trend following delusion shattered

Status
Not open for further replies.
NickelScalper:

I think the best argument against random success was the return I got in 2002. It's the one I'm most proud of. The markets were down big time, yet I ecked out a positive return, all because the moving averages were showing sell and I was in bonds.

The chart I posted for odmax was 8+ years. That's a pretty long time I think.
 
Thunderdog:

What I posted was anecdotal, yes, but at least for the charts on odmax and iwn it's verifiable that you can use simple trending techniques to beat Buy & Hold pretty handily. Using just about any trading software, you can reproduce the results I got.
 
Quote from postal:

Thunderdog:

What I posted was anecdotal, yes, but at least for the charts on odmax and iwn it's verifiable that you can use simple trending techniques to beat Buy & Hold pretty handily. Using just about any trading software, you can reproduce the results I got.
More...

Postal,

I am not disputing or trivializing your claim. I am on your side. An individual trader's personal trading performance is anecdotal by definition insofar as deriving broad conclusions about trend trading is concerned. It is "hank" with whom I have an issue. He has variable standards that he applies as he finds expedient. He completely disregards your anecdotal evidence, but he began this very thread with anecdotal evidence of his own. Hypocrisy would be much easier to swallow if it were produced in gel-cap form.
 
I always have a problem with this trend stuff. If someone sells me one Dec 2005 Corn contract for $ 2.29 / bushel the transaction carries no trend information. The only information contained in the transaction is price and volume. Possibly there is spread (bid and ask price) information. There is no inherent trend.

I also graph computer generated random numbers on Microsoft Excel. I see lots of trends. The trends are an illusion. Random numbers have no trend.My mind is seeing patterns that are not there.
 
Thunderdog, yes, I knew you were agreeing with me and I thank you; I guess the post about verification was for those who may not have thought it was possible to "see" a trend and trade it successfully and objectively using a purely mechanical system. That's why I posted the charts for odmax and iwn. BTW, if you want to trade a managed mutual fund like odmax but your trading system exceeds the fund family's trading limit, consider hedging with an ETF. For example, you can hedge odmax with EEM by going short EEM when the trend is down in odmax. These two funds are highly correlated. EEM is more difficult to trend trade directly because of its volatility, but it's excellent for hedging due to the higher alpha of odmax.

Hook, I agree with you that a single data point cannot define a trend, any more than your car's GPS will tell you where you're headed or where you've been. A trend by definition requires historical data to put the current data in context, aka a chart. If you really want to get into randomness vs trends, take a look at Mandelbrot's fractal analysis and chaos theory. Cool stuff.

But here's the interesting part about it, at least for me. The best securities for trend trading are those that are relatively illiquid. For example, there is no way IMO to trend trade SPY or QQQQ. Too many traders, too much volume. Any time a trend becomes evident in these two ETFs, the short term traders exploit it.

You can trend trade smallcaps and international, though, because the lack of volume keeps the scalpers away and allows the longer term trends to be exposed.
 
Quote from Hook N. Sinker:

I

I also graph computer generated random numbers on Microsoft Excel. I see lots of trends. The trends are an illusion. Random numbers have no trend.My mind is seeing patterns that are not there.
More...



bingo !


hank :)
 
some more trend nonsense found on the website of the "evangelist of trends"

<b>Trend Following Debate
There is a small minority who vociferously question trend following's validity. They believe that prediction is an absolute:

So, in other words, you trade according to a system the signals of which are not associated with whether price will move both in the right direction and to a certain minimum extent after the trader has committed to a prospective entry? Good luck. You're going to need it.

Discussion Board Post

This is correct. Trend followers can never have knowledge of some minimum extent move after an entry. How could anyone give a minimum?</B>


think about what this is saying...... if one can't determine a minimum extent of a move after entry..... the trend approach, as taught by the above gentleman, ( he is a good writer but seriously misinformed), MAKES ZERO SENSE. why not simply make reasonable random entries, if the trade goes your way--ride it, if it does not, cut it. this makes more sense than saying you are following the trend but dont know if it will continue untill your entry shows you.

since when did "buy and hold" change its name to "trend following" ? the heroes of trend following certainly don't cut their losses, sustaining 50% drawdowns at times--

the religion of trend following is becoming fanatical !



:D
 
Know what you're getting at, but for sake of argument, what's the difference between the above example and any other reason one may have for entering a trade? If you've done all your homework and enter a trade through some method you deem reliable, and it immediately goes against you -- does that give you any more excuse to stay in the trade as opposed to someone who entered at random?
 
Quote from illiquid:

Know what you're getting at, but for sake of argument, what's the difference between the above example and any other reason one may have for entering a trade? If you've done all your homework and enter a trade through some method you deem reliable, and it immediately goes against you -- does that give you any more excuse to stay in the trade as opposed to someone who entered at random?
More...



good point. however, some entry methods can be quantified objectively to an extent. higher highs and higher lows is not one of them.
 
Status
Not open for further replies.
Back
Top