SPI Equity Curve 4/12/08

SPI closed equity curve attached, up to 4th Dec 2008

SPI Equity Curve 4/12/08

 

Closed Equity Chart 4/12/08

Closed Equity chart to 4/12/08

 

Closed Equity Chart 4/12/08

Closed Equity chart to 4/12/08

 

SPI Trades Update 04/12/2008

Well, I have been busy this last month with work, been to Newcastle & back for work to visit a coal mine in the hunter valley with my job as a contract process engineer. Carbon Credits and coal bed methane are part of the new environmental rage these days in Australia.

Update for last month is attached. Have to say one trade was a slight disaster & another sub-system has bit the dust, once again I relied on too few trades to accept the system & suffered for it. I actually hit a system stop at about 300 points & exited early, if I had not, I would have had double the loss.

I have to take back what I said in beginning about not having a stop, I don’t have a set stop per say, I use limit orders with a target which adjusts for various reasons to exit including an assessment of whether the price action indicates the trade is wrong. Sometimes this means the limit order gets me out next open. In addition I do have a disaster or system stop when evaluating the performance just in case the system fails (as in this case) - I look at the maximum expected drawdown from testing, if it approaches that, I check the system responsible & exit if required.

While the total return is still excellent, that drawdown is not! So I can admit a mistake, that system is in the bin. One bad trade really hurts the Win/Loss ratio. This also is what trading mean reversion can be like, a lot of good trades but an occasional outlier can bite.

One thing I have done is reduce risk even further, I now trade ASX SPI CFD for $1 Points while I am forward testing. These are not too bad but I do have to be at open once again as the ASX CFDs are a new market & liquidity is low, so limit orders at open do NOT work, the market maker just hits your order not the actual SPI open price.

Ticker	Trade	Date	Price	Ex. date	Ex. Price	Points	Comment
SPIM_CCB	Long	30/09/2008	4494	1/10/2008	4885	391
SPIM_CCB	Short	2/10/2008	4875	3/10/2008	4670	205
SPIM_CCB	Long	6/10/2008	4575	7/10/2008	4685	110
SPIM_CCB	Long	7/10/2008	4420	7/10/2008	4685	265
SPIM_CCB	Long	10/10/2008	4080	14/10/2008	4500	420
SPIM_CCB	Short	14/10/2008	4500	16/10/2008	4055	445
SPIM_CCB	Long	24/10/2008	3903	29/10/2008	3980	77
SPIM_CCB	Short	30/10/2008	3917	6/11/2008	4176	-259
SPIM_CCB	Short	3/11/2008	4121	6/11/2008	4176	-55
SPIM_CCB	Short	5/11/2008	4330	6/11/2008	4176	154
SPIM_CCB	Long	11/11/2008	3980	21/11/2008	3600	-380	System Stop
SPIM_CCB	Long	20/11/2008	3405	21/11/2008	3434	29
SPIM_CCB	Long	21/11/2008	3240	21/11/2008	3434	194

Total	1596
No Win	10
No Loss	3
Total	13
Win %	77%
Avg Win	229
Avg Loss	-231
W/L	0.99
Expectancy	123

SPI Trades on Chart 11-11-08

 

SPI System Update 11-11-2008

Ticker	Trade	Date	Price	Ex. date	Ex. Price	Points
SPIM_CCB	Long	30/09/2008	4494	1/10/2008	4885	391
SPIM_CCB	Short	2/10/2008	4875	3/10/2008	4670	205
SPIM_CCB	Long	6/10/2008	4575	7/10/2008	4685	110
SPIM_CCB	Long	7/10/2008	4420	7/10/2008	4685	265
SPIM_CCB	Long	10/10/2008	4080	14/10/2008	4500	420
SPIM_CCB	Short	14/10/2008	4500	16/10/2008	4055	445
SPIM_CCB	Long	24/10/2008	3903	29/10/2008	3980	77
SPIM_CCB	Short	30/10/2008	3917	6/11/2008	4176	-259
SPIM_CCB	Short	3/11/2008	4121	6/11/2008	4176	-55
SPIM_CCB	Short	5/11/2008	4330	6/11/2008	4176	154
SPIM_CCB	Open Short	11/11/2008	4018	11/11/2008	3993	25
SPIM_CCB	Open Long	11/11/2008	3980	11/11/2008	3993	13

 

New trades since last update shaded

 

Statistics

Total	1791
No Win	10
No Loss	2
Total	12
Win %	83%
Avg Win	211
Avg Loss	-314
W/L	0.67
Expectancy	123

 

Comments - the system experienced a very large intra trade drawdown on that losing trade which was more than a tad annoying.  On review I found I had included a model with a very small trade size (about 12) which was obviously curve fitted to the recent short market.  Removing that model would have reduced the total return but W/L ratio increased & open drawdown decreased significantly.  Lesson of the day - don't get too greedy & curve fit data!