November has finally ended and I have been demo trading as per my previous posts with the express aim of addressing the specific weakness in trade selection identified at conclusion of posted results in October.
Summary
1) I am now using a very simple chart pattern and HGI signal to enter trades.
2) Several other methods for entry have been suggested on this thread and in my inbox. I have started to look at those and I am already prepared to discard one of them. Details in the next post.
Note please review these trade entry methods for yourself, if you are looking for a trade selection method. They may fit your trading style better than mine !
3) The trade selection method I am using produces far fewer trades than was the case when GTWV was “always in.”
The noteworthy thing is the fewer trades are producing better results !
4) The main problem identified at end of October was many trades were failing to become GT’s . My adapted trade entry method “seems” to have rectified this and
the win rate for seeds is much higher.
5) The account I am trading is the same one in use at end of October. This is important for context purposes. When I went offline I started throwing experimental trade selection methods at the account to see what was working and what wasn’t. By the time I settled on the method I am now using for trade selection I had input a lot of poor trades. The fact that the account is not only alive but doing well is testament to the viability of the new trade entry method in use.
In this post I will write up the test results for November. For readability purposes I will explain the chart pattern and signals I have been using for entry in the next post.
Results GTWV on H4 - EOM 30th November 2017
A :
Cumulative results (From start of open demo account 2/10/17 to End October - initial balance 500 USD)
This week cumulative
Expectancy per dollar traded =
-13.4%,
Expectancy per trade
-$0.23
Dataset 92 trades
Absolute
gain /
loss -$21.55 dollars i.e -4.31% of initial account
B :
Results (From start of open demo account 2/10/17 to 10th November 2017 - initial balance 500 USD)
This week cumulative
:
:
Expectancy per dollar traded =
-24.1%,
Expectancy per trade
-$0.53
Dataset 120 trades
Absolute
gain /
loss -$63.69 dollars i.e -12.74% of initial account
C :
Results (From 12th November 2017 to end November - initial balance 500 USD)
This week cumulative
Expectancy per dollar traded =
+69%,
Expectancy per trade
+$1.16
Dataset 36 trades
Absolute
gain /
loss +$41.66 dollars i.e +8.33% of initial account
D :
Results (From start of open demo account 2/10/17 to end November - initial balance 500 USD)
This week cumulative
Expectancy per dollar traded =
-9.45%,
Expectancy per trade
-$0.17
Dataset 186 trades
Absolute
gain /
loss -$22.03 dollars i.e -4.4% of initial account
Key :
A Shows the cumulative position at end October.
B Shows the results from start of account to after two weeks of experimenting with different entry methods.
C Shows the results for the 2 week period since I settled on a trade entry method that I will detail in the next post.
D Shows the cumulative position since start of demo account to date
Analysis :
Until end October we were only trading 6 pairs. In November, having recognized the limitations of the trade selection method then in use , we expanded the pool of pairs to 13, using the criteria detailed earlier in this thread, in order to find more trading opportunities using a more selective (hence limiting) trade entry method.
The results in November therefore come from a much expanded pool of pairs, but a also much reduced trading frequency because the new trade selection method is more selective.
Indeed during week commencing 20th November GTWV did not trade at all at H4 !!
Looking at the results in C, since I settled on the new selection method I note :
1)
+69% expectancy per dollar traded. That means for every dollar traded I could expect 69 cents back. Put in another way, the expectancy per trade was
+$1.16.
2) 24 of the 36 trades were seed trades.
3) Out of the 24 seeds, 16 were winners I.e +66.6%. This is a huge improvement from the +25% win rate in October!!
4) The average win in this period was
+$2.76. The average loss was
-$1.68
5) The overall win rate was 64% (Loss rate therefore 36%)
6) 36 trades in 3 weeks is an average of 12 trades per week or 2.5 trades per day. This is lower than the trading frequency recorded at end of October.
Looking at the results in D, which details results from opening of the account to date and includes trades using the new selection method I note :
1)
-9.45% expectancy per dollar traded. That means for every dollar traded I can expect a 10 cent loss. Put in another way, the expectancy per trade was
-$0.17. This overall expectancy into December is very encouraging given :
a) The large number of experimental trade entries in this dataset,
b) The poorly performing initial trade selection method until end October and
c) The small pool of pairs traded until end October.
2) The average win in this period was
+$2.34. The average loss was
-$1.84. This statistic is my favourite. If this extends into the future – the method will produce positive results as far as making money is concerned. That after all is the name of the game.
3) The overall win rate was 40% (Loss rate therefore 60%)
Conclusions :
Selective trade entry has increased the profitability of seeds
The method in use will be detailed in my next post
GTWV on a very small $500 USD account is performing well enough to continue as an experiment through December
The code for GTWV as indicated in my last post can now be simplified. I will be doing this in due course but this is not a hinderance to ongoing testing.
Charles