Buy a New Daily High or Sell a New Daily High?

[mrsoul]

Hello,

I wonder if it is profitable to either buy a new daily high or sell a new daily high in Forex.
Also I wonder if it is profitable to sell a new daily low or buy a new daily low.

I would make this trade and hold the position until the close of the day with no stop.

My instincts tell me that this set-up might be no better than 50%; however, there is a chance that it could be statistically significant and definitely worth a back-test.

Is anyone able to back-test this idea for a few years and if so, could you also let me know the overall profit or loss for the time period and the maximum number of winning and losing trades and pips in a row?

Thanks

 

[cr6196]

This idea of what to do at new highs/new lows was in "Street Smarts". My personal view is that it depends on context. This idea of selling a new high/low was called Turtle Soup ( i believe it was a 20 day high) and was intended as the opposite is what the Turtles did. I believe there are backtesting results in the back of the book.

The results showed that it worked well in bonds, this was presumed to be because bonds were mean-reverting ( i don't trade them, so i don't know) and that there was a desire, at the time, to fake people in on the breakout to create demand to create supply...the book was written when trend-following systems were extremely popular.
I think it is interesting idea but it depends on context and the current state of the market. In forex, at the moment fading breakouts (at the right points) would work quite well.

 

[mrsoul]

This idea was in "Street Smarts", my personal view is that it depends on context. This idea of selling a new high was called Turtle Soup ( i believe it was a 20 day high) and was intended as the opposite is what the Turtles did. I believe there are backtesting results in the back of the book.

The results showed that it worked well in bonds, this was presumed to be because bonds were mean-reverting ( i don't trade them, so i don't know) and that there was a desire, at the time, to fake people in on the breakout to create demand to create supply...the book was written when trend-following systems were extremely popular.

Interesting response cr6196.

Can you please clarify what you mean when you say that " This idea of selling a new high was intended as the opposite as what the Turtles did" ?

Did the Turtles sell or buy new highs?
I am not sure what you meant.

-Best of Luck-

 

[GladiatorX]

In stocks i've found shorting breakouts to be very effective compared to buying the breakouts... Usually they are false, when they're not, a pullback will go much beyond entry and on the rare occasion when a breakout is 'text-book' momentum, you lose.

Also on the E-mini S&P500, a setup i use is to wait for yesterdays high to be closed above, then i wait for a close back inside to short. I do add discretion to this however. In general a false breakout would create a large move down to the opposite end of yesterdays range.

 

[cr6196]

sorry, they bought new 20 day highs, another point, is they took a lot of losses, 30% of trades were profitable but many of them made 100%+ annually due to the size of the winning trades. Obviously this wasn't every year and there are exceptions etc etc.
If your interested in getting a copy of street smarts, PM me.

 

[mrsoul]

In stocks i've found shorting breakouts to be very effective compared to buying the breakouts... Usually they are false, when they're not, a pullback will go much beyond entry and on the rare occasion when a breakout is 'text-book' momentum, you lose.

Also on the E-mini S&P500, a setup i use is to wait for yesterdays high to be closed above, then i wait for a close back inside to short. I do add discretion to this however. In general a false breakout would create a large move down to the opposite end of yesterdays range.

Have you done any back-testing?

 

[GladiatorX]

Have you done any back-testing?

Yes I've done hours and hours of backtesting on pretty much everything i could think of so that i can look for setups of confluence, always with discretionary elements ofcourse

 

[pedro01]

In a ranging market - sell new highs.

In a trending market - buy new highs.

If you aren't sure which you are in, best to do nothing.

 

[shadowninja]

So much whipsawing goes on these days; I'd use new highs or new lows to close positions.

 

[GladiatorX]

So much whipsawing goes on these days; I'd use new highs or new lows to close positions.

I believe for anything to be viable as an exit to a current trade, it must be viable as an opposite trade... Otherwise really your exit doesn't have any edge, therefore regardless of where you exit. It won't help