LaSalle Trading Systems for Tradestation

[AB763]

I bought their Recession/Depression package of Indicators and Trading systems and WOW.
I have made good money with the RemaBands and CYOSC trading systems trading the EC and NQ. Haven't tried their other stuff yet but these are sure worth the $285.
Home ?(LaSalle Trading Systems)?
I used 89 tick bars on the NQ and 144 tick bars on the EC. I had not used tick bars before but they do seem work better (less noise). Followed the optimization instructions and then auto traded the systems on the simulator for most of the day before live trading. and Yup they work.:clap:

Just a heads up for those who want to try something new.
AB763

 

[StratOpt]

Be very careful accepting any optimization results that are the result of using a tick based bar of any duration. Also be very careful accepting any results from the simulator. Non-critical acceptance of computer results will often lead you to a very disappointing real time result. Know the pitfalls of all of your testing processes before you drop a single thin dime into the markets.

 

[pedro01]

Be very careful accepting any optimization results that are the result of using a tick based bar of any duration. Also be very careful accepting any results from the simulator. Non-critical acceptance of computer results will often lead you to a very disappointing real time result. Know the pitfalls of all of your testing processes before you drop a single thin dime into the markets.

Agreed.

In terms of backtesting a strategy with market orders - TS will always show a buy at the LOW of the bar you buy into and a sell at the high of the bar you get out of (reverse for shorts)

The more bars in an average trade & the larger the moves you are getting in on, the less this will matter.

Trading strategies that use limit & stop orders are much more reliable in terms of results.

 

[StratOpt]

If you are using next bar market orders then the price shown will be the open of the bar always. This is the type of order that will be the most reliable in terms of backtesting. Those that use limit orders are not more reliable in terms of testing since you are not certain if that order will have filled in real time and filled or not can have a very dramatic effect on the results. The proper testing of the strategy is the most critical aspect of the entire development cycle so don't take it lightly.

 

[pedro01]

Well - that's not what mine does !!!

Here - I have used buy/sell short next bar at market but as you can see the buys put my entry at the low & the sells put the entry at the high.

This, in my opinion, is why market orders are less realistic than using limit orders.

Also the market you are testing on matters - if you are trading the mini-Dow overnight, then you would not expect to have your limit orders filled each time. If you are trading the e-mini S&P, then you should not have any issues.

 

[StratOpt]

I am not trying to argue with you about this. I am trying to help out the original poster and anyone else who may be reading. A next bar market order on a strategy that is not using IntraBarOrderGeneration will always be the open of the next bar in TradeStation. It is when using IOG also, but it will be the next tick that is picked, and then later on with historical testing it will likely change. Next bar market orders are far more reliable than limit orders. This is not my first rodeo using TradeStation. I have written more than a couple lines of code. Also limit orders ALWAYS matter. You will go into the queue of orders when you push out a limit order and in the ES this can be hundreds of contracts deep. You push out a limit order to try and get a better price. If the market turns and heads off in your desired direction and the price doesn't trade through yours first then you will not get filled and it doesn't matter what the symbol is. the only time you can say that your order would have been filled is when price moves through your limit price. This failure to fill has quite a lot of effect on a trading strategy especially those who have entries that are conditional upon the trade having occurred. If you do happen to be using IntraBarOrderGeneration for your strategies then you need to take what I said about being careful when testing and double it.

If you frequent the TSW forums much you will also know me as _Nemo

 

[pedro01]

I am not trying to argue with you about this.

No need to argue - you are saying it works one way and it works in another on my machine. Let's also consider the fact that we can easily prove this with a simple 3 or 4 line script....

I am trying to help out the original poster and anyone else who may be reading. A next bar market order on a strategy that is not using IntraBarOrderGeneration will always be the open of the next bar in TradeStation.

Then how do you explain my post that shows this is not correct when backtesting ?

It is when using IOG also, but it will be the next tick that is picked, and then later on with historical testing it will likely change. Next bar market orders are far more reliable than limit orders. This is not my first rodeo using TradeStation. I have written more than a couple lines of code. Also limit orders ALWAYS matter. You will go into the queue of orders when you push out a limit order and in the ES this can be hundreds of contracts deep. You push out a limit order to try and get a better price. If the market turns and heads off in your desired direction and the price doesn't trade through yours first then you will not get filled and it doesn't matter what the symbol is. the only time you can say that your order would have been filled is when price moves through your limit price. This failure to fill has quite a lot of effect on a trading strategy especially those who have entries that are conditional upon the trade having occurred. If you do happen to be using IntraBarOrderGeneration for your strategies then you need to take what I said about being careful when testing and double it.

If you frequent the TSW forums much you will also know me as _Nemo

I am not new to programming either. I have been doing it since I was 12. I had my first commercial system running when I was 17. I am now 38 - that's 26 years of programming. I run a software company.

I have shown quite clearly how a "buy next bar at market" when backtesing puts you in at the low. I have also shown how a "Sell short next bar at market" puts you in at the high.

There are 2 possibilities here:

1 - my TS (8.5) & your TS are set up differently and are showing strategy entries at different points - in which case, it is in both our interests to understand the difference.
2 - you are wrong
3 - I am wrong

How about we prove this point ? Right here on this thread and get to the bottom of why you think a "buy next bar at market" gets you in at the open & why I think it gets you in at the low ?

I have no problem with being wrong - but you are supposed to be an expert and I have shown that the buy next bar has a set of rose tinted spectacles. As yet, you haven't shown anything...

 

[StratOpt]

I talked with one of my students today and he told me that there is a big problem with 8.5 and the way it is dealing with orders. I haven't messed with 8.5 much because TradeStation has now made it so that you can't easily import and export strategies backwards in version so I develop mostly in 8.3 and I alpha test 9.0. I will spend some time in that 8.5 version to see what this flaw is. It is a flaw or an oversight with some settings or something however. It was never designed to do that and it never will be. so I believe that you are seeing what you are seeing, but trust me it was not meant to be this way.

 

[StratOpt]

As I expected and have explained, if you are using next bar market orders then the price that you will get will be the open of the bar. This has not changed at all in version 8.5 or any of the many various versions all the way back to my first usage of the TS product from 2000i and it has not changed all the way up through version 9.0. So I have opened up a chart with a simple next bar market Stochastic strategy and I think I know what you, and my student, are doing as I can reproduce a chart that looks like yours. This will go back to what I also mentioned earlier dealing with intrabarordergeneration (IOG). If you put a simple coded next bar market strategy as I have described and the go into the Format Strategy ... Calculation tab and then select the checkbox for Enable IOG then you will get a chart that looks similar to what you are showing. You can also get to this by putting into your code set an [IntraBarOrderGeneration = true] attribute. I think that you will find if you uncheck that box or comment out your attribute that you will find what a next bar market order really looks like and was designed to be. Next bar market orders are the most reliable order type for backtesting that there is in TS. Limit orders are NOT.

 

[AB763]

StratOpt is correct about hoe Tradestation executes market orders. Backtesting in Tradestation will only get you in the ballpark at best. Forward testing by running your automated strategy in Tradestation is as close to the real thing as commercially available.

I had a job of forward testing in TS Sim for a Chicago prop shop. I would run 10 to 20 systems at a time watching for particular market conditions where the systems would break down. IMHO tick bars can provide much better results under broader market conditions.

LaSalle Trading Systems recommends this process: 1) back test and optimize, 2) forward test and prove the optimization and system is working, 3) watch it trade and watch the market for changes. be prepared to turn it off or modify its settings as the market changes. I used their programs daily and alwasy run the 3Line Trender to help identify the type of market conditions I'm in at any one time.