target. (But not with randomly placed trades!)
0 traders viewing now