Tails and the Art of placing stops

[tech/a]

Yeh like i said i need some software bad.

No you dont!

You can do it without software.
Mind you you'll learn more than you can imagine designing methods with software.

Dont think for one minute that profitability is exclusive to systems traders!

 

[theasxgorilla]

Nizar,

I can only tell you what I do and perhaps it helps. My trading method is discretionary and longer term...a position could be open for anywhere between 6 and 18 months.

On a given share I find the maximum distance travelled from peak to trough during an up trend. When I set a stoploss, initial OR trailing, I factor this in. If the maximum distance travelled from peak to trough during an up trend is 12% then I set two stoplosses...one is discretionary, the other is definite.

The discretionary I would therefore set at 12% and MAY execute on one daily close below at market open of the following day. The definite stoploss I will set at 15% and (should) execute no matter what. The definite stoploss is used to determine position size based on putting 2% of trading capital at risk (FPS).

In the example that you gave, EVE, the peak to trough distance travelled during an up trend can be as much as 25%. This means that if you entered at 21 cents you would have to set a stoploss at 16 cents to be just outside the noise.

If you trade these very low priced, low liquidity shares then this is par for the course. Your stoplosses need to be wider due to the inherently higher volatility and due to the lower liquidity tails are more prone to forming, particularly during the sale of larger positions.

 

[MichaelD]

As a result of your stops/exits; Were you stopped out of all your longs during May?

Last May I wasn't exclusively using a 6.5ATR exit. I actually tailored the exit to the last 12 months' price action, so I'd use anything from a 4.5ATR to about a 6ATR exit. I was stopped out of about 1/2 my longs.

Did you end up exiting near the lows?

For about 1/2 of the stopped out positions, yes, the exit corresponded with the low. However, looking at the exits as a whole in retrospect shows quite clearly that I was far better off taking the exits than not.

I really need to do some backtesting bad.

Backtesting is a very powerful tool to give you the confidence to stick with your trading rules.

Another question for you; is your entry (closed on all time high) based on a true all time high? ie. how about a high for the last 5rs, coz stocks like that can still run if the broken overhead resistance was significant eg.MLS at 4c.

I don't actually use an all time high entry as it's a fairly high drawdown entry. This is my entry filter;

CLOSE>Mov(CLOSE,175,EXPONENTIAL) AND H>Ref(HHV(H,2),-1) AND CLOSE>OPEN

1. Close above a long term moving average
2. Highest high value for 3 days
3. White candle

That's it. To date, I haven't found anything else that significantly improves the end result. To be honest, I'm not looking (see later in this reply).

And one more: i would imagine during the last 12-18 months you would have received more entry signals than your trading capital would allow, so how do you choose which trades to take?

Do you look at volumes?
Liquidity?
Share price between a specific range?

Yes, this filter produces many trading signals for my universe (ASX300). I simply started at the letter A and bought until the money ran out. As trades got stopped out, the funds went either to pyramid trades or to new trade signals. I'm now left with 1/2 "legacy" trades which weren't stopped out in May and 1/2 early alphabet trades. There is no performance differential between them. More often than not, the worst looking charts on entry are the ones turning into the biggest winners.

DId you do some sort of MonteCarlo analysis so any combination of trades taken (when the system was backtested) would been profitable?

Yes, I extensively MonteCarloed the system on all sorts of universes including some quite hard to construct ones. It was 100% profitable in all runs, a mandatory condition before I would trade any system.

A few further points I'd make;

1. All you need for consistent results is one or more systems that you know are profitable and that you have absolute confidence in so that you trade them consistently. How one gets to this point is individual. For me, this journey required TradeSim and a mechanical system.

2. Once you have absolute confidence in the profitability and consistency of your trading, then it is time to look at how to improve results in other ways (eg pyramiding/scaling in, leverage, other systems). This is where the real differences to your bottom line can be found, not in tweaking entries.

3. Psychology is truly the key-we are our own worst enemy. I was just recently congratulating myself on how long it has been since I made a trading error (= break system rules) and guess what, I just managed to screw up a gift of a trade in this market (WMT) by ignoring my rules in the heat of the moment. D'Oh!

 

[motorway]

Getting stopped out is also a function of where you are entering..

Buying at the supply line ( The upper trend line where supply is entering )
IS going to be a different experience to buying at the demand line ( where buying is entering )

Look at the waxing and waning of the volume as the trend unfolds

notice the Volume drying up at as it approaches the lower line
and then moves in harmony as the trend continues..

The points where the volume drys up are the points the selling waves were exhausted ... If the selling is exhausted you are probably not going to have trouble getting stopped..

Not maybe what you are looking for ( does require judgment )
mechanical Vs Discretionary.... But discretionary can be objective and methodical...


Ok Look at the Volume Now.... That clear earlier pattern has changed
selling has not exhausted so far... This suggests the angle of the trend will change and/or it's direction..

Markets react naturally as a trend unfolds.. Being stopped out too often means you are out of step with that natural rhythm

motorway

 

[It's Snake Pliskin]

Yes.
Because i get stopped out too much, too often.
This is because my exits and stops are too discretionary and i dont have enough of a knowledge or background to apply technical analysis proficiently.

Truly mechanical systems, like your long term trending system, on the other hand, do not rely on any sort of analysis.

I thought or used to think that such a wide stop like your 6.5ATR would result in too much profit giveback as you yourself have identified, but maybe in the long-term its the most profitable way than a tighter ATR based stop.

As a result of your stops/exits; Were you stopped out of all your longs during May?
Did you end up exiting near the lows?

I really need to do some backtesting bad.

Another question for you; is your entry (closed on all time high) based on a true all time high? ie. how about a high for the last 5rs, coz stocks like that can still run if the broken overhead resistance was significant eg.MLS at 4c.

And one more: i would imagine during the last 12-18 months you would have received more entry signals than your trading capital would allow, so how do you choose which trades to take?

Do you look at volumes?
Liquidity?
Share price between a specific range?

DId you do some sort of MonteCarlo analysis so any combination of trades taken (when the system was backtested) would been profitable?

Yeh like i said i need some software bad.

Thanks for your help.

Nizar,
Don't get frustrated with it. losing is part of the business. It is the long term effects that count.

Try to be conservative WITH WHAT you buy. WHY is it worth the RISK? What is in the background that warrants a purchase?

I believe stops are a tricky tool to set. They do what they are called which is why I ask the above.
I am referring to initial stops.
Take time out and refocus on your mission.
Snake

Read motorways post above mine.

 

[theasxgorilla]

Very cool post 'motorway'....I'm digging this volume analysis stuff.

The ASX Gorilla.

 

[nizar]

Thanks for the advice every1.

Tech/a - Just one question: where do you set your stop for your long term trending system? Below the 180dma?

Seems like an excellent place to a put a stop, wouldve kept you in PDN and ZFX and i mean the whole trend going back 3 years.

Michael - Thanks for your response. Yeah actually its not just my stop but exits in general are too discretionary for my liking. Id prefer it to be more systematic.

And dont mention WMT to me, please.

Snake - Good points. Yeh a stop is there to its job, but it must be placed in a strategic place so that the stop being hit means that the trend has changed ie. you dont want to be holding this because the reason you bought the stock is no longer there. Obviously you know all this, but yeh just reminding myself...

And also (though off topic) - is there always an equities bullmarket somewhere in the world?
Any stockmarkets out there in a bearmarket right now?

 

[MichaelD]

And dont mention WMT to me, please.

When the trade's over, win, lose or draw, I plan to write this one up as a great example of what not to do. Classic errors that I haven't made in a long time.

Oh well, at least I'm on the right side of the seemingly invincible WOW with a fully pyramided position (CFDs - even sweeter that most of it's not my money). It dwarfs any short term silliness.