Sell order based on study condition not triggering

[InfamousCookie]

While doing some testing with live trading, I was having an issue with conditional sell orders not triggering. I'm using Accumulation Swing Index as my sell study. The code is posted below. I have it set up to sell the stock when asiplot is below smaplot but doesn't seem to always trigger. Seems like the trigger rate is about 30% or so. I'm currently looking at a stock that should have sold mid day yesterday using this conditional selling code but it's still sitting there despite the chart clearly showing that the asiplot is below the smaplot. I have checked to make sure the times on the chart and condition are the same as well as the input parameters are the same. Any idea what I'm missing? Thanks for the help.

#TOS Accumulation_Swing_Index
declare lower;

input smaLength = 10;
input averageType = AverageType.WILDERS;
def limit = 30;
def AbsHighClose = AbsValue(high - close[1]);
def AbsLowClose = AbsValue(low - close[1]);
def AbsCloseOpen = AbsValue(close[1] - open[1]);
def K = If(AbsHighClose >= AbsLowClose, AbsHighClose, AbsLowClose);
def R = If(AbsHighClose >= AbsLowClose,
        If(AbsHighClose >= (high - low), AbsHighClose - 0.5 * AbsLowClose + 0.25 * AbsCloseOpen, (high - low) + 0.25 * AbsCloseOpen),
        If(AbsLowClose >= (high - low),  AbsLowClose - 0.5 * AbsHighClose + 0.25 * AbsCloseOpen, (high - low) + 0.25 * AbsCloseOpen));
def nRes = If(R != 0,
            (50 * (((close - close[1]) + 0.50 * (close - open) + 0.25 * (close[1] - open[1])) / R ) * K / limit) + if !IsNaN(nRes[1]) then nRes[1] else 0,
            0 + if !IsNaN(nRes[1]) then nRes[1] else 0);

def asi = nRes;
def sma = Movingaverage(averagetype, asi, smaLength);
def ASI_crossesbelow = asi crosses below sma ;
plot asiplot = asi;
plot smaplot = sma;

 

[irishgold]

Seems to be a pretty good exit indicator. I am assuming you are seeing the actual condition on the chart but the order not responding? I have heard and experienced myself issues with Conditional orders based on Studies not firing. It's an issue with TOS. Maybe try creating an alert based on the crossbelow and see what reliability of the alert is . And make sure you have set the timeframe ie 5 Min in the conditional order, it usually comes up Day.
I would use Plot instead of def for ASI_crossbelow then just say when that is true send alert, if that seems to be working then try again in conditional order but like I said I don't know that conditional orders work consistently.

 

[InfamousCookie]

Seems to be a pretty good exit indicator. I am assuming you are seeing the actual condition on the chart but the order not responding? I have heard and experienced myself issues with Conditional orders based on Studies not firing. It's an issue with TOS. Maybe try creating an alert based on the crossbelow and see what reliability of the alert is . And make sure you have set the timeframe ie 5 Min in the conditional order, it usually comes up Day.
I would use Plot instead of def for ASI_crossbelow then just say when that is true send alert, if that seems to be working then try again in conditional order but like I said I don't know that conditional orders work consistently.

You are correct; I see the conditions being met in the chart but the sell order not firing. Trying the plot instead of def for ASI_crossbelow is a good idea. I'm going to give that a try with the conditional order. I thought about doing alerts but would prefer to have it auto trigger due to having a full time job during the day, chances are good I couldn't get online to do the trade if the alert triggers. However, if the conditional order keeps giving me issues I might not have a choice. Thank you for the suggestion!

 

[MerryDay]

I used the ASI in conditional orders successfully in the past.

I am assuming that you chose to reference your custom study in your OCO.
That is allowed and theoretically should work if you wrote the OCO correctly.

You didn't provide the script for your conditional order so it isn't possible to comment on where the issue might be.

You could try referencing the ToS indicator in the conditional order and create the custom entry/exit condition. This is my preferred method and I have never had an issue with my OCOs.

What is OCO script? How are you creating your exit condition?

TDAmeritrade has several caveats listed as to the causes of OCOs not triggering. You could google their commentary and see if it applies to your situation.

 

[InfamousCookie]

I used the ASI in conditional orders successfully in the past.

I am assuming that you chose to reference your custom study in your OCO.
That is allowed and theoretically should work if you wrote the OCO correctly.

You didn't provide the script for your conditional order so it isn't possible to comment on where the issue might be.

You could try referencing the ToS indicator in the conditional order and create the custom entry/exit condition. This is my preferred method and I have never had an issue with my OCOs.

What is OCO script? How are you creating your exit condition?

TDAmeritrade has several caveats listed as to the causes of OCOs not triggering. You could google their commentary and see if it applies to your situation.

Hi MerryDay,

Here are several pictures showing what I have done. I did happen to look at TOS study condition page and the thing that stood out to me is that studies had to have one plot. So, I actually took the above code and marked out plots creating an ASI only and SMA only plot. Then I created the study based order using those two codes. Before, I was using the code above only. Of course, even creating separate plots didn't cause the sale to trigger on some trades.

ASMASIONLY("sma length" = 5, "average type" = "EXPONENTIAL") is less than ASMSMAOnly("sma length" = 5, "average type" = "EXPONENTIAL")

#TOS Accumulation_Swing_Index ASI ONLY
declare lower;

input smaLength = 10;
input averageType = AverageType.WILDERS;
def limit = 30;
def AbsHighClose = AbsValue(high - close[1]);
def AbsLowClose = AbsValue(low - close[1]);
def AbsCloseOpen = AbsValue(close[1] - open[1]);
def K = If(AbsHighClose >= AbsLowClose, AbsHighClose, AbsLowClose);
def R = If(AbsHighClose >= AbsLowClose,
        If(AbsHighClose >= (high - low), AbsHighClose - 0.5 * AbsLowClose + 0.25 * AbsCloseOpen, (high - low) + 0.25 * AbsCloseOpen),
        If(AbsLowClose >= (high - low),  AbsLowClose - 0.5 * AbsHighClose + 0.25 * AbsCloseOpen, (high - low) + 0.25 * AbsCloseOpen));
def nRes = If(R != 0,
            (50 * (((close - close[1]) + 0.50 * (close - open) + 0.25 * (close[1] - open[1])) / R ) * K / limit) + if !IsNaN(nRes[1]) then nRes[1] else 0,
            0 + if !IsNaN(nRes[1]) then nRes[1] else 0);

def asi = nRes;
def sma = Movingaverage(averagetype, asi, smaLength);
def ASI_crossesbelow = asi crosses below sma ;
plot asiplot = asi;
#plot smaplot = sma;

#TOS Accumulation_Swing_Index SMA ONLY
declare lower;

input smaLength = 10;
input averageType = AverageType.WILDERS;
def limit = 30;
def AbsHighClose = AbsValue(high - close[1]);
def AbsLowClose = AbsValue(low - close[1]);
def AbsCloseOpen = AbsValue(close[1] - open[1]);
def K = If(AbsHighClose >= AbsLowClose, AbsHighClose, AbsLowClose);
def R = If(AbsHighClose >= AbsLowClose,
        If(AbsHighClose >= (high - low), AbsHighClose - 0.5 * AbsLowClose + 0.25 * AbsCloseOpen, (high - low) + 0.25 * AbsCloseOpen),
        If(AbsLowClose >= (high - low),  AbsLowClose - 0.5 * AbsHighClose + 0.25 * AbsCloseOpen, (high - low) + 0.25 * AbsCloseOpen));
def nRes = If(R != 0,
            (50 * (((close - close[1]) + 0.50 * (close - open) + 0.25 * (close[1] - open[1])) / R ) * K / limit) + if !IsNaN(nRes[1]) then nRes[1] else 0,
            0 + if !IsNaN(nRes[1]) then nRes[1] else 0);

def asi = nRes;
def sma = Movingaverage(averagetype, asi, smaLength);
def ASI_crossesbelow = asi crosses below sma ;
#plot asiplot = asi;
plot smaplot = sma;

Please let me know what you think or if I can provide any other additional details. I appreciate your help. Thank you.

 

[MerryDay]

You have both a Wait Trg and Wait Cond on the Order above. Your issues have nothing to do with your conditional order.
The Wait Trg is taking precedence over a Wait Cond. Your conditional order is never being processed, I am assuming because of the hold of the Wait Trg/

Here is the Order of Precedence:

Do you see #2 in the description in the bottom of your order rules?
The particular OCO that you displayed in the above post, states that the conditional order must wait for the trigger that is on your group of orders.
As long as that wait trg is there, it will not execute your condition.

The bottom of your order should look like this. Nothing in the way of processing the OCO. No Wait Trg.

 

[InfamousCookie]

You have both a Wait Trg and Wait Cond on the Order above. Your issues have nothing to do with your conditional order.
The Wait Trg is taking precedence over a Wait Cond. Your conditional order is never being processed, I am assuming because of the hold of the Wait Trg/

Here is the Order of Precedence:
View attachment 1105



Do you see #2 in the description in the bottom of your order rules?
The particular OCO that you displayed in the above post, states that the conditional order must wait for the trigger that is on your group of orders.
As long as that wait trg is there, it will not execute your condition.

The bottom of your order should look like this. Nothing in the way of processing the OCO. No Wait Trg.

That makes sense to me. This is probably why if I did a cancel then replace on the conditional study sell with the exact same parameters, it tends to go through since I probably inadvertently got rid of that trigger. The question I have would have is, what's causing that wait trigger then? Before the actual submission of the order above, I see the #2 wait trigger in the part associated with the study order. If I delete the buy portion of the OCO order before actually submitting the order, then that trigger goes away so I would assume that the trigger is associated with the buy order somehow. However, I am currently holding a stock where the buy order went through but the study is still waiting on that trigger you mentioned. To get around this issue, should I be buying the stock alone then setting the sell order separately? Or maybe there is some other setting that I'm missing?

I appreciate your help, MerryDay; you're awesome!

 

[MerryDay]

@InfamousCookie I haven't any idea what you are trying to do. So I can't tell you how to do it.

If your buy and sell orders are not dependent on one triggering the other. Then don't set them up w/ First Triggers.
If they are dependent on each other then your current setup is working. The First Trigger is properly preventing your sell order from being filled

https://tlc.thinkorswim.com/center/howToTos/thinkManual/Trade/Active-Trader/AT-Entering-Orders

 

[InfamousCookie]

@InfamousCookie I haven't any idea what you are trying to do. So I can't tell you how to do it.

If your buy and sell orders are not dependent on one triggering the other. Then don't set them up w/ First Triggers.
View attachment 1107
https://tlc.thinkorswim.com/center/howToTos/thinkManual/Trade/Active-Trader/AT-Entering-Orders

In a nutshell, as soon as I buy a stock and the buy is filled, I want it to create an OCO with the conditional study order and the trailing stop loss with that stock immediately. But I will play with it. Thanks.

 

[Scott111111]

I just ran across this post. I as well have had study orders that I have saved as custom using the 'Save Custom Order Template'. After some playing around I do not believe that it is me (user error) and it has something to do with the custom order.

Just opening the custom order and hitting confirm and send the order will sit in a Wait Cond forever. However, if I go into the order then the study and hit edit. Change nothing hit then ok and then save it will execute using the correct condition.

 

[Michael Paul]

Just opening the custom order and hitting confirm and send the order will sit in a Wait Cond forever. However, if I go into the order then the study and hit edit. Change nothing hit then ok and then save it will execute using the correct condition.

I was having the exact same issue, and your observation solved the problem. Thanks for sharing!

 

[Scott111111]

I was having the exact same issue, and your observation solved the problem. Thanks for sharing!

Sure, that one had me perplexed for a while. The more I think about it is that we need to go into it to get the settings.

 

[Michael Paul]

Sure, that one had me perplexed for a while. The more I think about it is that we need to go into it to get the settings.

Yeah I think you're right, but it is annoying as my study is not changing each time with my saved order. Bunch of extra clicks for no real good reason.

 

[fisbo]

This work around is working out great but a real pain. I tested a few different scenarios and observed that the issue only comes up for custom built studies.

 

[zoyd]

I have reported this issue to TOS and they confirmed it as a bug. The template does not load the conditional script, you have to load it manually by clicking edit or the script name in the gear dropdown.