cunparis

I thought I answered it, indirectly.. in that I gave up the automation and learned to trade manually. So I stopped doing analysis. To me it made sense to learn how to trade manually before learning how to automate. Before I was shooting in the dark and curve fitting and then hoping it'd work real time. it didn't. I don't have the time or desire to have a lengthy debate about anything I said as I simply don't care enough and I'm too busy. I've settled into my routine, I come home from work, make my levels, set my audio alerts, and configure my automated strategies and I go have dinner and spend time with the family. When the audio alert goes off I come to my PC and look to time an entry for my trade. I'm hoping to automate this last step so that I can avoid missing trades (not hearing audio alert) and avoid waiting for the timing trigger at my PC. I take 0-2 trades per session, and I'm done. This works for me. In Oct - Dec of 2012 I averaged 1 pt per day traded (for 1 lot), but I didn't take trades every day. That was better than I did trading full time. In January I traded full time and was breakeven. so now I'm back to what I did in 2012 and will be happy with that. Lots of people claim to be profitable, and I'm not saying you're not. I just quit listening to all that. I quit forums totally actually, but once in a while I get an email about an update in a thread from a long time ago.. I don't expect anyone to believe me, but I sleep better knowing I've recommended people trade on simulator until profitable 6 months (consecutive). I never said that one would be profitable trading real money after that 6 months, but it will at least prevent people from blindly throwing their money away. If anyone wants to read more about my story, you can read a lot including real trades and real performance results that showed the ups & downs of my trading over the past several years, on my blog. I wish everyone the best of luck in whatever they do and good trading. Cheers.

I think it's a very realistic view, and the very few profitable traders I have known have agreed with it. I've learned to trade profitably (not insanely so, but positive) manually using my brain. Modeling everything I do into an automated strategy would be way too complex. There are so many inputs and factors and they're constantly changing. And if it could be done, then one could slowly work up to 1000 contracts and make millions. Wouldn't every fund in the world be working on such a thing? But I don't want to argue this point because people coming into trading are chasing their dream and they don't want to believe it's not possible. So when someone tells them the odds of succeeding are 1 in 10,000 they either believe that's not true or they believe they'll be the one who will make it. In my case I went back to consulting which I was doing before trading full time. And my trading improved. So now I get a steady paycheck and I focus on getting one good trade in the last 2-3 hours of the US session. I'm kind of rambling. My main advice for someone who doesn't believe what I say is this: Trade on sim until consistently profitable for 6 months. That will remove the possibility of financial losses. The remaining risk will be time loss. There is no easy way to remove that one. When a strategy (manual or automated) works and you're sure it works, run it real time on sim for 6 months to make sure it doesn't stop working. I'm sure you'll find it does stop working but at least you won't lose any money finding that out.

It didn't work. :evil tongue: The answer is yes optimizing profit target & stop loss is curve fitting, at least to an extent. If the stop and target are just random numbers that are optimized then it's curve fitting. I'm now working on a semi-automated strategy. I put in my levels and the basic idea and my strategy picks entries & stop loss. Since it uses my levels which are done manually, there is no backtesting and so no optimization. I've only been doing it for a few weeks and I'm working out little bugs but so far I'm encouraged. Market action wasn't great the past few weeks and I only run it the last 3 hours of the trading session so it'll take a while longer to get enough data. So if anyone is interested in automation.. I do not believe it's possible. The markets change. high volatility, low volatility, trend days, balanced days, etc. I've never seen anyone have a successful automated strategy long term, at least without massive drawdowns that would cause most to pull the plug. I think the best route is to give up trading, but if that's not an option, it's best to learn to trade manually and then look to automate parts of the trading slowly.

I no longer have TS so I don't know, but if 9.2 has second or greater precision on the timestamps (TS 8 had 1 minute precision) then it should be possible. :helloooo:

Well you know what they say.. Try it out and let us know.

I have prototyped bundling trades and to be honest I don't think it's that useful. The problem is one has to make a lot of assumptions that are often incorrect. Just because two trades occur at the same time doesn't mean it was the same trader who made them. That sort of thing. I'm curious to know how others have handled this. The programming is not hard but it seems we don't have enough information to work with.

I looked at your website but it wasn't clear what it actually does (even if you wish to keep the algorithms secret). I guess it's finding large disguised orders? If you have more examples, a video showing it work realtime, or a trial, I would be very interested.

To update the thread I have recently started increasing my size. I now trade 2-4 contracts on bund & ES. Once this is more stable then I will increase it slowly (after each month of good results).

My hosting provider had a major hardware failure and my site has been down for 3 days now. It should be up again today. It sucks but that's what one gets when one pays $6/month for hosting a website. Also I must say that I've changed my use of stops a bit.. I don't use tight stops any more. I don't use hard stops actually, I just get out if it's not working. This has taken a lot of experience but I'm doing much better this way. I think tight stops are better if one can be profitable with it, but everyone has to find out what works for them.

Can you explain 90% mechanical? I've been doing programming that generates visual & audio alerts and then I decide to place a trade. I know another trader who decides when to potentially enter and he clicks and then his software will enter when certain criteria are met and will exit with other criteria. I think both examples are interesting and something I'm currently researching.

This is the trade-off with automation. It can't be as optimal as an experienced professional trader. For me, trailing stops give up too much profit and profit targets are often missed by a tick or two. That's why, for me, it's important to be able to read the order flow and adapt. If I get long I take my first scale when I see responsive selling come in. Then for my other units I watch the order flow as we approach my targets and if I see strong selling coming in a few ticks from my target I'm not going to risk giving back all that profit just to try and get 2 more ticks. It's a poor R:R to do that. So I will exit a few ticks before my target. It took me a while to learn to do this, but after many targets being missed by a few ticks I started figuring it out. Since most edges are very small (mine has been 2 ticks over the last 4-6 months) getting an extra tick on a target has a huge impact. This is just one reason, IMHO, that mechanical trading will be sub-optimal. Some traders may prefer to live with a sub-optimal method because they feel they can never acquire the skills to trade manually with discretion. This is how I felt for a long time but I realized that I was circumventing the problem. Best to learn how to trade properly and then I don't need automation. This year I made a lot of changes to how I trade and I'm still working them out but for most of last year I had a profit factor of 2.0. I never could make an automated strategy that worked long term, especially not with a PF of 2.0. So if I'm performing better than my automated strategies it makes sense for me to continue developing my skills. Just a point of view of someone who's tried it and chosen the path of learning to trade with a little discretion. I have my game plan before the open each day. I know what I will do, where I will trade, etc. But when it comes to timing entries or passing on a setup, and managing my trades, that's where the discretion comes in. If someone can make an automated system then that's great and I'm happy for you. The risk is that someone will spend 2 years trying and have nothing to show for it. But that's the risk with learning to trade too. It's the risk inherent in any worthwhile adventure. And that's why only a few succeed. So even if it's possible and you see someone with proof they're doing it, that doesn't mean you can do it. The odds are stacked against you with a very slim chance of success. This is not meant to discourage people but rather to show some realism. if you want to do an automated system you have to accept the fact that 99% don't work and that it can take years to find out. If one accepts these odds and gives it a shot then I sincerely wish you good luck with your endeavor.

I'm not, that was the winner of the automated trading championship!!!

This pretty much proves my point that it's not possible. Here is my take on the winner: The Profit Factor is 1.15. There is no room for error. Commissions, slippage, market change, anything basically could make this go negative (which is already what happened if you look at the equity curve). It has 50% drawdown. If it continued it'd probably give back most of those gains. These systems have very short life-spans. If you're lucky they can do well in the beginning and then slowly stop working. Who can sit through a 50% drawdown to find out if it's just a drawdown or if it will never work again? I can't. I've had systems that did great like this but they all stopped eventually. Some after a few months, some after a few weeks, and one after a few years. The one that lasted a few years went into drawdown so I stopped and then it came back so i started again and then it went into drawdown so I didn't stop and guess what? Never worked again. I know most will keep searching for the holy grail so I don't want to be a negative nancy and will leave it at that. Good luck.

I gave up on that a while ago. Every once in a while I come up with some new idea and I test the heck out of it and of course it is not profitable. The problem is context. Context is everything. As an example: A system could signal a long but if we're a few ticks below yesterday's high then it's likely not to work. Context cannot be automated. It takes a human to analyze the context because each situation is unique. So for the past year I've been focused on learning to analyze the markets. Supply & demand, support & resistance, market profile, etc. I've come a long way towards understanding the context. However profiting from it is not easy. It takes years of practice. IMHO your time would be more productive learning about context and forgetting about anything mechanical. Count on it taking several years. If you're not prepared to put in several years then best to quit now.

I find this perspective really interesting. I recently started focusing on 30min bars and I use 1min bars to pinpoint entries and my trading improved. What you say about 1min & looking inside with 5 second is how I feel about 30 min & 1min. Just goes to show how the markets are fractal and how everyone has to find what works for them. Thanks for sharing.