[mattz]

Originally Posted by gero »

feng2088,
be careful feng2088, it is easy to open a tradestation protected tading system (russian are the most skilled at this ;-)
And ND agreement is worthless, once someone has opened your system, he can use it plain vanilla or modify it. You have no chance to do anything.
Do not send the code outside your pc. And even protect it with a good firewall and antivirus.
Be careful.
gero

I agree that protection is a MUST. But, many customers simply cant sit and supervise their systems all day, so we provide this service. We have no desire to reveal or figure out a customer's methodology. We have many that have trusted us as vendors, and customers. All we want to do is provide this as a service for our execution.

Optimus seeks to provide these to traders that also feel comfortable that someone with experience is running their system. An NDA is not worth anything only if the parties don't respect the terms, but we always recspected other people's intellectual property.

[lylec305]

Originally Posted by feng2088 »

Hello guys,

I designed a system using EasyLanguage that solely trades ES, however it makes less than 50 trades a year in average. I don't want to be charged for $100 platform fee and $20 data fee every month by tradestation. Are there any cheaper brokers out there that support easylanguage. If no, how do your guys manage to pay less fees and commissions? Do you convert your codes to other languages and use a different broker?

Also, I tried to make the test result look realistic..so I checked "Fill entire order when trade price exceeds limit price" (1st attachment), however I didn't apply any slippage to the market orders. If I put in $0.25, it will apply a tick of slippage to my limit orders as well which is not what I wanted. Does anyone have the same issue?

Thanks

What is your price fluctuation? I would be interesting to see your high to low price spread. A 10 year strategy is a long time to put money at risk...

[eqsys]

Originally Posted by feng2088 »

Also, I tried to make the test result look realistic..so I checked "Fill entire order when trade price exceeds limit price" (1st attachment), however I didn't apply any slippage to the market orders. If I put in $0.25, it will apply a tick of slippage to my limit orders as well which is not what I wanted. Does anyone have the same issue?

Thanks

Wait a BIG minute....your slippage calculation is completely wrong. I used to make the same mistake in TS before I realized my mistake.

In TS you can set the value for commissions and slippage in DOLLARS. With TS the commissions are $1.20 per contract, and my real time testing over many, many years and many, many contracts tells me to use the average bid-ask spread in DOLLARS as slippage.

For the ES the average bid-ask spread is .25. This is NOT 25 cents! This is $12.50. If you don't get this right your entire system calculations will be incorrect.

To correctly calculate the slippage you take the average bid-ask spread times the Big Point Value shown on the Symbol Properties page.

I have a TS Radar Screen workspace that continuously calculates the average bid-ask spread in points and dollars for a variety of futures contracts and another workspace for stocks. You absolutely have to get this number correct, otherwise all your system testing will be bogus...

One other thing, whenever I see these wonderful looking backtest results, they invariably leave out the drawdown stats. You have to have money management rules in place so that when you get drawdown you won't either go broke or get so scared and/or depressed that you stop trading. So, it's impossible for me to evaluate any system without knowing what level of drawdown to expect, and to adjust position size accordingly.

Let's say you don't want to lose more than 10% of your portfolio value when you get large drawdown. This means you have to allow 10x the max intraday drawdown in dollars to trade one contract. Then, you look at your monthly or annual gains and divide that by the 10x drawdown value, and you'll get an idea of the true profitability of the system.

Most novice traders don't do this, and most wipe out - it's almost guaranteed.

[fugu_master]

It's possible to run earlier versions of tradestation (e.g. 8.7. 8.8) and combine with OwnData with a feed from a broker. I personally use interactive brokers and take a data feed from the ASX (Australia) for swing trading options.

I also use IQ Feed for forex futures, which I also feed into TS.

I've looked at migrating to Multicharts too - and likely will do in the future.

[feng2088]

Originally Posted by eqsys »

Wait a BIG minute....your slippage calculation is completely wrong. I used to make the same mistake in TS before I realized my mistake.

In TS you can set the value for commissions and slippage in DOLLARS. With TS the commissions are $1.20 per contract, and my real time testing over many, many years and many, many contracts tells me to use the average bid-ask spread in DOLLARS as slippage.

For the ES the average bid-ask spread is .25. This is NOT 25 cents! This is $12.50. If you don't get this right your entire system calculations will be incorrect.

To correctly calculate the slippage you take the average bid-ask spread times the Big Point Value shown on the Symbol Properties page.

I have a TS Radar Screen workspace that continuously calculates the average bid-ask spread in points and dollars for a variety of futures contracts and another workspace for stocks. You absolutely have to get this number correct, otherwise all your system testing will be bogus...

One other thing, whenever I see these wonderful looking backtest results, they invariably leave out the drawdown stats. You have to have money management rules in place so that when you get drawdown you won't either go broke or get so scared and/or depressed that you stop trading. So, it's impossible for me to evaluate any system without knowing what level of drawdown to expect, and to adjust position size accordingly.

Let's say you don't want to lose more than 10% of your portfolio value when you get large drawdown. This means you have to allow 10x the max intraday drawdown in dollars to trade one contract. Then, you look at your monthly or annual gains and divide that by the 10x drawdown value, and you'll get an idea of the true profitability of the system.

Most novice traders don't do this, and most wipe out - it's almost guaranteed.

eqsys, TS reads this $0.25 as a tick instead of 25 cents. A tick = $12.5. When I put $0.25 in that box, it would actually reduce my profit by $25 or increase my loss by $25 ($12.5 * 2 = $25 buy and sell) It doesn't matter if I have limit orders. Thats why I said it's not accurate as it should only apply this to market order.

In regard to max drawdown, my # is a bit high since my initial investment is only $5000.
Max. Drawdown (Trade Close to Trade Close) = -$1,051.66.16 or 21.03% of my initial 5k captial.

Max. Drawdown (Intra-day Peak to Valley) = -$2,167.22 or 43.34% of my initial 5k captial. It happened on 11/7/08, however the acct had a 39k total net profit as of that day.

I will definitely start with more than 5k real money, so the number will look better than this if I would be getting the same performance. I do respect risk and I use stoploss.

Thank you all!

[eqsys]

Yes, the min move is .25 of $50, or $12.50. In TS there is a box where you can put in Position Slippage. This is in $, not points or ticks. If you put .25 into the Position Slippage box you get 25 cents slippage. You have to put $12.50 into this box in order to get a correct calculation.

I have traded live futures with both limit and market orders. My experience has been that limit orders work most of the time, but when a market really explodes, you don't get filled with a limit order. So, I eventually I went back to market orders. And, market orders = slippage, and that slippage can really add up.

If you only have 1k in drawdown this is a really good system. Just trade it in real time with real money for a few months and see how it does. I have come up with exceptional systems like this but it turned out they didn't work in real time trading. There was usually a reason - I overlooked something, used alternative chart types that didn't work in real time, etc. Perhaps your system will really work.

Best of luck to you...

[eqsys]

to feng2088....

Here's how I calculate the profitability of my systems, and I will use the figures you gave in the thumbnails as examples.

Your average monthly gain over the past 12 months was $685.19. Your max drawdown was $-2,167.22.

I can't take more than a 10% drawdown, so I multiply the 2167.22 by 10 and that's how much I want in my account to trade 1 contract.

So, 685.19/21672.20 = +3.2% per month. You may be comfortable with a 20-30% drawdown of your account equity, but I am not. And, when you are trading bigger accounts, 20-30% drawdown is a LOT of money. This is especially true when you are trading your own hard earned money.

I evaluate all of my systems this way. In order to make 100% per year you need to make +5.9% per month compound. My goal in system testing is to make 100% per year in backtesting. I know that in real time trading I'll only make a portion of this.

By the way, how many optimizable variables does your system have? I am only comfortable with systems that have 1 or 2 variables. More than that and I have found that the systems backtest really well, but fail in real time trading, because they are curve fit.