Exits: Our worst enemy | Page 4 | Forex Factory

Options

Go to
Permalinks
Print Thread

Similar Threads

Why your broker is your friend and your enemy 8 replies

bad broker..The worst of the worst ..reprise 56 replies

candle stick, your enemy or your friend? 14 replies

Volatility = enemy or a friend? 4 replies

Stops Exits And Trade Management 3 replies

Attachments: Exits: Our worst enemy
Exit Attachments

Exits: Our worst enemy

  • #61
  • Jul 14, 2007 7:15pm Jul 14, 2007 7:15pm
Quoting superdezign
Here you are Merlin:

Tom Basso designed a simple, random-entry trading system … We determined the volatility of the market by a 10-day exponential moving average of the average true range. Our initial stop was three times that volatility reading. Once entry occurred by a coin flip, the same three-times-volatility stop was trailed from the close. However, the stop could only move in our favor. Thus, the stop moved closer whenever the markets moved in our favor or whenever volatility shrank. We also used a 1% risk model for our position-sizing system. …
We ran it on 10 markets. And it was always, in each market, either long or short depending upon a coin flip. … It made money 100% of the time when a simple 1% risk money management system was added. … The system had a (trade success) reliability of 38%, which is about average for a trend-following system.

Source: Van K. Tharp, Trade Your Way to Financial Freedom
38% for a trend following system? With a little bit of discretion I believe the figure could be as high as 70%.
  • #62
  • Edited 9:56pm Jul 14, 2007 9:42pm | Edited 9:56pm
Quoting luqmanz
38% for a trend following system? With a little bit of discretion I believe the figure could be as high as 70%.
Good point..... it’s true that with a trend-following system, we can attain overall profit with a win rate < 50%, if our avg net win size exceeds our avg net loss size, by a sufficient margin. The Turtles operated that way. Profit factor is the product of both win rate and win size.

However, it should be mathematically self-evident that “cut losses short, let profits run” only works to the extent that markets trend. Think about it logically: if (hypothetically) prices trended indefinitely, then it wouldn’t matter when you entered, as long as you hold on for as long as possible, you will profit. Likewise, if price moves against you, then in a trending market, the probability that the adverse move will get worse exceeds the probability of recovery. So it makes sense to cut losses quickly. But if/when the probability of reversal (e.g. at support/resistance) exceeds the probability of trend continuation – if/where this can be approximately calculated – then the “cut losses short, let profits run” maxim breaks down. This will happen repeatedly during sideways price movement.

So the question becomes: does your chosen currency pair “trend” sufficiently, in your chosen timeframe, consistently enough, to allow you to overcome costs? If so, you will profit overall; if not, your account balance will fall.

Tom Basso is interviewed in Jack Schwager’s New Market Wizards, which was written during the early 1990s. Was Tom trading forex, or strongly trending stocks? According to Mr Schwager (see p 280), Tom was trading stocks, then moved into futures. I also recognize the quote from Mr Demark: it originally comes from Chuck Le Beau (see http://www.forexfactory.com/showthread.php?p=89109)
Then see my post here for some background:
http://www.forexfactory.com/showthread.php?t=38050
The equities market during the 1990s was trending strongly..... upward.

Under the right conditions, it might well be profitable to trade a random entry system, but why limit the overall probabilistic edge to exits? I still agree with Merlin** (although IMHO HalifaxCB makes a valid point regarding risk) that because entries and exits have to be made against the same market conditions, they are therefore equal in their value. If high probability entries obtain their value from robust technical/fundamental criteria, then for exits to add to the probability of profit, in the long term, they must necessarily do so likewise.

David

**may God help me when I have to hop to the mailbox on one leg... but then the mail is usually only bills, anyway
  • #63
  • Jul 15, 2007 12:29am Jul 15, 2007 12:29am
Quoting hanover
Good point..... it’s true that with a trend-following system, we can attain overall profit with a win rate < 50%, if our avg net win size exceeds our avg net loss size, by a sufficient margin. The Turtles operated that way. Profit factor is the product of both win rate and win size. ...If high probability entries obtain their value from robust technical/fundamental criteria, then for exits to add to the probability of profit, in the long term, they must necessarily do so likewise.

David
Great, in other words, entry and exits are both too damn important.

I guess that's why scalping is very hard for newbies. Trends in 1 min TF is much shorter than trends in Daily charts.
  • #64
  • Edited 4:52am Jul 15, 2007 4:37am | Edited 4:52am
Quoting luqmanz
I guess that's why scalping is very hard for newbies. Trends in 1 min TF is much shorter than trends in Daily charts.
Four (more) reasons why scalping is very hard:

1. The shorter the timeframe, the more random the price movement. Dr Forex really labors this point in his book Bird Watching in Lion Country.

2. Costs are relatively higher. If (for example) I trade 200 pip moves, with a 4 pip spread, I’m paying the broker 2%. If I’m trading 20 pip moves, with a 4 pip spread, I’m paying the broker 20%. Thus costs are 10x higher, or a difference of 18% per trade. Whichever way you look at it, I would say that’s significant..... and invariably underestimated.

3. More scope for broker manipulation: it’s easier for brokers to explain away their moving prices an extra pip or two to take out a stop, than it is for them to justify a contrived 20-30 pip ‘spike’.

4. Scalping requires rapid-fire decision making, hence it’s easier for the trader to become flustered, and make ‘emotional’ decisions, misread or overlook elements on charts, fail to take in the bigger picture, and/or make data entry errors when placing orders.

David
  • #65
  • Jul 15, 2007 6:16am Jul 15, 2007 6:16am
Quoting hanover
Four (more) reasons why scalping is very hard:

1. The shorter the timeframe, the more random the price movement. Dr Forex really labors this point in his book Bird Watching in Lion Country.

2. Costs are relatively higher. If (for example) I trade 200 pip moves, with a 4 pip spread, I’m paying the broker 2%. If I’m trading 20 pip moves, with a 4 pip spread, I’m paying the broker 20%. Thus costs are 10x higher, or a difference of 18% per trade. Whichever way you look at it, I would say that’s significant..... and invariably underestimated.

3. More scope for broker manipulation: it’s easier for brokers to explain away their moving prices an extra pip or two to take out a stop, than it is for them to justify a contrived 20-30 pip ‘spike’.

4. Scalping requires rapid-fire decision making, hence it’s easier for the trader to become flustered, and make ‘emotional’ decisions, misread or overlook elements on charts, fail to take in the bigger picture, and/or make data entry errors when placing orders.

David

I couldn't agree more with these points. Especially with the first 3.
  • #66
  • Jul 15, 2007 8:00am Jul 15, 2007 8:00am
Ok for those who say the exit has more importance over the entry I will try to use a sport's analogy. Since most here on FF are not in the US I will use Football (what we call Soccer) to try to make a point.

Which is more important most of the times to scoring a goal, the player making the perfect pass to another player in a position to score or the player who actually kicks it (or heads it) into the net. I say most of the times because I guess sometimes it can be considered unassisted when controlled for a period of time by the player scored with little help. Not often though.

I would say they are equally important although the scorer gets all the glory.

The object in trading is to make money and in sports is to score but to maximize either, both entry (assist) and exit (score) need to contribute as well.
  • #67
  • Jul 15, 2007 8:17am Jul 15, 2007 8:17am
Good point.... it’s true that with a trend-following system, we can attain overall profit with a win rate < 50%, if our avg net win size exceeds our avg net loss size, by a sufficient margin. The Turtles operated that way. Profit factor is the product of both win rate and win size.

However, it should be mathematically self-evident that “cut losses short, let profits run” only works to the extent that markets trend.

Cutting your losses short and letting profits run does not help you more or less in any situation.

if Losing trades > Winning trades on a consistent basis, you are losing money. Its common sense.

Think about it logically: if (hypothetically) prices trended indefinitely, then it wouldn’t matter when you entered, as long as you hold on for as long as possible, you will profit.

True, but impossible

Likewise, if price moves against you, then in a trending market, the probability that the adverse move will get worse exceeds the probability of recovery. So it makes sense to cut losses quickly.

That would make the "adverse move", a new trend

But if/when the probability of reversal (e.g. at support/resistance) exceeds the probability of trend continuation – if/where this can be approximately calculated – then the “cut losses short, let profits run” maxim breaks down. This will happen repeatedly during sideways price movement.

You just desribed a trend reversal twice "exceeds the probability of trend continuation" = reversal = possible new trend

So the question becomes: does your chosen currency pair “trend” sufficiently, in your chosen timeframe, consistently enough, to allow you to overcome costs? If so, you will profit overall; if not, your account balance will fall.

Thats why you define the trend per whatever you are trading as be x% of a move depending on how much that pair moves. Everything trends. If it didnt there would be no need for organized markets

Tom Basso is interviewed in Jack Schwager’s New Market Wizards, which was written during the early 1990s. Was Tom trading forex, or strongly trending stocks? According to Mr Schwager (see p 280), Tom was trading stocks, then moved into futures. I also recognize the quote from Mr Demark: it originally comes from Chuck Le Beau (see http://www.forexfactory.com/showthread.php?p=89109)
Then see my post here for some background:
http://www.forexfactory.com/showthread.php?t=38050
The equities market during the 1990s was trending strongly..... upward.

Under the right conditions, it might well be profitable to trade a random entry system, but why limit the overall probabilistic edge to exits? I still agree with Merlin** (although IMHO HalifaxCB makes a valid point regarding risk) that because entries and exits have to be made against the same market conditions, they are therefore equal in their value. If high probability entries obtain their value from robust technical/fundamental criteria, then for exits to add to the probability of profit, in the long term, they must necessarily do so likewise.

I would never, and have never suggested trading a random entry system. I think it is rediculous TO EVEN CONSIDER IT! BUTTTT what I was trying to prove when I brought that up is to help everyone see that entries are not as important as people make them out to be..

THEY ARE IMPORTANT!!!!! That is not what I am arguing, all I am saying is that people who stop putting so much emphasis on ACCURACY AND START THINKING ABOUT THE EXPECTENCY of a system they are building.. Look at the bigger picture and incorporate what is necesary to make yourself a successful trader...
  • #68
  • Jul 15, 2007 8:19am Jul 15, 2007 8:19am
I highly doubt that. If you come across one with even a 60% win rate over a decent sized sample of trades then I think you should be given the trader of the year award

Quoting luqmanz
38% for a trend following system? With a little bit of discretion I believe the figure could be as high as 70%.
  • #69
  • Jul 15, 2007 10:21am Jul 15, 2007 10:21am
Quoting hanover
Halifax, IMHO the "finite capital" problem is more of a position sizing, than an exit, issue. Assuming the original placement of protective stoplosses was based on sound technical criteria, then I agree that they should be adhered to, to limit risk. In the case of a making a profitable exit, risk is generally no longer an issue.

I'm not saying that we can calculate the probabilities with any high degree of accuracy, just that we can enhance expectancy by exiting (whether fully or partially) when the probability of a favorable continuation appears low (e.g. our upward breakout trade loses momentum upon hitting a confluence of pivot/Fibo/resistance). In other words, technical criteria, rather than simply taking profit at 50 pips because 50 was the nearest round number that fitted our back-testing results.

David
Hi David;
In that I was just addressing the issue that one should only get out when going the other way looked more profitable, which was what the poster befor me (IIRC - it's so many posts ago ) was saying.
I tend to look at the markets as a mixture of short deterministic periods interspersed with many drifting, rather random periods. These are clearest on the very short term; if you watch USD/CHF or EUR/USD over the last little while on something like the hourlies, you'll see what I mean in the stair-step type of behaviour - with short sharp movements that start with an initial acceleration, a period of high velocity in one direction, an overshoot, and a settlement back to a new level where the value drifts sideways for awhile.

My point is that once the deterministic phase is over, one is into a region where prediction with any degree of consistency seems meaningless, at least on that time frame and given whatever set of analytical tools one wants.

You can, of course, move out to a longer time frame to determine whether to stay in, or go the other way; but on the longer timeframes the determinsitic part of the trend becomes less pronounced relative to the apparent noise, which in turn increases the risk (and hence requires deeper pockets).

For example, on the euro/usd, if I was in long for the long run, I would probably be long now, which I am not. But to maintain that long, I would have to be willing to accept a risk of maybe a 400p down move before bailing, since it does look somewhat like we may be facing a minor rising resistance (based on the peak up closes from 2/April on, on the MQ4 - i.e. Moscow time) charts on the weeklies. The last time we hit this led to a 7 week decline of around 400pips. For me that isn't an indication of an imminent down move, but it does indicate enough of a risk of one to close out until I see a clearer indication of which direction the next largish move will be. (FWIW, I don't trade off these, I trade off my Kalman filters, but since no one else does, I have to translate my thoughts to more widely accepted indicators )


Cheers;
CB
  • #70
  • Jul 15, 2007 11:04am Jul 15, 2007 11:04am
~ 1200 bars of 5 min data I had lying around.
Attached Image (click to enlarge)
Click to Enlarge Name: Random Entry Exit XL.PNG Size: 109 KB
Attached File(s)
File Type: zip Random Entry Exit XL.zip   78 KB | 375 downloads
  • #71
  • Jul 15, 2007 8:01pm Jul 15, 2007 8:01pm
Quoting superdezign
I highly doubt that. If you come across one with even a 60% win rate over a decent sized sample of trades then I think you should be given the trader of the year award
I have seen people scoring that 60% using a system with their discretion. Perhaps, they are just lucky?

I have several solution to the issue you brought up. These solutions is for long-term traders using Daily charts to trade.

Quote
However, it should be mathematically self-evident that “cut losses short, let profits run” only works to the extent that markets trend.
Just pick a market that the Daily charts show a strong trend e.g EURUSD.

Quote
Think about it logically: if (hypothetically) prices trended indefinitely, then it wouldn’t matter when you entered, as long as you hold on for as long as possible, you will profit.
All trend will finish sooner or later. Solution, trade long-term because long-term trend continues for years like EURUSD.

Quote
But if/when the probability of reversal (e.g. at support/resistance) exceeds the probability of trend continuation – if/where this can be approximately calculated – then the “cut losses short, let profits run” maxim breaks down. This will happen repeatedly during sideways price movement.
This is the time where entry and trail stop is crucial. With good execution, you will be able to cut losses short, but since the market is sideways, you cant let profits run simply because the market doesnt let you to. Just be happy and stick to your plan (buy, trail stop,buy trail stop) until one day the trend resumes. You dont have to make money everyday. Patience.

Quote
So the question becomes: does your chosen currency pair “trend” sufficiently, in your chosen timeframe, consistently enough, to allow you to overcome costs? If so, you will profit overall; if not, your account balance will fall.
Hence the beauty of trading long-term (daily charts). You can hibernate for 3 months and the trend is still there. LOL.
  • #72
  • Jul 16, 2007 4:32am Jul 16, 2007 4:32am
I hope that all wasnt directed at what I said, for that statement was a quote someone earlier had said and my responses to it, or more so arguments to it were in bold. just so you konw

Quoting luqmanz
I have seen people scoring that 60% using a system with their discretion. Perhaps, they are just lucky?

I have several solution to the issue you brought up. These solutions is for long-term traders using Daily charts to trade.


Just pick a market that the Daily charts show a strong trend e.g EURUSD.


All trend will finish sooner or later. Solution, trade long-term because long-term trend continues for years like EURUSD.


This is the time where entry and trail stop is crucial. With good execution, you will be able to cut losses short, but since the market is sideways, you cant let profits run simply because the market doesnt let you to. Just be happy and stick to your plan (buy, trail stop,buy trail stop) until one day the trend resumes. You dont have to make money everyday. Patience.


Hence the beauty of trading long-term (daily charts). You can hibernate for 3 months and the trend is still there. LOL.
  • #73
  • Jul 16, 2007 9:21am Jul 16, 2007 9:21am
I'm still here Guys.

I opened a new thread for developing an exit strategy for price action bar trading

http://www.forexfactory.com/showthre...99#post1496199

Right now i only use Psychological levels (supp/res/Fibo proj) for exit.

I'm looking to enhance these and also filters for entry.
100% of traders are losers. Just that some win more than they lose!
  • #74
  • Jul 16, 2007 2:46pm Jul 16, 2007 2:46pm
Quoting superdezign
I would never, and have never suggested trading a random entry system.
Superdezign, Not sure why you thought I was suggesting that you were advocating random entries. I started out replying to luqmanz, then I’m afraid I digressed somewhat, responding to no post in particular. I agree with virtually everything that you say, although much depends on what each of us means by “trend” or “trending”.

I agree that everything trends, from one tick to the next. But with a trailing stop, we always give back some profit. If the market doesn’t happen to trend far enough, so that the profit given back exceeds any gain made thus far on the trade, the trade becomes a loss. That’s what I mean when I say that “letting profits run” will only operate profitably to the extent that prices trend. If the market moves sideways, or prices move (for example) into an area of strong, confluent S/R, it may prove more profitable to take profit quickly, instead of letting it run; a reversal may be imminent.

There are a lot of very different, but equally valid, ways of interpreting price behavior, and formulating trading strategies. And anyway, to whatever extent I’m wrong, well, I’m wrong; it really doesn’t matter! I’d rather invest my energies into making my trading more profitable, than trying to score points debating semantics on a forum. But I would like to apologize if I sounded high-handed; I don’t want to upset anyone.

Clockwork71, thanks for the feedback.

HalifaxCB, thank you for your reply. I believe I understand where you are coming from. As you say, on certain timeframes, price tends to burst strongly, and then drift into a sideways, indeterminate or random pattern. If I understand correctly, this would equate to Merlin’s “flat” state (see post #53). At such a point it might be wise to take some profit. I know a successful trader who opens his position when price enters the short deterministic periods, then exits it, in full or in part, when momentum decreases.

I’ve experimented with Kaufman bands before, but never heard of Kalman filters (similar names that confused me, but no doubt completely different concepts). I'll have to do a Google on that one!

David
  • #75
  • Jul 16, 2007 2:51pm Jul 16, 2007 2:51pm
Quoting superdezign
It wouldnt produce the same results, only because it would not allow you to follow one of the most important rules of trading: let your profits run and cut your losses short
you love trading cliches :

actually this one has some truth to it, but i think it is superceded by another cliche - "follow your plan!"
Relax and be happy.
  • #76
  • Jul 16, 2007 2:56pm Jul 16, 2007 2:56pm
Quoting hanover
(although IMHO HalifaxCB makes a valid point regarding risk)
agreed. risk considerations can definetely make entry and exit carry different characteristics. but as far as expectancy goes, i still see no good argument for why exit/entry should be viewed any differently.
Relax and be happy.
  • #77
  • Jul 16, 2007 3:29pm Jul 16, 2007 3:29pm
This thread my friends, is what makes this place so good. We are up to 76 posts without anyone getting personal.

Although the discussion isn't a "system", it is something more. It's an idea.

I have recently learned that in order to make some pips, following someone else's ideas isn't the way to go. However, it is discussions like this that get you thinking. And thinking helps, it helps you to look at things differently. To look at things differently is a bonus, because in the end your system becomes your own. And let's be honest, my system can only be unique because of the combination of various parts. These parts are available, and borrowed 90% of the time from other traders. It is unique in the combination, not the parts.

Let's say that I agree 100% with everything Merlin has posted about exits. That in fact, exits are essentially taking the other side. (One of the three available.) So if I feel that the market is running out of steam, I decide to sit on the sidelines. That I am taking the position of not having a position. In a sense, that IS a trade I believe. Maybe I feel that I am done for whatever reason, and the trend won't continue, at least for the time being. Maybe as I exit out of my EUR/USD (as I see sideways), I decide my money is better served to go into USD/JPY....that is one way to look at it as well. It's all about putting your money to work if work is available.

Anyways, as I peel away my crutches, or as some call them indicators, it is discussions like this that I believe will help me get to the next level. Thanks guys!
  • #78
  • Jul 16, 2007 3:35pm Jul 16, 2007 3:35pm
Quote:
Originally Posted by superdezign http://forexfactory.com/images/buttons/viewpost.gif
...but did it backwards, where you exited randomly and entered using a well devised strategy, I bet you would lose money in the end.

its a bet

now we just have to find the experiment!
__________________


Actually, if you think about it, you already know the answer. Beginning traders do this all the time. They get a good entry, and grab 12 pips. Why? Because its a random profit, and they are used to losing, so winning is something to hang on to. Or, there are a lot of newer traders that will aim for 10 pips. or 20. We all know what that ends up with. I know that in a sense aiming for 10 isn't purely "random", but the price would be. It wouldn't be based on TA, it is based on a goal. Goals are good, but if there is no resistance for another 40 pips, then its an arbitrary figure.....kinda like being random I suppose.




Quoting merlin
its a bet

now we just have to find the experiment!
  • #79
  • Jul 16, 2007 4:11pm Jul 16, 2007 4:11pm
Quoting superdezign
I hope that all wasnt directed at what I said, for that statement was a quote someone earlier had said and my responses to it, or more so arguments to it were in bold. just so you konw
Hehe .. I'll be happy if what I said can be useful to anyone. It wasn't really directed to anyone. I believe other people may have the same questions too.
  • #80
  • Jul 16, 2007 7:19pm Jul 16, 2007 7:19pm
Quoting clockwork71
This thread my friends, is what makes this place so good. We are up to 76 posts without anyone getting personal.

Although the discussion isn't a "system", it is something more. It's an idea.

I have recently learned that in order to make some pips, following someone else's ideas isn't the way to go. However, it is discussions like this that get you thinking.


agreed, and i much prefer debating ideas than systems

Quoting clockwork71
I am taking the position of not having a position. In a sense, that IS a trade I believe.
haha!! yes this is how i see it too! a "trade" to me isnt one-entry-one-exit. a "trade" is anything i do to adjust my position. once your positions get complicated (ie multiple systems trading multiple instruments in multiple markets with complex MM strategies and scaling in and out) it actually becomes quite hard to view a trade as a single entry+exit.
Relax and be happy.
1 trader viewing now