Turtle Strategy- for daytrading?

[UKtradergirl]

I am writing this on the assumption that a lot of people will have heard/ read about the Turtles and their strategy.

Anyway, i just wondered if anyone has tried applying similar rules to a daytrading/shorter term strategy?

e.g. The turtles used a 20 day breakout strategy- and a 55 day breakout strategy. How about replacing the daily bars with 5 minute bars, or 1 minute bars? Also adapting the other rules as appropriate. (ATR, stops, etc).

I've not looked at this in any detail - although a quick glance at my open chart (which happens to be oil) shows that this would have caught the big downward move that happened between 6.30pm and 7.30pm (UK time).

Although this strategy is prone to whipsawing- it obviously worked for the turtles- as jumping on board a decent trend can more than make up for a multitude of losses.

Anyway, just wondered if anyone had any opinions on the above?

Also.. does anyone trade breakouts of any kind- of the variety of breaking out of x number of previous bars range? I want to hear your experiences here too.

 

[rags2riches]

Something vaguely similar over here

The System II (Yes with irony)

Have a browse.

 

[Trader333]

Although this strategy is prone to whipsawing- it obviously worked for the turtles- as jumping on board a decent trend can more than make up for a multitude of losses.

The Turtles didn't day trade and this is important to consider. Day trading is a very different animal to EOD in my view and capturing intra-day trends is fraught with major difficulties. It is highly unlikely that an approach that works for EOD will work intra-day in my view.


Paul

 

[Rhody Trader]

The basic problem with daytrading and the Turtle method is that by definition daytrading implies closing out positions before the end of the day. The Turtle method would require holding if the trend is still on. That's going to mean overnight at times.

 

[chilltrader]

plus they had REALLY deep pockets...

 

[nine]

rags2riches is right.

The System II (Yes with irony) by feb2865 is refreshing in a site dedicated to indicators. The section covering turtle like breakouts is late in the thread. Its not a straight channel breakout system but it is useful if you're thinking about breakouts for day trading.

You could use turtle in the trendier daytrading markets but the downside of a breakout system is large stops and/or low win rates.

 

[UKtradergirl]

I checked out the thread linked to above. It is a good thread- have read about 5 pages and will spend some more time reading it too. I have always traded pullbacks, but normally enter sooner than this guy does- like, when the market starts going back up again, not when it gets past the last swing high.

Will keep reading the thread there anyway.

Some good points made above too- thanks.

 

[BSD]

Hi there Donna, why not ask at at the source, after Richard Dennis himself Curtis Faith was the most successful Turtle who turned sthg like 100K into around 30 million in one or two years and is a very accessible, friendly guy.

This is his website / forum, I am sure either he himself or others there have backtested for using the system intraday:

Trader's Roundtable :: Index

 

[UKtradergirl]

Hi there Donna, why not ask at at the source, after Richard Dennis himself Curtis Faith was the most successful Turtle who turned sthg like 100K into around 30 million in one or two years and is a very accessible, friendly guy.

This is his website / forum, I am sure either he himself or others there have backtested for using the system intraday:

Trader's Roundtable :: Index

Curtis Faith's book is very good. I will check out the link you posted, thanks a lot BSD.

 

[hackenbush78]

I'm bringing this old thread up because I was wondering the same as UKtradergirl ... ie could the Turtle Strategy be successfully implemented into day trading ? I take on board the points raised that trades must be cut if they are running towards the close, or else run the risk of overnight volatilty and of course that would defeat the purpose of such a well drilled and rigid proven trading stategy.

However, if you take the example of the Turtle's 20 day system and equate say a 1 minute chart to their 1 day chart, you find that 20 minutes is 3.3% of a 10 hour trading day, whereas 20 days is 5.4% of one year. I wonder what the average lifespan of a Turtle's 20 day system trade actually was. I bet very few were held over a couple of months or so !

Granted, there will be occasions when a trade set up may present itself later on in the day and it will either need to be declined or run to the close and then cut. So in essence the Turtle System cannot be replicated exactly in the day trading environment, but I can't see why there cannot be ample other opportunities ..perhaps 90% of opportunities could be taken up ?

I'd also imagine that perhaps a one minute chart isn't ideal, though 5 mins may be a tad long. The 1 minute could get rather frantic, especially in terms of working out "N" every 20 minutes. And I'm sure trading multiple currencies/futures contracts etc would not be for the faint hearted.

It would be great to be able to follow the Turtle's system to the letter in the world of day trading. I'm enough of a realist to think that it's not 100% possible, but also enough of an optimist and have trawled enough 1 and 5 minute charts to think that it might be worth investigating further.

Btw, BSD, Trader's Roundtable looks excellent ...thanks for posting the link.

 

[hackenbush78]

Just had a reality check ... if one minute represents one day, then N, True Range and Dollar Volatilty etc would need to be updated every minute. Even on a 5 minute chart it's too frequent.

So, no I don't think the Turtle System can be replicated for day trading, but elements of the system definitely can be ...especially the breakout trading rules, adding to positions, stops and profit taking rules.

But it's obvious that one's own risk management and assessment of volatilty would need to be implemented.

 

[rathcoole_exile]

J

So, no I don't think the Turtle System can be replicated for day trading,

check out Hubert Senter's** approach using breakouts/downs from Darvas Boxes for day-trading.
It's something along these same lines....

(**tradethemarkets.com - and NO, I'm not affilliated with them in any way)

 

[hackenbush78]

Many thanks for posting, Rathcoole ...will check this out later.

 

[meanreversion]

The reason the Turtle system worked when it was applied (although check this link http://www.automated-trading-system.com/turtles-just-lucky/) is because trends occurred and using wide stops like 2N (e.g. this would be 350 pips in GBP/USD) meant the system could get through market noise.

For day trading, the noise is a much larger component of the move. I've tested turtle on shorter time frames and have not seen good, consistent results.

The two main lessons from turtle are 1) cut losses/run winners and 2) the trend is your friend. These should both be applied in day trading .. the usual recommended approach is identifying the underlying trend and then looking for retracements to enter in that direction. This strategy has a higher chance of success than a breakout approach, which (in day trading) more often than not means you'll be selling the low/buying the high with a tight stop which the "noise" will almost certainly trigger.

 

[Adamus]

You need to define whether you mean 'day trading' as brokers mean it - i.e. you exit trades at the end of day - or you mean intraday time frames, with no requirement to exit EOD.

 

[hackenbush78]

Thanks for the replies.

mainrevision, you make some interesting points and I appreciate that you have obviously spent a lot more time researching day trading results using Turtle Strategies than I have. I would assume you are right and there are false breakouts where one buys at the top or sells at the bottom, but due to their stringent stop out criteria there would be minimal losses, one would think. It could of course get ugly in prolonged sideways markets with the inevitable false breaks but only the first tranche or so of units would be taken, but one would think there should be few occasions where one would enter trades in those conditions anyway, would there not ?

As someone who has always been a day trader looking to buy eg on retracements of an uptrend I am comfortable with what you are saying there, but isn't that totally opposed to Turtle Strategy as they are only looking for breakouts after prolonged (relatively)sideways action only ?

Adamus, I meant day trading as brokers mean it, ie exit at close of day. Therein lies the inherent problem ... ie what happens if one gets a fantastic signal in the last hour before close of trading ? Well a decison has to be made whether to decline, to take and then cut at close even if the Turtle Strategy would indicate to keep, or, as Forex is a 24 hour market place in the most part, to hold and nurse it around the clock. I haven't yet decided what the best policy is in that regard.

Finally, £/$ was an absolutely great market to exploit the system today. Looking at 1 min and 5 min charts there was a breakout at around 7am UK this morning, though some of the profit would have been given up on a retracement. I think there was also a chance to short it afterwards.

If day trading, using 5 min charts, one must get their head round the fact that a day is actually more like 3 months in the Turtles' world.

 

[meanreversion]

You need to do your own research. This forum is good for getting pointers, but you MUST backtest all these things yourself. Forget that it worked this morning in cable, how did it perform over the last six months? And then test it across different asset classes etc. Until you truly believe that your system has positive expectancy, there is no point attempting to trade it (you will just abandon it when drawdown becomes uncomfortable).

There are no short cuts here. Get decent software, learn to program and test until you're blue in the face. Only that way will you have confidence to execute your plan.

 

[hackenbush78]

Absolutely, when I mentioend this morning's cable I just gave it as an example of a trending market that would have been caught by the system. Swallows and summers though, of course.

When you mention software, is there software out there that can calculate things like True Range, N, Dollar Volatilty off of ultra short term charts, ie 1 min and 5 min and can then be indicated to a trader quickly and instantly ? If not I suppose a half decent programmer could drum something up.

 

[meanreversion]

I use Amibroker, but ProRealTime is also good, eSignal etc.

Don't pay for a programmer. You need to have confidence in the system, and if someone else has written it, you won't.

 

[hackenbush78]

Many thanks ...appreciated.