Larry Hite on "why a high % of speculators lose money ?

Discussions about Money Management and Risk Control.
Post Reply
PTCM
Roundtable Fellow
Roundtable Fellow
Posts: 82
Joined: Fri Dec 19, 2003 2:58 pm
Location: Hong Kong

Larry Hite on "why a high % of speculators lose money ?

Post by PTCM » Thu Aug 04, 2005 2:59 am

Larry said " assume that they are better than average at picking winners and assume that they are right 50% of the time. They run their losses until they have lost 50% of their capital and take their profits when they are up 10%. They have almost guaranteed themselves a loss of 20% over time!" he exclaimed.

"Now assume we reverse these habits, cut out losses at 10% and run out profits until we have at least doubled our money. We don't need to be right more than 25% of the time to still generate a positive return of 17.5 over time. The important thing is that we have skewed the return distribution to the right by changing a bad habit." he said

I thought what Larry Hite said was very intriguing. I am trying to apply his logic to a simple trending following system with around 50% accuracy. On TradeStation, I can code up his logic pretty quickly. Just wondering if anyone here has tested similar ideas ?

TrendMonkey
Roundtable Knight
Roundtable Knight
Posts: 154
Joined: Fri Apr 22, 2005 9:14 pm
Location: Vancouver, Canada

Post by TrendMonkey » Thu Aug 04, 2005 9:22 am

Isn't that like saying that people are fat because they eat too much? (In other words, isn't this obvious?)

Anyway, if I am not mistaken, cutting losses at 50% is NOT the opposite of letting profits double. I have lots of stocks that are up 50% but none that are up double, which would be 100%.

Paul King
Roundtable Knight
Roundtable Knight
Posts: 207
Joined: Mon Feb 23, 2004 9:13 am
Location: Vermont, USA
Contact:

Sounds simple but this is the key to making money trading

Post by Paul King » Thu Aug 04, 2005 9:24 am

All my trading systems use these ideas to create their 'edge':

Have a win% (slightly) better than 50% (usually between 50%-60%).
Keep losses small (compared to winners) by sticking to my initial stops.
Allow winners room to grow i.e. don't trail stops too tightly with winners.


If your average win is bigger than your average loss then you don't need an entry signal that is any better than random.

This is simpy the practical solution to the old 'Golden Rule' of trading that everyone knows which is 'Keep your losses small, and allow your winners to run'. Simple to understand but not so easy to do - kind of like losing weight; you know how to do it (eat less and excercise more) but it is not easy to do it with discipline and consistency for a long period of time.

Paul

TrendMonkey
Roundtable Knight
Roundtable Knight
Posts: 154
Joined: Fri Apr 22, 2005 9:14 pm
Location: Vancouver, Canada

Post by TrendMonkey » Thu Aug 04, 2005 4:18 pm

That is an amusing analogy, I have often thought all the trading discipline stuff bears a lot of parallels to weight loss. They both sound easy and anyone ought to be able to do them with their eyes closed, and they both are prone to periods of discouraging results.

Paul King
Roundtable Knight
Roundtable Knight
Posts: 207
Joined: Mon Feb 23, 2004 9:13 am
Location: Vermont, USA
Contact:

Post by Paul King » Thu Aug 04, 2005 4:29 pm

The one big difference with dieting versus trading is that trading has no guaranteed results.

If you eat less (in calories) than you use up (in excercise), the laws of conservation of enery dictate that you have to lose weight.

If you diligently adhere to your trading plan, there is no guarantee that you will make money (either becuase you have a negative expectancy system, or it is simply going through a losing period), and that is what makes it difficult.

Most people could not tolerate a 'job' where some months, instead of getting your paycheck, your boss asked you for the last 2 months pay back, and other months he gave you a bonus equal to the last 3 months total pay :-).

My conclusion: dieting may be difficult, but it is easier than trading, but trading is a whole lot more interesting than dieting is :-)

Good luck in your trading (and dieting)

Paul

PTCM
Roundtable Fellow
Roundtable Fellow
Posts: 82
Joined: Fri Dec 19, 2003 2:58 pm
Location: Hong Kong

how did Hite do it ......

Post by PTCM » Fri Aug 05, 2005 12:55 am

The concept is simple, but if everyone could do it in the real trading world with their eyes closed, I guess we all would be very rich already.

From my testings, I found that it's not difficult to add values to a single paremetre strategy by changing the money management rules on a yearly basis. However, if you look at the equity curve on a fixed holding period basis, ie 6-month, consistentcy is hard to find.

Hite said his simulations indicated a probability of over 80% that his combined strategies portfolio would be profitable, and that's what convinced MAN to back MINT's approach. 80% is way too high in comparison to the Turtle System. I believe the Turtles succeeded because of their win/loss ratio and profit factor, but the drawdowns could be very significant.

Paul,

I went to your website. Could you produce consistent + expectancy with your eye closed without chaning the paremetres in 04? BTW, what software are you using to combine equity curves ? Just curious .... PortfolioMerge seems to be good, but it only support TradeStation 7 or 8, not 2000i. Most of the hedge funds here in Asia are using 2000i.

Paul King
Roundtable Knight
Roundtable Knight
Posts: 207
Joined: Mon Feb 23, 2004 9:13 am
Location: Vermont, USA
Contact:

Post by Paul King » Fri Aug 05, 2005 8:17 am

2004 was a difficult year and my systems 'only' made just over 11% (my target is 20%, and I already have a better return through July 2005 than I did in all of 2004).

My trading is not 100% mechanical, and with so many thing in trading, getting exactly what you want off the shelf is not possible, so all my software is proprietary/custom built. I have found that the 5% you can't code/automate can make a big difference to your performance.

I use very little optimization so the 'parameters' I use do not generally change much from year to year. The more solid an idea is, the less optimization is required, and the greater the range of parameter values that work OK.

Roscoe
Roundtable Knight
Roundtable Knight
Posts: 250
Joined: Sat Jan 24, 2004 2:06 am
Location: Houston TX

Post by Roscoe » Fri Aug 05, 2005 5:39 pm

Hello Paul,
Paul King wrote:I have found that the 5% you can't code/automate can make a big difference to your performance.
Could you describe some examples of the things that you can't code/automate please? I would like to try with my setup just out of curiosity.

BTW, thanks for your e-books. I have read Expectancy and am now in the Market Types book, which is proving to be very interesting indeed.

Paul King
Roundtable Knight
Roundtable Knight
Posts: 207
Joined: Mon Feb 23, 2004 9:13 am
Location: Vermont, USA
Contact:

Post by Paul King » Fri Aug 05, 2005 7:18 pm

Ross,

Glad you are finding the eBooks useful :-)

As for the 5% discretionary part of my trading, I use it exclusively to skip low risk/reward trades i.e. I never enter a position that has not been signalled by one of my systems, but I may use discretion/experience to avoid signals under some circumstances.

Examples of discretionary entry signal avoidance (that is difficult to code) include:

When I know I am unable to calculate as accurate estimate of prevailing volatility. This can occur when there is a 'significant' event pending such as an earnings call, or when it is a recent IPO.

When an instrument is not currently liquid enough for my position size. I have a specific set of criteria that determine my list of 'tradable' instruments, but I 'eyeball' each trade to see whether it is currently liquid enough - this part is purely based on experience when I look at the last 5 days 1 minute chart and can 'tell' if I want to trade it or not.

When the opening of a US equity traded on the NYSE is delayed due to an order imbalance it exhibits certain behaviours that are not easy to code, but which I use to determine if a gap open is 'real' or not.

If you have any other questions regarding any of my eBooks, feel free to email me.

Good luck in your trading.

Paul

Vinny
Contributor
Contributor
Posts: 4
Joined: Sun Aug 07, 2005 2:41 pm
Location: Chicago

Re: Larry Hite on "why a high % of speculators lose mon

Post by Vinny » Tue Aug 09, 2005 4:55 pm

PTCM wrote:Larry said ". . . Now assume we reverse these habits, cut out losses at 10% and run out profits until we have at least doubled our money. . .
And as I understand (and try to follow for my own trading) true trend-following would not limit profits to DOUBLING our money.

PTCM,
I would be interested in reading in what context Mr. Hite said this. Could you give me a link to the entire article?

Post Reply