Pete's Place

Here is my first stab at MM rules. I'd like to get some feedback.
One note before I dump the rules: I firmly believe that my trades are correlated with regard to winners/losers. Now that my exit and entry rules are firmed up I can actually test this hypothesis, and those results will ultimately affect these MM rules.

Anyway, here goes:


MONEY MANAGEMENT

"Here to Stay"

1. Max position size determined evening before. Hard limit, may not exceed during the next trading day no matter the circumstance.

a. Size vs. Account Equity:
i. Increase size at "x" dollars.
ii. Reduce size upon falling back to "y" dollars.
iii. Make "y" > "x" whenever possible.
iv. Never make "y" < "x".

b. Size vs. Market Behavior:
i. Have pre-planned "winner-rich environment size"
ii. Standard size
iii. "Loser-rich environment" size


2. Pyramiding is OK as long as each individual entry conforms to the regular entry rules, and the total size conforms to the normal MM rules.
.
3. A 10% loss in account size for the calendar month = done for the month. End-of-day check, not intraday.
 
An unexpected by-product of writing down detailed trading rules: I have lots of time for other (non-trading) activities. I can't believe I traded before without having a complete plan on paper.

Also, every once in a while now, when I'm trading and I don't know what to do, I realize I have a hole in the rules, or they are too sketchy in places. At those times I firm up the rules. Usually all it takes is something simple.
 
Pete, I am just curious. Since writing down your plan do you use it as sort of a "flight plan" and checking your plan before every move you make? I have a few holes in my trading as well that I feel can be filled with a "pre-flight check list". I am finding it a hard transition albeit a rewarding one.

Chris.
 
Not currently using a checklist format, but sounds like a great idea. I'm going to try it.

I think a checklist will add even more consistency to my results.

And really, I think consistency is all that matters. Only then can you tweak the system and see the effects.
 
Instead of a checklist I went with a "trade ticket".

I have a simple form I fill out with the entry and exit criteria.

The idea is to use discretion in filling out the form, but after that, execute it robotically.

My emotions have been getting a hold of me after i'm in a trade lately. Now all I have to do is stick to what the ticket says, no thinking necessary.

The tickets also support the ambush mentality, because I have to plan the entries ahead of time.
 
Sim trading this week.
Working on making the initial phase completely mechanical, since that is what gives me the most angst.

As a starting point, this week I am trailing 1 pt behind the 5 min chart for 1/2 ctcs and 0.5 pts behind for the other half. I trail until I get to breakeven, then I use the preplanned exit strategy.

Today this strategy jerked me out of two great trades. It also limited my losses on two bad trades, but not enough to make up for what I missed.

I'll trade the whole week this way and make adjustments over the weekend. I'm pretty happy with entries and profit taking exits, just need to polish up this initial phase.
 

Attachments

Goals of the initial phase

(1) Be bigger when i'm right, than when i'm wrong

(2) Get out if i'm not proven correct, rather than wait around to be proven wrong


Solution so far
-------------------

(1) Enter full size
(2) 1/2 size must perform VERY well right away
the other half is given a little more leeway


So i'm working on the technical implementation of #2. Like I said I don't want there to be any thinking involved, at least for now. The thinking part is over; I determined the entry and the profit targets. I want a systematic way to get to breakeven. There are a million ways to do this.

  • go to b/e after a certain amount of favorable movement, otherwise keep the initial stop intact
  • trail the stop
  • bar behind (variable offset)
  • time stop
  • body stop - get out if there is a complete body below (for longs) my s/r level

I guess i'll like one that gets me out of 80% of losers early for the 1st half, and one that keeps me in 80% of winners for the second half. Starting today i'm keeping track, should have some good statistics in a few weeks.
 
To study simple methods for handling the initial phase, first I've divided my trades into "winners" and "losers"-- looking just at the entry point, not where I actually exited. (You have to decide for yourself what you consider a "winning" and "losing" entry, based on your tolerance for pain.) Anyhoo, then I made a matrix with four columns: "Pass winner", "Kill Winner", "Pass Loser" and "Kill Loser".

"Kill Loser" means it exits before the initial stop is hit, in the case of a loser. "Pass loser" means the initial stop gets hit. "Pass Winner" means it gets the stop to BE and you get a chance to close for a profit. "Kill Winner" means it cuts off a trade before it has a chance to develop, and it never hits your initial stop.

In the rows i've listed all the simple methods for handling the initial phase, to get that stop to breakeven. E.g., trail a 5 point stop.

Then, I take each entry and mark the result for each method, in the appropriate column.

My Results? So far trailing a stop works great and everything else mostly sucks. Two caveats: (1) the results are highly dependent on how you trade. (2) this is *not* a profit taking exit, again i'm just talking about doing something automatically to get the stop to breakeven.

Well if anyone else does this kind of analysis it would be nice to hear how it works out for you.
 
Just went through a great exercise...

I'm looking at my past entries, winners vs. losers, printed out on the 11 min chart I now trade off of (since dumping the less efficient trendline method, where I used tick charts).

Then I printed out 2 minute charts around each of the entries.

The setup looked the same on the 11 min charts, winners vs. losers, however on the 2 min. charts there was a lot of difference. For the winners, contained in the 2 min charts were multiple failures to break through the s/r level, whereas for the losers there was only an initial pullback.

Thus I have modified my entry rules-- now i'll look for the usual signal on the 11 min but drop down to the 2 min for the actual entry. Looks like candle patterns like engulfings, or morning/evening stars, usually unreliable on a 2 min chart, are quite good in the context of a long shadow on the 11 minute.
 
Here's the 11 min chart from Thursday.
If a trader took one of the first two long lower shadows to go long, he'd be starting the days with two losses, perhaps. The third one works out but not before some chop, which would probably stop him out, only to see it run up later. Ain't trading grand?
 

Attachments

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