Rol's Trading Journal

Excel 2007 Power Programming by Walkenback & Mr.Excel forums...all you ever need. The help you get there is amazing. Stuff I would pay for is literally free.

My universe is currently 1000 names, and all that is DDE links and VBA scripts, then sent direct to the platform. If you have a near top of the line computer like I do, it can handle that many with out a problem. I am not sure about 1400, as I havent tried. I have a AMD Phemon 6 core 1090 with 8 gigs. So I suppose if you have a better computer it should be able to handle your whole universe with out any problems.
 
Realized good steady gains for the week, but the strat took losses on about 4 stocks today (friday), which were underperforming. While it always hurts to take significant losses (-593.33 today), overall the numbers still look good. Profit Factor >2, and % profitable > 66%. Avg. Winning Trade < Avg. Losing Trade, but this is due to larger losses on a relatively smaller # of stocks. I hope to maintain an APR of > 100%, but I don't have a fixed #, because I expect that # to be variable. It is interesting that the Buy & Hold return is only 0.05%. This number represents the % return of holding all the stocks for the entire trading period as a comparison to the strategy return. It feels good parting with the losers anyway, because now I can reload for next week. :) My "edge" did not materialize with them and by holding them outside the strat rules, I would be putting myself in "no man's land".

I think I will narrow my list of stocks down to 1000 and increase my position size slightly to make up for the fewer signals I may get. I will put the 1000 symbols in the RadarScreen and let the strategy go to work. This way I won't need to perform nightly scans, and will just monitor the 1000 symbols in real-time. This will make everything even more hands off.

Total Net Profit $2,936.22
(Per Share) $0.19
Gross Profit $5,486.55
Gross Loss ($2,550.32)
Profit Factor 2.15
Total Number of Trades 134
Percent Profitable 74.63%
Winning Trades 100
Losing Trades 34
Avg. Trade Net Profit $21.91
Avg. Winning Trade $54.87
Avg. Losing Trade ($75.01)
Largest Winning Trade $199.96
Largest Losing Trade ($369.89)
Max. Consecutive Winning Trades 15
Max. Consecutive Losing Trades 4
Total Shares/Contracts Held 15260
Total Commission $351.38
Return on Initial Capital 5.87%
Annual Rate of Return 114.08%
Buy & Hold Return 0.05%
Trading Period 18 Dys, 6 Hrs, 28 Mins
Percent of Time in the Market 78.78%

attachment.php
 

Attachments

" My strat does not look at support or resistance levels, or moving averages, or volume other than to screen out low volume stocks. The couple of indicators I use are not used in the typical ways. It seems to come down to an edge I have discovered that backtests well or even better on out of sample data on a portfolio of about 1400 stocks (no futures, currencies, options, etc.). My theory was that if it worked on around 68% of the stocks out there, why not minimize individual stock exposure by diversifying into multiple symbols. "

I have been following your post with great interest. My compliments on a great equity curve.

Is symbol selection based on just price action?

Can you recommend a favorite book that you found most helpful in developing your trading expertise.

Thanks
 
Quote from Camdo:

" I have been following your post with great interest. My compliments on a great equity curve.

Is symbol selection based on just price action?

Can you recommend a favorite book that you found most helpful in developing your trading expertise?

Thanks
More...

I think one's equity curve should be the definitive gauge of one’s trading prowess. In a sense, all indicators are based on price action if they use price in their calculations. I just think people get bogged down with too many indicators, hoping they will reveal a magic winning trade. I get dizzy when I see, for example, charts that include Bollinger bands, multiple moving averages, support and resistance lines and candles. I like just one or two indicators below the chart so I can compare them to what the o/h/l/c is doing on a daily basis.

My entries are made by looking at daily bars, and seeing what the price has been doing for several days prior. There is just too much noise in intraday charts for me to see a setup occur. Also, it is difficult to back test a portfolio using intraday data over a period of years. I like to back test over years to try and capture all the various ways stocks behave during diverse market conditions. Another thing is the setups I look for need time to develop and play out, and this usually takes several days.

I don’t have any books to recommend and have never considered myself an expert. Maybe when my equity hits 1 mil. I will consider myself an expert. :D Most of what I have learned has been on my own, by observing price changes on daily charts. I didn’t even know what swing trading or rtm trading was in my early years of trading. I think “buying the dips” best describes what I do. I would suggest to someone in order to get a feel for this type of setup is to identify days when a stock price rose significantly and then look at what the price was doing the few days before that, and seeing if any patterns emerge. Do this for a stock over a 10 year period, and then do this for 9999 more stocks, and it will become ingrained in your subconscious as to what a good setup looks like. Come up with indicators, optimizing various parameters through back testing, which will alert you to these setups. And then automate the strategy.
 
Quote from Rol:


Let me just say that if you trade futures or forex, don't use TradeStation.
More...

Have you used any other platforms other than Tradestation? Specifically for futures autotrading, do you have any advice?

I like the journal, by the way. Thanks for sharing.
 
Quote from benwm:

Have you used any other platforms other than Tradestation? Specifically for futures autotrading, do you have any advice?

I like the journal, by the way. Thanks for sharing.
More...

TS is the only platform I have used for autotrading. I have sort of locked myself in with it, as I have only just recently begun looking at the other brokerage, api, and coding language options. Actually, from what I have been reading about it in these forums, what I have is not so bad after all.

My focus has been quite narrow, but it has allowed me, I think, to nearly perfect my strat using TS RadarScreen. I know nothing about backtesting of day trading strats, and don't even really want to, because with day trading I feel you have to be monitoring your system much closer, and will often be complaining about bad fills and data issues. In TS sometimes my intraday charts will lock up, but RadarScreen just keeps humming along. I can imagine what would happen if I actually had autotrading applied to them.:mad:

My strat works very well with futures, it just does not produce many signals. There were only 28 over the past 10 years for the Eminis.
 
Is this a long-only strategy? Sorry I wasn't so clear on that from your posts. For example, if the market tanked 20% would you be getting net short along the way? Or would you just get less entries and expect to make less money than normal?

Out of interest, how did the strategy backtest during the May 2010 flash crash, or even during 2008-Spring 2009 bottom?

I agree with the others that this is a really high quality journal. I don't trade stocks but the thought process is really impressive.

Were you influenced by any of the other posters or threads on ET?
 
Quote from benwm:
Is this a long-only strategy? Sorry I wasn't so clear on that from your posts. For example, if the market tanked 20% would you be getting net short along the way? Or would you just get less entries and expect to make less money than normal?
More...
benwm,
Yes this is a long only strategy for a few reasons. When I applied my rules to the short side the strat did poorly. Also, I need the 2X buying power, and when selling a stock short you need 100% cash, (at least with my broker), and this would totally mess up the position sizing calculations of the strat. There is no shortage of signals when watching 1000+ symbols, so I have good market exposure in up, down, or sideways markets. The strat does not attempt to predict market direction so I wouldn't know when I should be getting net short anyway (today, would have been a good day with the S&P down the most in 6 months). I view overall market direction as nothing more than the sum of what all the individual stocks turn out to be doing that day. If advancers = decliners then the market may look as if it went nowhere, but I may be having a good day anyway, because I happened to be holding more of the advancers. I have rather stopped caring about market direction, as well as individual stock direction. The strat will follow it's instructions, regardless of market direction. It will be trading the same way whether the DOW is at 10k, 20k, or anywhere in between. My time horizon is only a few days. If I turned my entries off, and just had the strat monitor the exits, I would be 100% cash in a few days.
Quote from benwm:
Out of interest, how did the strategy backtest during the May 2010 flash crash, or even during 2008-Spring 2009 bottom?
More...
Here are some charts and stats from Amibroker of how my strat performed during those periods. This is the equity curve from the May 08 highs where the E-minis just kissed the 200 dma, to the March 06, 2009 low. The starting equity was $50,000. There are some considerable drawdowns, but it is during a bear market. You can see that the overall trendline is still intact.
attachment.php


Quote from benwm:
Were you influenced by any of the other posters or threads on ET?
More...
I joined ET a couple of years ago, but never read it much, being immersed in my strategy development. I'm not sure why I came back recently, but I became interested in the journals and began reading through lescors journal. Many things he said and was doing resonated with me, such as risk management and having rules in place. I then decided I would start my own journal to record this experiment I am conducting.
 

Attachments

You must log in or register to reply here.
Back
Top