|
PureBytes Links
Trading Reference Links
|
I had responded to the originator of this thread privately, but I
may as well do it here. I agree with most of what Gary Fritz says
below.
I have been experimenting with the fixed ratio position sizing
strategy using ProSizer, the Monte Carlo simulation tool I developed
just for this purpose (see http://unicorn.us.com/trading/prosizer.html
if you're curious about it).
I compared fixed fraction, fixed ratio, percent risk, and percent
volatility position sizing models. In all cases I adjusted the
parameters so that the average of the Maximum Drawdown from all
the trials came out to 25%. Then I looked at the return. I did
this for trades generated by two different trading strategies. My
assessments are as follows:
* Fixed ratio usually performs better than fixed fraction.
* One will likely find that the %risk or %volatility models described
by Van K Tharp superior to fixed ratio, for the same drawdown.
* Fixed ratio is dangerous: higher standard deviation of drawdowns,
higher probability of ruin. It fails to account for equity size
or risk per trade.
* I think it's irresponsible for Ryan Jones to promote this method
to beginning traders, who won't understand the risk involved. On
the other hand, I have noticed that sometimes fixed ratio is the
best-performing model for small accounts.
I'm not surprised that Ryan Jones has traded himself into a 95%
drawdown. I think that's partly due to the high risk of fixed
ratio, and partly due to the fact that you take every trade
regardless of risk, volatility, or account size.
I disagree with Gary that one should take the basic fixed fractional
approach. My own experiments suggest that there are other models you
can use, and combinations of models, that give superior results.
-Alex
>From: "Gary Fritz" <fritz@xxxxxxxx>
>To: omega-list@xxxxxxxxxx
>Date: Thu, 25 Jul 2002 20:35:25 -0600
>Subject: Re: Any Ryan Jones Money Management Fans?
>> Just curious to hear experiences of traders who are using
>> the Fixed Ratio money management method of Ryan Jones.
>
>I posted the following on the Code list last year. Now this will be
>in the Omega archives so I can just point to it there. :-)
>
>==========================================================
>
>I read "The Trading Game" last weekend, and I was NOT impressed.
>
>First off, I find it remarkable that people can get so much attention
>for the radical concept of position sizing. I mean, are there that
>many people out there who don't understand that you'll make more
>profits (assuming a positive expectation) if you trade multiple
>contracts!? Jones talks like scaling your size up is a huge
>revelation.
>
>But ignoring that, I think his Fixed Ratio approach is bogus. IMO
>his entire premise is flawed: he looks at the per-contract profit it
>takes to move from 1 to 2 contracts, and he says that it should take
>the same per-contract profit to move from X to X+1 contracts. I.e.
>if you need $10k profit to move from 1 to 2 contracts, you should
>need $10k profit **per contract** to move from 100 to 101. You'd
>need $1M total profit to increase by 1 contract.
>
>I think this is flawed for 2 reasons: first, it relies much too
>heavily on the size of the contract. The entire leverage structure
>he computes would be totally different for, say, $250 SP's vs. $500
>SP's. But the big flaw is his use of additive growth instead of
>percentage growth. Moving from 1 contract to 2 isn't equivalent to
>moving from 100 to 101; it's like moving from 100 to 200! I think
>simple fixed-fractional approaches handle the position sizing much
>more logically.
>
>What really honks me off, though, is the way he cooks the books to
>make his approach look good. Fundamentally what he's doing is using
>very high leverage when the account is small, and backing off as the
>account gets big. This has the advantage that it gets the small
>account off the ground & running quickly. But it also exposes you to
>a lot more risk early on. He uses all kinds of examples to show how
>the FR approach can take a $X per contract loss with a much lower
>drawdown than FF -- but he constructs his examples so that drawdown
>happens AFTER he's scaled back the leverage. He conveniently
>neglects to mention that the FR approach would BANKRUPT you if that
>same per-contract loss happened early on with higher leverage.
>
>Add to that a host of logical and math errors, and I was SERIOUSLY
>underwhelmed.
>
>My advice would be to use a basic Fixed Fractional approach. Decide
>what leverage works for you, taking into account your risk tolerance,
>the Optimal F of your system (make sure you trade far UNDER the
>"optimal" F value), etc, and just risk a constant percentage on each
>trade. As your account grows, you may decide to back off on the
>leverage a bit. You can do all that without the Fixed Ratio
>complexities.
>
>Gary
>
>
>PS: According to
>http://groups.google.com/groups?hl=en&selm=9c957579.0201081404.5c89813
>c%40posting.google.com
>Jones traded himself into a 95% drawdown, presumably using his own MM
>techniques. I can't verify the accuracy of that claim, but it
>wouldn't surprise me at all. As I explain below, all it takes is a
>drawdown ***early in the account's life, while Fixed Ratio has you
>using dangerously high levels of leverage***, to produce results like
>that.
>
|