My SPM + SCT child

F

frostengine

Without a doubt this thread will probably get a lot of heat just because of its subject matter. Please refrain from turning this into a "war" between the two methods. The whole point of this thread is that combined they make a decent strategy.

If anyone has followed my most recent posts you will notice I have posted about SPM and about SCT.. I have actually been attempting to automate both separately...

Well a light bulb went off last night and I ended up combining the two to where now they really are nothing alike either method.. but is something entirely different..

Both methods final goal is to exploit price action... using the 2 together allowed me to find a "sweet" spot... Granted this is my first attempt at the combination and I am certain there is a LOT of room for improvement. What shocked me the most was I wasn't able to make either profitable on their own, it was just the combination of the 2 concepts that worked for me.

$5 round trip commission has been added to each round trip, as well as .25 slippage for each trade.

Below is the backtest results:

Trades: 1889
Win%: 39.76%
Profit Factor: 1.33
Sharp: .53
Profit: +$45,573 per contract

Long Only:
Trades: 1052
Win%: 44.2%
Profit Factor: 1.38
Sharp; .60
Profit: +$27,918 per contract

Short only:
Trades: 837
Win%: 34.17%
Profit Factor: 1.27
Sharp: .27
Profit; +$17,655 per contract
More...

Granted not the best strategy in the world, but that is actually pretty good numbers considering nearly 2k total trades.

The long only method seems to be more solid at this point.. Its even doing well in this current major bear market... The short portion of this strategy needs a bit more work..

BTW backtest is from: 2003-2008 ES and biggest draw down experienced at any point was -$3,500 which happened recently in the huge volatilty moves
 
Interesting frost. I've been reading the same threads. Now, I wonder... those threads are built on the premise of sharing and helping. Are you going to do the same here or just take it & run?

:)
 
Quote from brownsfan019:

Interesting frost. I've been reading the same threads. Now, I wonder... those threads are built on the premise of sharing and helping. Are you going to do the same here or just take it & run?

:)
More...

I second the question. There's a line starting. Don't fret, the ES has enough room for every ES trader on ET.
 
brownsfan,

For the most part it will be a "take it and run" approach. Although depending on the question I will answer some questions concerning it, but will in no way lay out what I did. Simply because in those threads the methods are somewhat vague and require the users to "learn" it to be able to profit from it.. My combination of the two on the other hand is really not vague at all. In fact its very mechanical, so much so that I was able to code it to do that back test, and could trade it completely automated. Thus if I shared this it would be entirely different than what they shared, as all of ET could have it automated within a day and no telling what effect that would have on the strategy ;)
 
Godzilla,

I'll make this my only reply concerning my decision to add SCT to SPM. After this I would appreciate it if we can stay on track and not turn this into a dispute between the two strategies.

I was unable to make EITHER strategy profitable on their own. That is not to say that neither are profitable on their own. You have shown in other threads that you can trade SPM profitable. However, what you do is not as mechanical as I need to develop an automated strategy. You rely on price action that you have "learned" while trading. That is not something I could code.

SCT also is related to price action. By using SCT with SPM I was able to "substitute" the unknown price action with something more "mechanical" that I could automate.
 
Quote from frostengine:

However, what you do is not as mechanical as I need to develop an automated strategy. You rely on price action that you have "learned" while trading. That is not something I could code.
More...

FTR, price action has been and can be coded very easily.

But you would have to understand it first.
 
<i>"as I need to develop an automated strategy..."</i>

All of the time, effort and money spent trying to develop a profitable system (that will hold up for years without puking) could have instead been spent learning to visually read = interpret price action.

Mechanical trading is by far the longer, harder road that relatively few ever create lasting success with.
 
Still struggling with the short side of the equation.. but have vastly improved the long side. Look at this PL chart for this long only, its amazing that is a LONG only strategy. Especially look at the last few months how its making money hand over fist in this bear market....

updated stats for long version of the strategy;

Profit: +$37,027
Trades: 432
Win% 58.1%
PF: 2.02
Sharp: .69

You will notice this strategy now takes FAR less trades, so I became concern that I may have curve fitted it, so to alleviate my fears I tested the strategy on ER2 and NQ

ER2 Results: +$33,580
NQ Results: +$10,701

Granted that does not prove its not curve fitted, however it does help a lot.

<img src=http://elitetrader.com/vb/attachment.php?s=&postid=2142589 width=800>
 

Attachments

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