May Trading Summary

Note: Only traded 8 days
Net breakdown (contracts traded):
ZS $240(2), 6E -$614(53), TF -$961(12)

RESULTS FOR MAY
Contracts:	67
Net $P/L:	-1335
Wins:	27
Losses:	21
Win%:	56
$Commissions:	293
Avg$Win:	78
Avg$Loss:	-164

Fri. 5/31

3:55pm CDT - TF testing today. Didn't go too well but as usual with live testing, I thought of some ideas for improvement.

RESULTS FOR DAY
TF Contracts:	12
Net $P/L:	-961
Wins:	1
Losses:	6
Win%:	14
Avg$Win:	107
Avg$Loss:	-178

Tues. 5/28

11:30am CDT - OK, this may conclude my discretionary test drive.  Time for more coding and backtesting to find a more mechanical method to my madness!

RESULTS FOR DAY
6E Contracts:	10
Net $P/L:	-685
Wins:	3
Losses:	4
Win%:	43
Avg$Win:	54
Avg$Loss:	-212

Fri. 5/24

9:10am CDT - Just $71 to show for my test drive this week of a new method.  Sure beats sim trading though which I absolutely hate.  You really do learn so much by actually live trading a system. 

RESULTS FOR DAY
6E Contracts:	7
Net $P/L:	117
Wins:	4
Losses:	3
Win%:	57
Avg$Win:	67
Avg$Loss:	-51

Thurs. 5/23

10:30am CDT - Choppy morning in the red and decided to hold final trade of day, short, with wide 25 pips between my stop and target. It came within 2 ticks of target then shot up and took me out at high tick of day 1.2924.

RESULTS FOR DAY
6E Contracts:	15
Net $P/L:	-559
Wins:	5
Losses:	4
Win%:	56
Avg$Win:	105
Avg$Loss:	-271

Wed. 5/22

10:10am CDT - It took some time to get used to an actual moving market after the last 2 days. I missed quite a few trades and didn't hold for as long as I should have on many of the ones I did take.

RESULTS FOR DAY
6E Contracts:	10
Net $P/L:	265
Wins:	5
Losses:	3
Win%:	63
Avg$Win:	105
Avg$Loss:	-87

Tues. 5/21

10:15am CDT - Bank holiday in Europe explains the dead market yesterday but it was equally dead today...

RESULTS FOR DAY
6E Contracts:	5
Net $P/L:	101
Wins:	4
Losses:	1
Win%:	80
Avg$Win:	33
Avg$Loss:	-30

Mon. 5/20

9:35am CDT - Taking a break from coding strats for a couple hours each morning to test drive a discretionary method I have been working on.  Not sure how long I will do this for but I guess results will dictate that decision soon enough.  And it may very well turn in to a mechanical system!

RESULTS FOR DAY
6E Contracts:	6
Net $P/L:	147
Wins:	3
Losses:	0
Win%:	100
Avg$Win:	49
Avg$Loss:	0

Another day...

I never considered that learning Ninjascript would lead to depression!  Perhaps my rose-colored glasses I used to wear clouded my realistic view on trading and I am now being forced to see the true reality that developing a good mechanical strategy is even that much harder than I thought (and I thought it was really hard before).  I'm still optimistic but now that I can find out in a day or two the results of a given trading system with years of cold hard data, some coding, and dozens of near instantaneous Ninja-generated P/L curves, leads me to realize how naive I have been up until now.

Once again, another system curve example below from TF 1-min. charts that fails to meet any sort of acceptable criteria for a trading system.  I think the Avg. Profit per trade was $10 - and that's with no slippage or commission.  Generate enough of these day in and day out for various systems I've coded so far, and one could certainly justify some depression.

A 100% mechanical trading system may be my ideal goal but some discretion on taking signals might not be such a bad thing.  When I look at the trades the above system took, I can see many reasons why some are just not good.  Sure I can edit the code to "filter" some of these out but then some good trade signals are missing that I would have taken.  It's impossible to try to program everything this human is thinking. It's a fine line.

In the meantime, the work continues and I'll just keep my sanity and optimistic view on life by keeping the live bald eagle cam playing on one of my monitors.  There's something relaxing about hearing the eagle chirps, watching the chicks lie around all day, and seeing the live fish being delivered by one of the parents to the nest.  The simple life! (here's direct link: http://humwild.org/humboldt-bay-eagle-camera)

Live stream by Ustream

Martingale Trading System

Since my crazy trading ride in 2007 which involved a martingale system, I have sworn never to try anything like that again!  As any trading professional will tell you, it is suicide for the retail trader.  It's like playing Russian Roulette but worse because you don't get put out of your misery at the end!

Ever since then, however, I've always told myself that if I learned how to program strategies and could automate the backtesting of them, I would want to see for myself the results for a martingale type system.  Well that day is finally here since I can program in Ninjascript and I'm sharing a simple strategy with anyone who wants to play with it, in NinjaTrader's Strategy Analyzer, and see for themselves why you should NEVER trade a system like this.

This chart explains and shows how the strategy trades, fading moves looking for a reversion toward the mean.  The default is a max position of 128 contracts.  Lots of the variables can be changed but it doesn't make any difference.  There are no settings that avoid huge drawdowns over time.  And sure, you can recode the strategy to go with a trend instead of fading it, but again, it won't matter in the end.

As I state in the strategy code, "This is a martingale strategy and is not recommended for anyone. There are ZERO ways to configure the variables to avoid an eventual huge drawdown. Recommend backtesting with ES only (due to liquidity) and using years of 1 min. data."

Here are a couple examples of results using 6 years of ES 1 min. data. In the first, system is looking for just 2 ticks of profit, on net position, on every trade:

Same settings but looking for 6 ticks of profit instead:

Feel free to download here, import the single Ninjascript assembly (not the 4 files) into your NinjaTrader, and explore for yourself. Let me know if you have any problems or questions.  And if you find any Holy Grail settings, let me know! :)

Fri. 5/10

11:50am CDT - It's been way too long so had to take a couple scalps after the crop reports came out.

RESULTS FOR DAY
ZS Contracts:	2
Net $P/L:	240
Wins:	2
Losses:	0
Win%:	100
Avg$Win:	120
Avg$Loss:	0

TF System

Last month was the first month in memory that I made no trades.  I have been building and testing strategies in NinjaTrader and that work continues.  I just haven't found anything I'm satisfied with to actually start trading.  Hopefully that happens soon.  It's really really hard to find something that nets more than a few ticks of profit per trade per contract with minimal drawdowns!

As mentioned a month ago, I decided to focus more on TF since that is the best bang for my commission dollar.  That said, I thought I'd share some data for one strategy that I recently finished coding but am still hoping to improve and make better. If you have any thoughts about whether you'd trade this, let me know.

Initially just backtesting the past year, results looked pretty good.  But then when I tested 3 years of data, the drawdown became unbearable.

Curious what system optimized for min. drawdown would look like for past 5 years yielded:

There are only 2 variables in this system which determine settings for profit and stop on each trade and any open trades are exited at close each day. I haven't done anything yet with walkforward testing but that may be something to look at to improve results.

Or perhaps the next idea I test will be much better?  Who knows...