Has anyone tried Turtle methodology intraday? (here are the rules)

[1a2b3cppp]

System 1:

Enter long on 20 period donchian channel breakout to the upside, short opposite. Exit long when 10 period donchian channel low is crossed. Exit short when 10 period donchian channel high is crossed.

System 2:

Enter long on 55 period donchian channel breakout to the upside, short opposite. Exit long with 20 period donchian channel low is crossed. Exit short when 20 period donchian channel high is crossed.

If you get a signal for System 1, but the previous System 1 signal was profitable, then you don't take the current signal and instead wait for a System 2 signal. Dunno why that's the rule, but that's the rule.

So for example:

System 1 entry = stopped out for a loss. Then you get another System 1 entry and you can take it.

But:

System 1 entry = win. Then y ou get another System 1 entry. You cannot take it cuz the previous System 1 entry was a win.

The rules for adding to winners are like... take the 20 period ATR and add one more unit every time price goes up (in a long trade) or down (in a short trade) by the value of the ATR. Ok so like 20 period ATR is 2 and you enter at 100, that means your next add would be at 102, and 104, and 106, up to a maximum of 4 "units."

Stops: twice the ATR below a long position or above a short position, and this should not exceed 2% of your account (so you'll have to play around with your position sizing). Every time you add another unit (cuz price moves up by ATR) you move your stop by 1/2 the ATR.

Yes, this means that even after a fully loaded 4-unit position, your stop will still be BELOW your original entry price. No locking in profits. This takes BALLS. But this system counts on hitting the occasional home run, so no locking in profits and getting stopped out.

So anyway, the Turtles did this on the daily charts.

But charts are fractals, right? Timeframe doesn't matter. So theoretically this should work intraday.

Except people say it doesn't work anymore, or at least System 1 doesn't work anymore. Beats me.

How were the markets in 1983 or whenever they were doing this?

Like everything else, seems like this would be awesome in a trendy market and worthless crap in a choppy market.

Someone do this or code it and backtest it. I saw it once backtested on Forex but it was from an EA vendor so I don't trust it.

 

[Roark]

You are trying everything and nothing is working. Why don't you try one thing and try it long enough to see if it works or not. Try something simple, like opening gaps. Out of the last 100 trading sessions, how many times did the opening gap fail to close during the session?

 

[1a2b3cppp]

Nah this was just kind of a secondary discussion I wanted to have.

SPY gaps didn't close very often i the last 6 months. I never "got" gap trading. I understand how it's supposed to work, but it seems like if the gap doesn't close then you end up on the wrong side of a runaway. But SPY is trending like a mofo over the last 6 months, so yeah.

 

[Roark]

The Turtle system did not work for me intraday.

Concerning gaps, you are not limited to just SPY. There are other securities and markets.

For example have a look at OPEN, over the last 100 trading sessions it gapped up 57 times and down 42 times. It closed the gap during the session 86 times, i.e., 86 out of 97 times, it closed the gap during the session.

With respect to only gap ups, it closed the gap 49 times out of 57, or about 86% of the time. Looking at only downward gaps, such gaps were closed 37 times out of 42, or about 88% of the time.

The average gap is 0.58, including one monster gap of 9.68. Might be something to exploit there.

 

[psytrade]

You obviously don't understand trend following if you are looking for trends in mean reverting time frames.

Ed Seykota had a post on his Trading Tribe site saying you could trend trade all the way to single bursts of light if you wanted to. Mind you, the transaction costs would eat the profits.

 

[1a2b3cppp]

Quote from Roark:

The Turtle system did not work for me intraday.

What about anti-turtle intraday :eek:

Concerning gaps, you are not limited to just SPY. There are other securities and markets.

I know. I like to stick with the indexes tho because they're all I've traded for the last few years and I'm comfortable with them. There's too much crazy stuff that can happen with individual stocks, plus I tend to spend hours playing around with stock screeners looking for good ones.

For example have a look at OPEN, over the last 100 trading sessions it gapped up 57 times and down 42 times. It closed the gap during the session 86 times, i.e., 86 out of 97 times, it closed the gap during the session.

With respect to only gap ups, it closed the gap 49 times out of 57, or about 86% of the time. Looking at only downward gaps, such gaps were closed 37 times out of 42, or about 88% of the time.

The average gap is 0.58, including one monster gap of 9.68. Might be something to exploit there.

I'll leave that to other people.

Do you just enter at the open if there's a gap? Do you average into the price assuming that it will close? Where do you put your absolute hard stop in case price runs away?

I'm just curious. I don't really think I'll ever trade gaps.

 

[1a2b3cppp]

Quote from psytrade:

You obviously don't understand trend following if you are looking for trends in mean reverting time frames.

How is intraday any more "mean reverting" than hourly or daily charts?

If I posted 10 charts without labels could you tell me which ones were intraday and which ones were daily?

Ed Seykota had a post on his Trading Tribe site saying you could trend trade all the way to single bursts of light if you wanted to. Mind you, the transaction costs would eat the profits.

Now I agree with you there. That's why I prefer longer term trading to day trading: less noise and stuff to deal with and your costs are proportionally smaller. I don't think there's a different between daily and 1 minute charts other than the percentage of your position size comapred to slippage and commission.

In other words, if a swing trader gets filled 5 cents below what he wanted, he probably won't care.

If a scalper does, that might mean the difference between a profit and a loss on that trade.

 

[1a2b3cppp]

I just realized something that may be detrimental to day trading this strategy. What if you're still in the trade at the end of the day. Do you hold overnight? That's a big no-no in day trading land.

 

[jokepie]

Quote from 1a2b3cppp:

I just realized something that may be detrimental to day trading this strategy. What if you're still in the trade at the end of the day. Do you hold overnight? That's a big no-no in day trading land.

IT can be implemented based on the fact that the time frames do not matter. Problem is in execution and your response time.
You practically have to trade the same strategy in 30-20 minutes, that you deploy in 6months !!!
Its like time trave..!!!

 

[1a2b3cppp]

I read somewhere else it was tested in "Trader Magazine" (or something like that) intraday and it didn't perform well.

So if it doesn't work you could modify it into a fading/averaging down system (with hard stops of course so you don't blow your account).

But we might just be getting back to the argument of trend following/RTM systems and the key is knowing which one is going to work ahead of time.

So fading it would be like... price breaks 20 period channel to the upside, short. Target is low of 10 period channel.

Average down at the normal "add" places.

Hard stop somewhere... I dunno.


Damn, just from eyeballing this on SPY on Monday, this missed a lot when you follow the normal rules. Enter, stop out for loss. Enter, stop out for loss. etc etc. I guess that's what "trend followers" do tho, huh, take a ton of trades and hope for the big one.