They know something...I can Feel it. Let's get in on it." ...reminds me of trading gurus and the sheep masses. Or any euphoria bubble. Herd psychology. The house always wins in the end, they have no skin in the game. Just letting all the degenerate gamblers wage their bets. But I like Louis Winthorpe's calmness amidst the turmoil. He has a good feel for the market, let things settle and fizzle out and reveal themselves the ideal opportunity to take action. Be a Sniper, on Stakeout...waiting patiently, calmly for the right moment. -- Don't be a Black Friday shopper bursting through the door trader. It's important to understand emotions and volatility within a fixed environment blended in with reasonable expectations and market facts.
Hello MidwesternTrader, Very good post you have here. I am also learning systematic trading and developing/programming strategies for Intraday trading only. I agree with you on this, one system will not work for long time. For stop loss placement, I am also experiment using a ATR of 10 bars during Intraday trading. For your case, you can experiment with daily ATR. Good posting, I look forward to your journal. Thanks for sharing.
This looks like a great journal. I hope that you continue posting. Last year I did something similar. I traded GC, CL and LE concurrently using an automated strategy. I didn't blow up but that account ended down ~10%, mainly because GC and CL both had a long sideways phase in the summer/fall that chopped up my trend-following system. It's funny because not long after I turned my system off, GC and CL both started making decent trends again.
Hello MidwesternTrader, Are you using the same type of trading system on all 8 future markets? Thanks
No, I am not running the same exact signaling methodology on all 8 instruments. I have 4 variants that I use. Although each of the 4 variants is based in my core belief system about how markets move, how to identify trend continuation, and how to identify possible trend reversal. They are not radically different in how they predict the next trade I should make. I run back testing on all 4 variants for each instrument, then select the variant that has the best performance history to use for live trading. Sometimes multiple variants will trade about the same for a given instrument, so I just pick one to use. Other times there appears to be a clear best variant to use on an instrument. I run all variants against each instrument daily to track their continued performance as to which one remains top dog for both 2 year and 6 month time frames. I believe different variants work for different instruments because not all markets have the same texture or personality. Each has its own unique characteristics that influence how it moves. CL and GC are global commodities, trading nearly 24 x 7 (1 hour break) NG is a global commodity, but most trading for NG is in the North American market hours HO is mostly a North American usage of pertoluem and traded almost exclusively in the North American market hours. But it is hyper correlated to CL because CL is used to make HO. So it will track CL outside of US hours too. KC is another global product, but has reduced trading hours than CL, NG, GC HE, LE are global commodities too, but primarily North American based instruments that only trade 8:30 AM - 13:05 PM CST. This market also does not use a premarket where trades actually occur, only bid / ask price jockeying to set the opening 8:30 print. So because of the bullets above (and I'm sure plenty of other reasons I don't personally understand), I do not believe you can use the exact same thing to be successful across all futures instruments.
Thank you MidwesternTrader for the well written response. For your intra day strategies, how many year(s) or trades do you back test for? I ask because I develop intra day strategies as well. Thanks
CL, NQ and GC were back tested 3 years: 2015 - 2017. 2 years for NG, HO, KC, LE and HE 2016 - 2017. I've found one benefit of the journal is trying to articulate ideas and concepts for others also helps me more fully understand what I'm trying to achieve with each nuance.
Thanks MidwesternTrader for replying. I agree with your back testing sample years as mine is only 1-2 trades depending on number of trades. Yes, everyone here is helpful and lookforward to your journal results. once i find an edge(s) and go live, I will start a journal as well. The journal helps for documentation and extra help. Did you perform any walk forward testing as well?