Lots of noise out there. Lets keep it simple. I dont mean to post over Scott, but need this to go out today due to the FDOM (first day of month) stats. Please read Scotts post.
I (volar) just got caught up with vacation, and I always re-read my favorite books on vacation. Many traders look for trading answers, I look for understanding. What do I mean by that? Well I dont read books that tell you “how to make money.” I suggest New Market Wizards- the piece by William Eckhardt is the best written material I know of to this day (next to some of Taleb’s stuff).
I know some of you dont like stats, but if you dont understand true market distribution, well then how can you trade? If you are betting on mean reversion you need a platykurtotic market. Go google it, I am not your mother or hired teacher. If you are a trend trader, you need a leptokutotic market. Now, dont get me wrong, trend traders need edges; they have to be crafty to keep drawdowns low to catch the trend. This is hard, but it is necessary to traders who want to win. For those betting on mean reversion, you may win often (in little bits/pieces), but you will be bankrupt if you dont know how to get out and STAY out. I am sure there is an edge shorting mutual funds that have very consistent returns for that exact reason.
Now lets look at the seasonal distribution of weekly returns (rolling 2 week average) for this time of year.
Go figure- not bullish, or bearish. Just utterly fat tailed.
This means that you should be careful fighting trends- IMO bollinger bands may need to be used inversely this time of year. Or if you are betting on convergence, you need to be very careful with a very tight stop.
Now on to a timely piece of data.
Notice the first day of SEP is highly positive, but has a terrible sharpe. Which means, I dont take my usual FDOM trade.
Bottom Line: the market is deceiving. You need to know why you win and why you do not. As for the FDOM, I dont see that edge in September (even tho there is a high chance we do close higher).
Best of luck trading,