PDA

View Full Version : Why Can't people understand this?


Matt Bowen
08-16-2007, 07:20 PM
First, I want you to read something:

============================

A Cherokee elder sitting with his grandchildren told them, "In every life there is a terrible fight -- a fight between two wolves.

One Wolf is evil: he is fear, anger, envy, greed, arrogance, self-pity, resentment and deceit.

The other Wolf is good: joy, serenity, humility, confidence, generosity, truth, gentleness, and compassion."

A child asked, "Grandfather, which wolf will win?"

The elder looked him in the eye. "The one you feed."

============================

Now, why did I have you read this?

Because, it's one of the main reason why so many traders fall victim of making BAD CHOICES. An example of a bad choice is looking for a high percentage of winning trades. Think about it for a second... if you keep chasing after a system of high percentage of winning trades, you are feeding into a losing proposition (you might not know it), because you are a new trader and don't understand how trading really works.

Let me give you an example:

Earlier this week, I received an e-mail from a trader who did a little experiment with profit targets and he found out first-hand why profit targets are not the most profitable way to exit a trade. Not only that, if you average profit per trade is too small they will actually work against you... but before we get into that let me show you his e-mail:


From: **** ***** [mailto:XXXXXXX@earthlink.net]
Sent: Sunday, August 12, 2007 1:41 PM
To: Matt Bowen; Steve Griffiths
Subject: MTPEod Backtesting


Dear Steve and Matt,

A while ago I sent you an email asking about whether or not MTPEod was backtested and how I could do it myself. Matt replied that he tried to backtest S&P stocks (was it S&P index futures?) and it was difficult to do it and time consuming. I understand this. Unlike other trading systems, to backtest MTP one would have to find all setups that had developed in the past and see what the results would have been depending on how they were traded. So the task would be: 1. to find the setups and 2. to trade them.

So, I decided to do backtesting manually for a continuous gold futures contract (GC-067) going as far back as I could. My starting account was $40,000. Using the training mode, I was able to go back to the beginning of 2002. Then I moved forward one day at a time until the program showed that a setup had developed. For each setup that showed R/R more than 2:1 I took a trade and filled out a trade sheet. In that sheet I noted the following:

- Entry price
- Stop trigger
- 100%R price
- 200%R price
- 300%R price
- Total $ risk (for this exercise I used 1% of my account for position sizing)
- Total point risk

If I "entered" a trade, I followed 5 different lines of action:

1. Always exit if the stop loss was hit.
2. If the stop loss was not hit before 100%R price was reached, take a 100%R profit
3. If the stop loss was not hit and 100%R was reached, move the stop loss to breakeven and continue to hold the position. If the price reaches 200%R price, take the profit. If it does not, exit on a breakeven trailing stop.
4. If the stop loss was not hit and 100%R was reached, move the stop loss to breakeven and continue to hold the position. If the price reaches 300%R price, take the profit. If it does not, exit on a breakeven trailing stop.
5. If the stop loss was not hit and 100%R was reached, move the stop loss to breakeven and continue to hold the position. Then follow the price action positioning the trailing stop below market retracements as I would have done had I been actually trading and watching the market every day. The exit (and profit) would be taken when market hits the trailing stop.

It took me about a week working an hour or two a day to fill out forms for each of the 23 trades I found for gold from 2002 to the middle of 2007. Then I prepared excel spreadsheets for each method of trading outlined in 2 to 5 above (#1 applies to every course of action). I attached an excel spreadsheet that shows the summary for this study. It is self explanatory but I would be happy to answer your questions if you have any. My main goal was to see if your "Isolation Approach" was really as good as you said and now I proved it to myself that it really is. The backtest also proved once again that it is not worth taking many "easy" small profits and have a high percentage winning trades. One really needs to let the profits run. This experience and results will undoubtedly make me a better trader.


Have a good day,

**** *****

Here was my response:

Hi ****,

Thank you for the report and hope this exercise was a valuable lesson... I'll get to that in a minute, but first let me correct something you said in the e-mail: "Matt replied that he tried to backtest S&P stocks (was it S&P index futures?) and it was difficult to do it and time consuming".

Matt didn't try...Matt Did! One of the first things I had to do (Before I started working with MTPredictor) was to prove to myself that the software worked. I never take anybody at face value when it comes to putting real money on the line and just like you did, this experience will teach anyone a few very important lessons:

# 1. Profit targets are NOT the most profitable way to exit a trade. In fact, they severely limit the amount of money made with a system the has less than 50% win rate.

# 2. Loss control... you now understand why it's an ABSOLUTE MUST to control or limit the DOWNSIDE of any trade!!!

# 3. Letting profits RUN is what makes BIG money....PROFIT TARGETS are for people who are trading scared money and our afraid to manage a trade. Their FEAR actually cost them money (and most of them don't even know it).

# 4. After running this exercise....there is NO DOUBT that MTPredictor was profitable and is a professional trading product because it focuses on risk control and trade management.

I'm glad you too the time to do this exercise... because in order to become a profitable trader, you must go through an exercise like this to SEE FIRST HAND whether or not YOUR BELIEFS are valid and if you will make money or limit your success. See, 95% of most traders will not even do what you did here with this test... because the THINK what they are using works (they assume it works) and they don't want to question or test their own belief systems. When in fact, testing is the ONLY WAY to know if what are using is the BEST thing for you, not only in profits, but in risk control.

Good for you ****, this just shows you are well on your way to becoming a true professional.

Ok ****, let me ask you something: "Now that you have run this experiment or test, does it show you now very clearly what you once thought otherwise?" In other words, you might have liked the idea of a profit on 74% of your trades (17 out of 23 trades)... BUT LOOK HOW MUCH MONEY YOU LEAVE ON THE TABLE, by not letting your profits RUN!

By clicking on the link below you can see clearly that almost 5 times (5X's) the amount of money is left on the table by simply forgetting about how many trades you win on and letting your profits run using a trailing stop (like the ATR-stop in MTPredictor).

Click here to see why profit targets do not work:

http://www.mtptrader.com/GoldTrades.xls

(Don't worry, this is not a virus I have opened it on my computer)

I'm convinced, this is not hard for people to understand...it's hard for people to accept! If you want to become a profitable trader you need to stop listening to all the bull-shit out there and start learning from people who know what they are talking about.

Are you ready to here the truth?

If so... then click here: http://www.mtptrader.com/Tharp.pdf

My advice is print this document out and read it three (3) times. When you do this is will start to burn an impression in your head about what REALLY WORKS IN TRADING. I would save this e-mail because it's one of those e-mails that should get your attention. If by chance this e-mail does not get your attention, then you are just not ready to move forward in your trading. Don't worry, you'll pick it up soon enough, just stick around here for a few months, leaning to become a profitable trader is not something you find in a box or a book...you find it by hanging around somebody whose doing it.

As always... My very best to you,

Matt Bowen

Rob E
08-16-2007, 09:21 PM
Well done Matt!!!

As usual... thanks 4 the HARD HITTING TRUTH!!! ;)

Matt Bowen
08-17-2007, 04:29 PM
Hi Rob

Thank you...

As usual... thanks 4 the HARD HITTING TRUTH!!!

The Hard Hitting Truth is: It's not hard for people to understand...it's hard for people to accept!

This goes for anything related to money management!

Have a great weekend... I'm off to the Country Club to get in 9 holes before dinner :D

Matt

Steve Griffiths
08-18-2007, 06:33 AM
Hi Everybody

I agree with Matt here and this is exactly what I found when initially developing Mtpredictor many years ago, in that approximately 70% of the time the automatic set-ups reached “at least” the 100% initial risk zone. So in this respect, yes MTPredictor is 70% correct in it predictions………. And that is where many developers would have left it and as such marketed a “high % winning system”

But we are different, at MTPredictor we are not interested in a high % of winners, we are interested in making money, and once I added in the WPT targets (unique to MTP and NOT just random profit targets) then the $ bottom line increased significantly.

As you can see from this test, by also introducing the “break-even” rule your number of winners dropped, because a number of trades were break-even BUT the bottom line $ profits increased. This is professional trading.

Since then we have added the ATRStop, which has allowed the big winners to run even further, and recently we have introduced the STF “strength band” which will tell you when to use the WPT targets (in a sideway choppy market) and when to run your trades using the ATRStop (in a strong trend). Strong trends only happen about 30% of the time. Both of these are designed to help maximize your bottom line $ profits.

In other words, we at MTPredictor are only interesting with how markets “really unfold” and making as much money in the “real-world” rather just designing a system that amateur traders “think” (wrongly) they need.

So you have a simple choice, accept that way the “real-world” works or stick your head in the sand and fail like 97% of amateurs traders…..

Thanks

Steve

prh
08-18-2007, 11:30 AM
Matt and Steve:

This concept is extremely valuable. Some time ago I posted a similar analysis for all markets for an entire year. The conclusion was the same, in that letting the profits run produced the highest P/L as opposed to other algoritms which set predefined profit targets. The losses were roughly the same in all four trading schemes. As Steve correctly points out, the basic tenet of letting the profits run with a trailing stop is the best method. Losses are unavoidable and are an intrinsic part of trading (and a mere nuisance).

Philippe