What do you think about “cognitive biases as an edge”?
One story we can tell about the markets and coronavirus is this: It was not hard to come to the conclusion, by mid-to-late February, that a global COVID-19 pandemic was extremely likely, and that it was highly probable it would cause a massive catastrophe in the US. A few people managed to take this evidence seriously enough to trade on it, and made a killing, but the vast majority of the market simply didn’t predict this fairly predictable course of events. Why not? Because it didn’t feel like the sort of thing that could happen. It was too outlandish, too weird.
If we take this story seriously, it makes sense to look for other things like this—cases where the probability of something is being underestimated—because it seems too weird, because it’s too unpleasant to think about, or for some other reason.
For example, metaculus currently estimates something like a 15% chance that Trump loses the election and refuses to concede. I we take that probability seriously, and assume something like that is likely to lead to riots, civil unrest, uncertainty, etc, would it make sense to try and trade on that? On the assumption that this is not priced in, because the possibility of this sort of crisis is not something that the market knows how to take seriously?
Sure that could be an edge, but likely only if it doesn’t happen too often. E.g. people are often averse to exiting their losing trades. I think in most markets an Intuitive trader wouldn’t find an easy way to profit on that because Algorithmic traders already took care of that.
What do you think about “cognitive biases as an edge”?
One story we can tell about the markets and coronavirus is this: It was not hard to come to the conclusion, by mid-to-late February, that a global COVID-19 pandemic was extremely likely, and that it was highly probable it would cause a massive catastrophe in the US. A few people managed to take this evidence seriously enough to trade on it, and made a killing, but the vast majority of the market simply didn’t predict this fairly predictable course of events. Why not? Because it didn’t feel like the sort of thing that could happen. It was too outlandish, too weird.
If we take this story seriously, it makes sense to look for other things like this—cases where the probability of something is being underestimated—because it seems too weird, because it’s too unpleasant to think about, or for some other reason.
For example, metaculus currently estimates something like a 15% chance that Trump loses the election and refuses to concede. I we take that probability seriously, and assume something like that is likely to lead to riots, civil unrest, uncertainty, etc, would it make sense to try and trade on that? On the assumption that this is not priced in, because the possibility of this sort of crisis is not something that the market knows how to take seriously?
Sure that could be an edge, but likely only if it doesn’t happen too often. E.g. people are often averse to exiting their losing trades. I think in most markets an Intuitive trader wouldn’t find an easy way to profit on that because Algorithmic traders already took care of that.
Right, that’s a good point.