Reader question: Sentiment analysis?

Reader Tom asks:

I have been reading your blog for a few weeks and I don’t see any mention of sentiment. I also see you trade futures but you do not talk about Commitment of Traders like on the Euro recently. Is this stuff you use but don’t write about? Not judging if you do just wondered why you don’t talk about it. Thank you! Great blog that has become one of my favs.

Thank you, Tom. I spent many years looking at sentiment data, and the data is very convincing. In fact, one of the strongest relationships you will find is for price to lead sentiment (and not, as people assume, the other way around)! It make sense when you think about it–the public extrapolates trends. When the market (any market) is going up, people feel good about it, and vice versa. There might be value in a beginning trader using sentiment to understand the extremes of market psychology, but, even then, it’s all baked into the price anyway, so why not learn to read it there?

https://openclipart.org/image/300px/svg_to_png/184850/Comedy_Tragedy.pngHowever, this is yet another one of those things that doesn’t work like people think. There is no statistical support for the power of sentiment to predict market moves–objectively speaking, it simply doesn’t work like that. Like many things in stocks, mean reversion rules: if stocks have been going up, they are probably due for a pullback. Sentiment will also be strongly bullish, so people who do not actually look at the data remember this and draw false connections between sentiment extremes and pullbacks. This is another one of those things in technical analysis that I think would die a quick death if more people did actual analysis instead of just using visual relationships of lines on charts. Look at a chart is not analysis.

COT data is even more complicated. Without going into details, it doesn’t show what people think it does, and naive analysis will get you in trouble. Also, you mentioned COT on currencies, and that is something that has always seemed silly in the extreme, to me. The futures are, depending how you calculate it, roughly 1% of the size of the spot market. Do you think that significant “commercial” interests in currencies reveal their hand through COT in currency futures? It would seem to be a serious misunderstanding of market microstructure to assume so.

Bottom line–sentiment is, at best, a coincident indicator, and simply reinforces what we see in price. My continual quest is for simplicity (or, perhaps more properly, parsimony) in trading, so I work hard to eliminate anything that does not add analytical value. The last time I looked at sentiment, aside from a few analytical projects, was 2004, so, no, it’s not something that figures in my analysis in any way.

Thanks for the question!

AdamHGrimes

Adam Grimes has over two decades of experience in the industry as a trader, analyst and system developer. The author of a best-selling trading book, he has traded for his own account, for a top prop firm, and spent several years at the New York Mercantile Exchange. He focuses on the intersection of quantitative analysis and discretionary trading, and has a talent for teaching and helping traders find their own way in the market.

This Post Has 3 Comments

  1. Eric

    I understand and agree with the reference to COT in terms of FOREX. But there are other markets, in particular ag markets, where buyers and sellers usually use futures to hedge their positions, and they have to report this. Some authors have found a correlation between COT data and subsequent performance (over months, not days). Therefore, doesn’t this depend on the type of data we are analyzing?

    1. Adam Grimes

      Good point and valid criticism. I’ll make a small edit to the article. Thank you!

Comments are closed.