Backtesting Results Match Real-time History, but...

[GratefulJohnny]

We have been trading real-time the same system since mid 2000. It is a fairly simple trend following system that trades breakouts for each of 35 markets. Exits are also based on highs or lows a number of days back. Position size is calculated using a simple formula based on volatility and is consistent for each contract. The hard stop is also consistent for each market and based on initial volatility.

We have also written a backtester in visual basic that simulates our trades since 1984 using continuously backadjusted futures contracts from CSI. Our back tested results closely match our real returns, but previous to this the results are too unbelievable, very large returns with minimal drawdown. We have exhaustively combed through the calculations and trades and have found no errors. We have also tried to use a minimum volatility measure so our position size isn’t unrealistic. We have also run the traditional turtle parameters of initing on a 50-day breakout and exiting on a 10-day channel trailing stop. These returns are also extravagant without much giveback.

We are looking for any ideas to poke holes in our methodology or rhetoric to defend it. I have attached these results.
Thank you very much!

 

[TheBramble]

GJ - you've certainly come to the right place for major hole-poking! It's almost an art form here.

However, some folk may ask what your methodology comprises in detail before they assume the effort of assisting you in breaking it. You may well find, and I wouldn't blame you, that you are unwilling to give us much detail as some would suggest would be required for this sort of exercise.

Having said that, if it works, it works! Trade it.

BTW - welcome to t2w.

 

[rols]

Be grateful Johnny!

We have been trading real-time the same system since mid 2000. It is a fairly simple trend following system that trades breakouts for each of 35 markets. Exits are also based on highs or lows a number of days back. Position size is calculated using a simple formula based on volatility and is consistent for each contract. The hard stop is also consistent for each market and based on initial volatility.

We have also written a backtester in visual basic that simulates our trades since 1984 using continuously backadjusted futures contracts from CSI. Our back tested results closely match our real returns, but previous to this the results are too unbelievable, very large returns with minimal drawdown. We have exhaustively combed through the calculations and trades and have found no errors. We have also tried to use a minimum volatility measure so our position size isn’t unrealistic. We have also run the traditional turtle parameters of initing on a 50-day breakout and exiting on a 10-day channel trailing stop. These returns are also extravagant without much giveback.

We are looking for any ideas to poke holes in our methodology or rhetoric to defend it. I have attached these results.
Thank you very much!

Hi Johnny

Not quite sure what you are looking for? As Brambles says. Don't fix what ain't broke! If it's trading with real money in real time and making real profits then I'm real pleased for you.
......unless you're selling something?

 

[GratefulJohnny]

Not selling anything. Just trying to save an account. Our only account.
I have confidence in what I've done. My problem is that our investor doesn't have the confidence in our backtest results. I'm using this backtester to come up with a profit taking system. I'm really stuck in the mud right now. He's threatened to pull the money from the account which would close us down. We can't move forward on anything because he doesn't believe the results. Of course anything that we would move forward with would be even better than what he already considers unreasonable.

 

[GratefulJohnny]

The problem is that there is no drawdown after huge returns in the backtest prior to the actual trading.

 

[fxmarkets]

Threatened? Tell the client to take his/her business elsewhere, If you know , then you know.... some people just want need other things.

Best with it Johnny.

 

[rols]

The problem is that there is no drawdown after huge returns in the backtest prior to the actual trading.

check your PM inbox

 

[GratefulJohnny]

We've ran the backtester real time alongside our actual trading and they match up. Like I said the results of the backtester nearly match our actual trading results. We've gone in and manually checked every calculation that is made.

 

[GratefulJohnny]

We want to enhance the system and develop new ones. If these backtest results aren't believable we can't convince him of the validity of any of our research

 

[GratefulJohnny]

He's been trading with us the whole time, but now he's wondering if we've all been lucky all this time

 

[TWI]

This guy is the wrong investor for you. Send me your monthly broker statments real-time and I will be able to help you get funding. This is ONLY if you have a real-time, real-money track record.
What markets do you trade?
Other point I find confusing is that if you have been doing this for 6 years and the results are so stellar then why are you still living off your investor?

 

[GratefulJohnny]

This guy is the wrong investor for you. Send me your monthly broker statments real-time and I will be able to help you get funding. This is ONLY if you have a real-time, real-money track record.
What markets do you trade?
Other point I find confusing is that if you have been doing this for 6 years and the results are so stellar then why are you still living off your investor?

Thanks. You have a pm.
Real time performance attached.

 

[Tuffty]

Your recent performance would appear to be normal among the class of long term trend followers:

www.futuresmag.com/library/2006/02/01_0206cstradingtechpruitt.pdf

Maybe a reasonable period of draw down will shake the loose holders of trend following systems out, and their out performance will continue again.......or maybe not! Looks like you may be part of that process. I would think the huge amount of liquidity going into hedge funds in the last few years has something to do with the recent poor performance.