Trader Performance 2006 - 2009

[oildaytrader]

Found a spreadsheet with weekly results. This was me, 2003 - 2007 (in total return terms):

....and this is the SP 500:

Apparently, I was tracking the SP 500 quite closely before disaster struck. A bit behind, but not by much, which is fine, since I never risked everything, not by a long shot.
I do remember that once things really began to go south, I stopped keeping this spreadsheet, which was, I am sure, just one of the many stupid things I did in 2008.
OTOH, I did make a ton by accident in a different, more speculative account. The accident was a bit like Mrs. Packletide's Tiger (but without the incidental expenses). I'll describe that in my journal, where I suppose I should introduce my trading history a bit more fully.

How did you do it?Just kept adding to losing positions,averaging down , no stops?

How would it work out in the future , the same type of incidents kept repeating ?

 

[Benton D Struckcheon]

No averaging down, that's just insane. Just small repeated losses, because the system based itself on lognormal returns, as I noted earlier, which also means it assumed mean-reversion, which of course did not happen at all in 2008. Get enough of those together and it adds up to real money.
The account didn't actually blow up, but I stopped trading those systems after Sept of 08, and then merged it into my options account, and concentrated hard on trading GDX and only GDX based on much simpler T/A, and gradually built myself back up again.
In 2009 I finally got around to reading The Black Swan, out of sheer curiousity: I actually didn't know that it was all about what had happened to me. Turned me around completely (what amazed me was that the thing was written before the crisis; it reads like it was written after), and now I just do simple stuff, and of course am using many more stats on my system (or two), not to build them but to track them and make sure I'm not letting things get out of hand.
I also put much much less stock in anything based on the assumption of any sort of normal distribution, which pretty much includes most financial math, unfortunately. (Sharpe and Sortino ratios discussed earlier are both based on the assumption of normal distributions, for instance. It's useful info, but both seriously underestimate real-world risk. My 2003 - 2008 systems had beautiful Sharpe ratios, but of course in the end that meant nothing.)
Interestingly, I'm finding that a pet theory of mine seems to work just now: because of new stuff like Vix options, lots of new volatility indices, the securitization of the exchanges (all the US exchanges are now listed stocks) and the widespread use of ETFs, there's a lot more data now on sentiment. All of this stuff gives you a real-time read on sentiment. I'm finding it very useful, but I'm treading very carefully, since I now know that I have a bad tendency to trust numbers more than I should. It's what I'm going to write extensively about in my journal.

BTW, I should of course have noted that that graph was based on an index of total returns that started at 100 at the beginning of 2003.

 

[oildaytrader]

BTW, I should of course have noted that that graph was based on an index of total returns that started at 100 at the beginning of 2003.

If you had traded purely in the direction of the trend, would you have suffered as much?

You must have heard a thousand times "the trend is your friend,until .........................".Whats wrong with following trends?

Mean reversion systems have poor risk reward ratios.Trending systems have greater r/r ratios.Trend trading systems give me about 50% p a , contrary systems give me zero.

I always lose when I don't follow these rules.

http://www.dacharts.com/articles/_22rulestrading.htm

 

[Benton D Struckcheon]

If you had traded purely in the direction of the trend, would you have suffered as much?

You must have heard a thousand times "the trend is your friend,until .........................".Whats wrong with following trends?

Mean reversion systems have poor risk reward ratios.Trending systems have greater r/r ratios.Trend trading systems give me about 50% p a , contrary systems give me zero.

I always lose when I don't follow these rules.

http://www.dacharts.com/articles/_22rulestrading.htm

I don't know the answer to that. I have tested all kinds of things, and from what I can see, markets do tend to reverse more than they tend to trend. Basically, underlying it all is the fact that if a trend gets going, more people will hop on until, finally, there are just too many people on one side of the trade, at which point it reverses. Sideways markets do this very frequently and so get stuck in a range, trending markets, or what we call trending markets, have lower frequency and greater amplitude.
My current investigations are based on this idea. There used to be an old guy on the old FNN (got merged into CNBC) who tracked sentiment via what was available back then: traders polls. His idea was that when his measure went over 50, he got on until it got to an extreme.
One good example: look at this data from the CBOE on 12/4, a day when gold cratered: very high options volume, a record as a matter of fact, that was twice as high, almost, as the previous, with a very high ratio of calls to puts. I look at both: to me, high volume is a measure of bullish sentiment, because no one has to be in gold, after all, and a high ratio of calls to puts on a high volume day is twice as significant.
That was a sentiment extreme, and it told you that the decline that day was the beginning of a sustained down move in gold. I covered that short on Thursday, and so wound up losing money on Thursday and Friday, so as of right now it looks like the new long trade won't be making anything, but we'll see. That short was a nice one though.