My money management (advanced)

Hello.

I am sharing my hundreds of hours of work with you in a concise thread...

The crux of the problem is here:

http://www.forexfactory.com/showpost...6&postcount=22

I only take about 200 trades a year, and my goal is to compound make account into several 1000% by the end of the year. High risk, high return. To do this with fixed fraction, I'd need to risk between 8 and 15% per trade. The problem is, after a few losses, you are below 50% of your highest balance, which is unacceptable.


My MM goes like this.... I look back at the last 8 years of my trades, and find the highest PIP DRAWDOWN I incurred. I divide my HIGHEST BALANCE by that (and then by 10, to account for $/lot), and will trade that lot size until a new higher balance is achieved, in which case I recalculate the lot size.

I do not position size my trades by dividing a percent of my account by the stop.


This method eliminates the problem I posted the hyperlink to. Unfortunately, historical drawdown should be simply a GUESTIMATION of what drawdown is expected.... new higher drawdowns in the future (compared to the past 8 years) are inevitable. So, to compensate for this, if my account draws down 33%, I increase my "drawdown divisor" by a factor of 1.5..... the same is true if the account draws down 66%.... then my drawdown divisor increases again to 2X the original.

Is there risk for bankrupting the account? YES! Is this a happy medium between bankrupting the account and making huge profits? YES!

This money management is strictly concerned with pip drawdowns.... and new pip highs ARE CRITICAL. you are being paid for performance.

Here is 2008 with this method. the graph is based on about 2500 green pips on the year.

PLEASE MAKE SURE YOU READ THE BOTTOM HALF OF THIS POST BEFORE LEAVING:
http://www.forexfactory.com/showpost...2&postcount=16

Hi Tdion
I'm not clear on how you eliminate the problem. Regardless of betting methodology; if I have a 1000USD acct and lose 200USD, (20% loss), it will still take 25% of 800USD to get back to my starting point of 1000USD.
Maybe I'm just misreading something here.

I also trade w/ lot sizes and not fixed fractional.

Hi Green let me elaborate (and thank you for the question.)

First, does money management ever give a trader an advantage? NO! Let's make that clear first. The best we can do is mitigate our adversity as much as possible.

For me, this MM technique was one of the greatest discoveries in trading forex. Here's an example:

8 year max drawdown in pips: 750
8 year green pip tally: 19,000
R:R: between 1:3 and 1:5, depending on the setup

Let's say I start the year with 10K. My very first trade will be 10,000/750/10, or 1.3 lots.

With fixed fraction, if you lost your first trade, your next trade would be a typically be a smaller lot size. But in this case, the value per pip remains constant (the key to the method.)

Let's say I lose the 1st, 2nd, and 3rd trades.... but I win the fourth. With 1:3 R:R, I am back to breakeven. The same would not be said for fixed fraction, because FF is bound to the current account size.

In order for the account to draw down 33%, I will have a series of trades that net 250 red pips (750/3).... so then my lot size decreases.... 10000/1125/10 (1125 is from 750*1.5).... The new lot size is now 0.9 opposed to 1.3.... This is my best defense.... the losses can continue, but drawdown slows the erosion of my account...

Suppose drawdown continues at this point to 66% (or $6666 in terms of money.... this would require a total drawdown of 620 pips.... 250 for the first 33%, 370 for the 2nd 33%) Now the lot size per trade is reduced to 0.7 lots per trade... allowing for 476 more red pips until complete account wipeout. (if you are unfortunate enough to draw down more than 66%, you'd need to add some extra margin to trade the third 33%)

So, historically I had a max drawdown of 750 pips, but this MM allows for up to a 1096 pip drawdown.



Let's say you drew down 40%, and then recovered back to a 32% drawdown.... the lot size will increase from 0.9 back to 1.3 because you are out of the second level.

As the account reaches new high balances, you would start the process over again by dividing the new high by 750 and then by 10. Both fixed fraction and my MM are "pay for performance" MMs.... you will lose money without new pip total highs.... but it is irrational and illogical to think we can make green money with red pips.

Only in-as-much as it allows the law of large numbers to play out. If that's against you, then money management only guarantees a loss.

You're basing your MM on a historical max drawdown during some period, the assumption being that new losses won't exceed history. Have you ever considered expanding that theory? Or perhaps putting some kind of loss limit in place around there? I might be tempted to use a statistics approach to drawdown instead, going for say 2nd dev pricing instead.

While that won't protect you against big moves, I'm assuming you'll have some kind of stop loss in place too to prevent catastrophic loss. How are you factoring in stop losses?

The benefit here seems that you're using a (relatively) fixed trade size, instead of a per-trade sizing. I do something similar, but using starting balance. I take my weekly balance and take 1% of that instead of re-evaluating balance every trade.

The method you describe reduces the number of trades to return to BE and not the mathematical fact you need 25% to recover from a 20% draw down. I understand where you are coming from, I hadn't looked at it from this angle before, but you are correct.

I re balance my lot size on a weekly basis regardless of increase or decrease, which appears to differ from your re balance on basis of account size. I also use min lots for greater granularity.
gosh Tdion, we seem to be agreeing on something LOL

Hi rabid

From post #1:

"Unfortunately, historical drawdown should be simply a GUESTIMATION of what drawdown is expected.... new higher drawdowns in the future (compared to the past 8 years) are inevitable. So, to compensate for this, if my account draws down 33%, I increase my "drawdown divisor" by a factor of 1.5..... the same is true if the account draws down 66%.... then my drawdown divisor increases again to 2X the original."

Hey Green I've always been a fan of being buddies with people I something in common with

From post 1:

"Is there risk for bankrupting the account? YES!"

I am in the camp that the entire account is viewed as "risk money." If I weren't willing to lose every dime in the account at one time or another, it would be put into other investments. It's a price to pay for larger returns.... fixed fraction prevents bankruptcy with its parabolic erosion of the account during losing streaks, but will not produce massive gains when the winning streaks return.

Rabid, I said every trade is between 1:3 and 1:5 R:R. That intrinsicly means I have stops in place....

Here are identical trades using 8% risk per trade.... about 1/2 the total return for the year.

I could risk 10 to 15% per trade, but margin gets ugly. That is another nice thing about my MM.... the size of the trades are not huge.... but rather medium sized in the worst case..... and therefore my 100:1 leverage doesn't give my problems with margin calls.

I've given lots of info and I know it is too much to digest. BUT.... if you are serious about trading, it is worth exploring and rereading.

-TD

Laff. I'm not criticizing, merely clarifying. It's good to know our assumptions, and even better to know that we are assuming them... in case there's a way to fix them.


Yea, but you didn't specify how they're coming into play.

For instance, fixed risk (variable lot) sizing:
$1000 balance. 1% risk. 30 pip stop.

($1000 * .01) / 30 = .3333 dollars per pip. That'd be 3 micro lots (assuming USD base).

Naturally as balance decreases your trading lots would decrease, which is what you're trying to avoid. But if you're sizing your lots based on drawdown and getting a fixed lot size, your risk is variable per trade if your stop loss is variable. So are you using a fixed or variable stop loss?

For me, it is variable. Some setups are X pip SLs, while others are Y pip SLs.... All that really matters is, when the stops are hit, are they accumulating beyond my red pip threshhold.

I've attached an image showing

1) lot size increasing as the balance increases
2) lot size decreasing after a 33% drawdown, and then increasing after the balance recovers

hopefully this visual will bring clarity.

Tom, I guess by that logic we're not buddies any more. .......

.......

I did my thoughts to death in the thread in your link in post #1, so I won't repeat them in full here. LOL.

Very quick summary, before I disappear from this thread: IMHO, there's nothing wrong with departing from fixed fractional. By sizing higher you'll accelerate recovery from drawdown, i.e. alleviate the "down 20%, need 25% to recover" problem, just as you say. But, if the losing streak happens to continue, then your higher position size will drive you deeper into drawdown, more quickly. So long as you're prepared to live with that risk, there's no problem.

Position size is a double edged sword. There's no way around that.

David

(Gotta love those emoticons, that crybaby one just looks like me after my stoploss has been taken out)

This MM strategy is a great idea tdion, something I have been looking for!

What if you use this strategy only on the "market's money" and use a fixed fractional whenever you drawdown to your initial deposit?

P.s. For how long have you been using this type of MM successfully?

As for fixed fraction.... nobody says you can't do that.... but fixed fraction during any sort of drawdown will hurt the compounding that can later occur during a winning streak.... so I am willing to take more risk in return for a faster recovery and higher compounding.

http://www.forexfactory.com/showpost...71&postcount=8



One of the most important things I did not write earlier on this thread! I am pushing MM to the extreme. In my example, I used a 750 pip drawdown.... With 750, you might want to start with a 1500 divisor.... which will cut your risk down substantially (but will also eat at your potential profits.)

You can also simplify things by never recalculating the divisor, and just use an initial divisor that is higher than your max historical drawdown. (Otherwise margin calls will become a concern sooner)

It really just depends on your appetite for risk. Results will vary.

I am attaching the RAW PIP PERFORMANCE of my strategies for 2008 that my graph in post #1 is derived from....

Each trade was 0.1 lots, so every dollar is the equivalent of a pip (either green or red.)

Although it looks very good and linear, there was a sizable drawdown of about 500 pips at about a quarter of the way through the graph.

EDIT) Please, nobody start drawing trendlines on my graph

I can only imagine that in real time this is a MM stragety which will not only subject the trader to a large drawdown, but also wipe out.

Indeed, wipeout is a possibility (which has already been discussed a few times...)

It all comes down to philosophy.... and appetite for risk.... the size of divisor needs to be adjusted to meet the trader's own risk level.