Holy Grail?

[skanker]

Did I find the Holy Grail of trading? I backtested my system in Amibroker (past 10 years) and I would have been able net over $29 million dollars had I started out with $50k. Actually, I can get close to $1 billion if I don't restrict my position size per trade. This is an EOD system where I buy/short and sell/cover at the next day's open. 59% win rate and but mostly wins because it catches major trends and stops out if only trading in a range.

Should I start my year-round vacation plans or am I being a tad naive? You tell me.

 

[jonty123]

Coulda Woulda Shoulda, it's a whole new ball game once you put real money on the line.

You cant backtest greed, fear, emotions.

You need to start off trading a small account over a decent period of time, and then slowly add to it and increase stake size if you are achieving similar results to your back testing.

 

[skanker]

this is an EOD system. I get about 5-10 signals per day telling me to get in or get out. I might be anxious but it's not like I am trading intraday. FYI, I can never do intraday trading. I just don't have the makeup. But considering your response...So you're saying there's a chance that my system may be somewhat profitable?

 

[N Rothschild]

i wouldnt pay to much attention to back testing tbh. start trading it live with a small amount of money and see how it goes..

 

[neil]

Use a small amount of real money to experience the emotion of putting money on the line

 

[trendie]

skanker,

10 years seems pretty comprehensive.
50K into 29Mill?
So, if you seeded an account with a modest 5K, you anticipate 2.9Mill?? worth a pop.

have you done the usual robustness checks such as applying slippage tests, etc?

PS: how many PMs have you received so far?

 

[skanker]

Well, that was one of my issues. I need to make this system and my backtesting settings more realistic. How do I account for slippage? What other settings do I need to consider in order to make this a more realistic backtest? Thanks for everyone's replies.

 

[skanker]

Also, this system was optimized for the stocks on the S&P 500. So for S&P 500 stocks, I am able to be almost $30 million. But this has worked on different markets. On full NASDAQ list, I am still able to get into the high $20 millions. I am working on testing this on the full list of NYSE but I can't seem to find a comprehensive ticker list in comma delimited form.

 

[skanker]

This system works in bull and bear markets. Generally looks for bearish trends in bear markets and bullish trends in bull markets. I am also looking to change my position size of each buy and trade signal depending on the overall market condition. However, I haven't been able to figure out how to do that with amibroker.

 

[Lord Flasheart]

Hi Skank. Why dont you reveal you system and we can analys it,there may be something your missing. You can take it from me that even if it is a great system is does not mean it can easily be copied once emotion and everything else gets involved, so you have nothing to lose. Some of the best and most workable systems that are available here and elsewhere are "improved" as soon as they are looked at and turned into losing systems.

 

[astoria]

Also, this system was optimized for the stocks on the S&P 500. So for S&P 500 stocks, I am able to be almost $30 million. But this has worked on different markets. On full NASDAQ list, I am still able to get into the high $20 millions. I am working on testing this on the full list of NYSE but I can't seem to find a comprehensive ticker list in comma delimited form.

Uh huh..when you anticipate marketting it/

Need to do drawdown analysis...sharpe ratio and average profit/loss per trade..what were your conseq. losses...profits don't matter...

that is a start...good luck

Astoria

 

[skanker]

Julian,

As you might understand, I am quite skeptical of revealing my system. It isn't the most complicated thing. Actually, it only has a few rules. But I would like to keep it to myself...for now.

 

[skanker]

Can someone analyze this?

Avg. Profit/Loss 4435.91

Winners 4313 (60.87 %)
Avg. Profit 12209.15
Avg. Profit % 6.48 %
Largest win 923883.76
Max. Consecutive 34
Largest win 923883.76

Losers 2773 (39.13 %)
Avg. Loss -7654.25
Avg. Loss % -4.21 %
Max. Consecutive 18
Largest loss -644048.34

Max. trade drawdown -644048.34
Recovery Factor 20.07
Profit Factor 2.48

 

[Lord Flasheart]

How is someone supposed to analyze it without the proper facts. You mentioned about slippage,depending on what it is, it is possible that all of your profits will dissapear when slippage is included. As its an EOD system have gaps been accounted for etc,etc. Theres so much someone would need to know. You are never short of genuine help on this site, but do you really think you have discovered the holy grail? Be realistic and give someone the tools to help you.
I am not talking about myself, all my trades are descretionary and I dont have the time. Good luck

 

[skanker]

No, I don't think I have found the holy grail. I am actually trying to prove that it is not because it is so unbelieveable. That said, I can't give away my system in case I got really really lucky. If you tell me what you need to know, then I can try to provide that information to help you evaluate it.

How about this first question? What do you do to account for slippage in an EOD trading system?

 

[skanker]

If I say that I get charged 5 cent commission per share (to account for slippage), I net about $23 million over 10 years. But I don't know if that is a realistic slippage amount. Currently, I pay .0035 per share for commission.

 

[episteme]

What timeframe data are you testing this against? I really doubt that this is for real. My guess is that you are opening trades based on some post hoc information in the data that has crept in due to the testing method. Surprisingly easy to do.

 

[skanker]

I generally test using 1/1/2000 to 1/1/2010 EOD data from Yahoo via Amibroker. My system revolves around MA. I just see how the different MAs are in relation to the close. I don't have anything other than that and trade/position size/stop settings. Can you give me an example of post hoc information that could corrupt my backtest?

 

[skanker]

Hey, I doubt that this is for real too. I want to prove it out.

 

[skanker]

If I only buy and sell on the open, would it be too much to just limit my position size to 10% of average daily volume? Should it be set lower than that? I intend to use it on MOO orders only.