Up Key Reversal Breakout Strategy (EasyLanguage) for YM

[swansjr 10]

In an attempt to get more activity on this “Automated Trading” forum I’m posting a little strategy that I found in a PDF on strategy design. Unfortunately, I don’t have the original PDF or the author’s name at this time. Anyway, the set-up is centered on up-key-reversals (UKR). A UKR is where the most recent daily price bar is 1) making new low over the past X days and 2) the close is greater than the previous day. A buy limit order is then placed just above the daily high of the trigger bar. The position is exited either when a profit target is hit or X days after the initial purchase.

 

I coded this concept in EasyLanguage under TradeStation up and you’ll find it below. I only tested it for YM since that's all I trade at this time. I only tested it for the past 3.5 years. It shows a positive results with a Profit Factor of 3.3. I hope others will find this interesting and perhaps propose improvements. I'm eager to see what others may modify.

 

You’ll notice one of the inputs is to activate a fix ratio position-sizing algorithm. I put this into the code so you can easily see what a position-sizing algorithm can do to a trading system.

 

Oh, do I personally trade this system? Not at this time.

 

Jeff

[dede 10]

Interesting strategy........wondering if you have looked at it in a shorter timeframe such as a 2 or 5 minute chart?

[swansjr 10]

Nope. I've not looked at shorter or longer time frames. I should make a point to test it out sometime in the future.

 

Jeff

[Guest forsearch]

. I put this into the code so you can easily see what a position-sizing algorithm can do to a trading system.

 

Could you, or any forum members, post a copy of this .eld here in PLAINTEXT? Some of us don't have access to TS or MC to decompile the code, but might find it useful to look at it anyway.

 

Thanks.

[daedalus 10]

Wonder how hard it would be to make it so the system could sell as well as buy...

 

Interesting system though. If you used it into some kind of defined trend it could do well... You just don't want it buying every down surge as it has been doing in this recent bear market.

 

And the 2k max stop out stuff kills the results... I wonder how big of a drawdown is really necessary because that can really kill the results.

[sevensa 10]

Could you, or any forum members, post a copy of this .eld here in PLAINTEXT? Some of us don't have access to TS or MC to decompile the code, but might find it useful to look at it anyway.

 

Thanks.

 

 

{ UKR Breakout }

Input:
  ProfitFactor( 9 {0 = no target}),
  StopLossInDollars( 2000 { 0 = no stop}),
  BreakoutAmount( 10 {Purchase limit order at high+breakoutamount in points}),
  LL( 11 {Look back period for the lowest low }),
  BSI( 5 {Bars Since Entry - when to exit trade}),
  UseFixedRatio( false {Enable/Disable fixed ratio position sizing});

Variables:
  ATR(0),
  TotalContracts( 0 ),
  TotalProfit( 0 );


Condition1 = Low < Lowest( Low, LL )[1];
Condition2 = Close > Close[1];

If ( Currentbar > 1 ) And ( marketposition = 0 ) And ( Condition1 ) And ( Condition2 ) Then Begin

  // Ryan Jones Fixed-Ratio Position Sizing
  If ( UseFixedRatio ) And ( NetProfit > 0 ) Then
     TotalContracts = IntPortion( 1 + SquareRoot(1 + 8 * NetProfit/4000) * .5 )
  Else
     TotalContracts = 1;

  Buy TotalContracts contracts next bar at (High+BreakoutAmount) limit;
  ATR = AvgTrueRange(LL);

End;

If ( BarsSinceEntry(0) = BSI) Then sell("expiration") next bar at market;

If ( StopLossInDollars > 0 ) Then SetStopLoss( StopLossInDollars );
If ( ProfitFactor > 0 ) Then SetProfitTarget( ATR*ProfitFactor ); 

[swansjr 10]

If you used it into some kind of defined trend it could do well... You just don't want it buying every down surge as it has been doing in this recent bear market.

 

I was thinking the same thing. Maybe a 200 SMA could act as a filter. Only take trades when price is above. Or if slope of 200 SMA is positive.

 

I would like to spend some time exploring this but I've been too busy.

 

 

Jeff

[goody01 10]

Hello Swansjr,

 

thank you for the strategy.

Did you tried it intraday too and also with other futures like ES or NQ?

 

Thx + good luck,

 

Goody

[brakkar 10]

Hi,

on your chart I see that orders are entered at next bar open, not when price goes over previous bar high.

 

ISn't this a problem for backtesting accuracy ?

How do you deal with this ?

 

Thanks;

Brak

[swansjr 10]

Hello Swansjr,

 

thank you for the strategy.

Did you tried it intraday too and also with other futures like ES or NQ?

 

Thx + good luck,

 

Goody

 

 

I did try it intra day with YM without much success. I did not try it with other markets.

 

 

Jeff

[swansjr 10]

Hi,

on your chart I see that orders are entered at next bar open, not when price goes over previous bar high.

 

ISn't this a problem for backtesting accuracy ?

How do you deal with this ?

 

Thanks;

Brak

 

I don't think it's an issue for backtesting when using daily or minute bars. I understand there is an issue if using tick charts. However, your comment about placing an order over the previous bar high may be an interesting idea to test.

 

I don't trade this system. It's simply an interesting idea I found and thought others would be interested in seeing it.

 

Jeff

[ALBULA 10]

Hello Swansjr,

 

thank you for the strategy.

Do you know this Key Reversal Up strategy in Common Technical Analysis Language (CTL) ?

 

Thanks,

 

Albula

[flyingdutchmen 10]

systems like this one do not tend to do very

well with a fixed stoploss or profittarget, it does not give them

any sensetivity to changes in volatility.

i would recommend use a x-percent of the atr for both sl and tp,

it will give the overal performance a boost

[swansjr 10]

Hello Swansjr,

 

thank you for the strategy.

Do you know this Key Reversal Up strategy in Common Technical Analysis Language (CTL) ?

 

Thanks,

 

Albula

 

Sorry, can't help you with this. I only know EasyLanguage.

[swansjr 10]

systems like this one do not tend to do very

well with a fixed stoploss or profittarget, it does not give them

any sensetivity to changes in volatility.

i would recommend use a x-percent of the atr for both sl and tp,

it will give the overal performance a boost

 

You might be right. My best performing automated strategy uses both a fixed profit target and a fixed stop loss. Both parameters have not been modified in over 18 months. I've yet to find a dynamic target or stop that produces better performance on a net per trade basis.

 

In short, it's important to test all options.

[flyingdutchmen 10]

i am sorry swansjr, i have just read your complete

el script and have viewed your range based target,

i guess i was a litle fast with my previous comment