Nassim Talib’s book entitled Fooled by Randomness, is a great read. The central premise is the idea that people mistake randomness for something meaningful. We always want to find an explanation or a cause for something, even if there isn’t one. Markets are inherently noisy. It isn’t always possible to understand “why” a market moved (I admit, I’m probably guilty of this!). Luckily, in trading we don’t need to be right all the time. Let’s say you were serving burgers. If you burger was safe to eat 75% of the time, and unsafe 25%, you would be shut down. Even if it was 90% safe, and 10% not, that basically wouldn’t be good enough. Success in many disciplines requires us to be very close to 100%.
In trading, you’ve “just” got to be right just over 50% of the time. Or if the returns of your strategy are skewed positively (ie. gains are much bigger than losses), you could be right less than 50% of the time. Even though the standards of success are somewhat lower than in other domains, as we have noted, markets are very noisy making your task difficult. Does the noise mean that markets are always random? I would suggest not. Take for example the ECB meeting earlier this week. The fact that there would be a significant amount of market volatility when Mario Draghi started speaking was something anyone with an economic calendar would know. For a vol trader to profit from this though, we’d have to gauge what our expectations are of the “surprise” in volatility against market expectations. However, if our models had no understanding of scheduled market events, the massive spike in volatility on Thursday, would appear to have been totally random. The more time we spend, the more factors we can model to try to explain a bit more of the price action.
The way EUR/USD spot reacted to his various comments, rising when he said something hawkish etc, again were not surprising. Of course, this doesn’t mean we can profit from these moves, unless we are somehow quicker in reacting than other market participants or have a view about what he is going to say. What I’m trying to say is that whilst there is a lot of noise in markets, there are occasions where it appears to have more structure reacting to certain events or behaviours, which we can explain. I’ve simply given a central bank meeting as a very specific example of an event. We can also have other types of events spread over longer periods of time, such as large scale shifts in flows. For currency markets, these flows can have many origins, whether it is central banks managing their reserves, M&A flows, portfolio flows etc.
The key of course is trying to distinguish between the randomness, and factors which can help to explain price actions. Where we have the inklings of what appear to explainable market reactions, is where we can concentrate our forecasting efforts. Our focus needs to be on getting the rough direction of the forecast. We can accept that markets are often random, and provided we can explain at least some of the price action and that might well be good enough for us.