Daily range script scan - Error / Omission

[SCALPER]

Hello,

Today 4/9/21, I found an error / omission using the following daily range script -

(((highest(high[1],20) – lowest(low[1],20)) / lowest(low[1],20)) * 100) >= 100

Specifically, it omitted BTX with a 20 day range from 3.59 - 10.28 or 186.4%

Any help with a modification or working script is greatly appreciated.


Direct link to the Scan Query
Screenshots of the results w/ and w/out the script -

 

[rad14733]

@SCALPER Do you have a link to the original post you are referring to...???

 

[SCALPER]

@rad14733 ... Hi, I asked for basic code to determine the highest / lowest values in the thinkScript Lounge chat in TOS and pieced together the script using a percentage formula to define the range. I wasn't able to find much regarding ranges in here and hope someone is able to solve the error.

 

[rad14733]

@rad14733 ... Hi, I asked for basic code to determine the highest / lowest values in the thinkScript Lounge chat in TOS and pieced together the script using a percentage formula to define the range. I wasn't able to find much regarding ranges in here and hope someone is able to solve the error.

I don't bother with Thinkscript Lounge... I've visited several times but have found it to be a waste, overall...

That being said, we can't help if you don't post images of the Thinkscript or Condition Wizard code from your custom filters...

 

[SCALPER]

I hear you and appreciate the assistance. I'm a noob when it comes to coding but I believe the logic is sound.

Here is the Scan Query. To simplify things I deleted the additional custom filters except for the predefined parameters and the custom range script -
(((highest(high[1],20) – lowest(low[1],20)) / lowest(low[1],20)) * 100) >= 100

The screenshot shows BTX is not listed when the custom range script is run, yet it should be included since its 20 day range is 3.59 - 10.28 or 186.4%. Chances are I am using the wrong code to define the highest high and lowest low of the range resulting in the error / omission.

 

[MerryDay]

@SCALPER Good News Bad News
Good News, your logic and syntax in your script is doing just what you want it to.

In my personal experience, pasting scripts directly into the scanner can hide issues and sometimes give unreliable results. So I put your code into a study.

input length = 20;

plot LowerBand = Lowest(low[1], length);
plot UpperBand = Highest(high[1], length);
plot Diff      = UpperBand - LowerBand ;
plot range = Diff / Lowerband *100 ;
range.hide();
plot test = range > 100 ;
AddChartBubble(IsNaN(close[-1]), close, "HighestHi: " + Highest(high[1], 20) + " |" + "LowestLow: " + Lowest(low[1], 20) + " | " + "Range: " +range  , color = Color.ORANGE);

I ran the study through the scanner. It got really ugly. I got the following error AND it froze and corrupted my TOS app.

My experience has been that scripts that result in the above error message mean that your code will not run in the TOS scanner. It might be taking too many resources. I have no idea, when people start explaining program errors to me, my brain turns off. If you are intensely interested in why this code will not run, you could try calling TOS support.
Sorry, it was a neat idea for a code.

 

[XeoNoX]

its the golden study! they dont want you to have it! its the secret to profitability!

yea you will have to call tos support if it errors on your TOS as well, since im not willing to corrupt my TOS app im not really trying to test it, ill take @MerryDay 's word for it

 

[SCALPER]

@MerryDay - Thank you you for putting in the time to narrow the error down to a source. Hopefully TOS has a solution.

@XeoNoX - Haha. Definitely avoid the study. It resulted in the same error message after running this -

 

[krishna_yadu]

Is this high probability success pattern ? just trying to understand what is the significance of 20 day range ?

 

[MerryDay]

@krishna_yadu The script and the usage of a 20 day range is the definition of Donchian Channels
You can read more about them:
https://usethinkscript.com/threads/iso-adjusted-donchian-channel-for-tos.4901/
HTH

 

[SCALPER]

@MerryDay I piggybacked off the idea to reformat. The updated study yields the omitted ticker without an unexpected error. @XeoNoX behold the "golden study" -

# Range >= %
# Scalper - 04/11/2021
# Change 20 / 50 in example below to desired time interval / percentage
# Maximum time limitation undetermined

Def H = highest(high[1],20);
Def L = lowest(low[1],20);
Def Range = ((H – L) / L) *100;

Plot scan = Range >= 50;

If anyone finds a glitch / omission please chime in.
Cheers

 

[MerryDay]

Turns out there was maintenance going on when we ran into issues so it might have been a situational problem.
Good for you @SCALPER for sticking w/ it.

 

[krishna_yadu]

@krishna_yadu The script and the usage of a 20 day range is the definition of Donchian Channels
You can read more about them:
https://usethinkscript.com/threads/iso-adjusted-donchian-channel-for-tos.4901/
HTH

Thanks @MerryDay, heard of Donchian Channels before never thought it is one of the high probability success rate one.

 

[Docbrown83]

Question about the “mark” category, may be a very newb question, but what does this entail?

 

[SCALPER]

@Docbrown83 Mark %Change is useful for tracking real-time % change during pre and post market. The %Change column only tracks regular market hours.

 

[SCALPER]

The formula below has a glitch. It omits any result that has traded less than 90 days, even if the range parameter is met -

Def H = highest(high[1],90);
Def L = lowest(low[1],90);
Def Range = ((H – L) / L) *100;
Plot scan = Range >= 200;

Example -
• ALF - has traded for roughly 35 days and the percentage range is over 580%.
• Several recent IPOs would likely be omitted as well.

When a second formula is added, below the 35 day threshold, ALF is still omitted -

Def H90 = highest(high[1],90);
Def L90 = lowest(low[1],90);
Def R90 = ((H90 – L90) / L90) *100;

Def H20 = highest(high[1],20);
Def L20 = lowest(low[1],20);
Def R20 = ((H20 – L20) / L20) *100;

Plot scan = ( R90 >= 200 OR R20 >= 100 );

The only way ALF will show up as a result is if the number of days it's actively traded is greater than the specified number of days in the formula(s).

I attempted to add a IF condition that results in double.nan if the days traded were less than the specified number in the formula.
These lines were added to the code above and tested separately without any luck -

R90 >= if barnumber()>=90 then 200 else double.nan );
Def R90 = if barnumber()>=90 then (((H90 – L90) / L90) *100) else double.nan;

Out of options... any help is appreciated in finding a solution.