What an impressing journal you have got here! I decided to grab the results posted here and run some statistical analysis on them. I will share the results here:
Number of trades: 119
Mean points per trade: +4.3 Standard deviation of trade: 32.7 points
Average winning trade: 31.6 points
Average losing trade: -21.6 points
Winning percentage: 47.9%
Risk/reward ratio: 1.47:1
I assume about 240 trades per year in average (you had skipped some days due to volatility conditions).
I have also attached the picture of monte carlo analysis here. Monte carlo simulation is ran with 100000 iterations.
Blue line is showing average result, green lines are 95% confidence intervals, red line is the worst result in 100k trials and black line is a random result. The lines represent arithmetic returns (you are not cumulatively increasing lot sizes as your capital increases).
Monte carlo results yielded also chance of positive score: 92.2%
Overall very good results so far and I will be very interested to see how this develops. I'm impressed by the way how you declare before actual trading what you will do during the day (entry time, exit time and direction). I haven't seen anyone doing this before. Keep up the good work!
Edit: Added handwritten data (GFIs1_first_119_trades.txt)
@GFIs1 if you can supply me the data from the beginning of this trading system (I couldn't find your twitter account) I could make the same analysis with bigger sample size that would yield more accurate statistics.
Last edited by rounder8; August 31st, 2012 at 01:50 AM.
The following 5 users say Thank You to rounder8 for this post:
Thanks @rounder8 for the statistic and the nice comments.
Started publicly with the system on december 19 last year.
There is more data (plus 12 months more). The results can not
be taken for a statistic as the rule set was developed in that
period. Will send you all the twittered results though.
You can find the messages on twitter: https://twitter.com/gfis1
or search for gfis1.
Since when do you consider your system to be fully developed to the form it is now? Is it automated system or are there some discretionary elements? Earlier in this journal you told that you have your results on a spreadsheet, it would be easier for me to calculate the statistics if I could have that resource so I don't have to manually collect the nuggets of information from twitter feed, in case you are curious to see...
The following user says Thank You to rounder8 for this post:
The system is final since the beginning of publicly announcing the trades. It is automated - but not programmed
to open a trade. The closing / and stop is fully automated. The ruleset is quite difficult to program as it includes
situations like holidays - important numbers on a day etc.. So these are not "discretionary" elements - but these
elements do influence time of starting and/or closing time of each trade. Unfortunately this is not easily to be
transcripted into a full auto trading system.
Hope this helps :-)