Logical Fallacies and Trading
As I learn this business of trading I have been wondering what logical fallacies ( formal and informal), if any, might have a bearing on trading psychology.
Here are some initial thoughts on a few.
1) Fallacy of conjunctive probability.
Literature shows that humans ( and rats btw) tend to judge the probability of two events occurring together as more likely than those events occurring separately. The classic example is the Jane fallacy.
The probability of two distinct types of events occurring together though is never greater than the probability of them occurring separately.
So for example if there is a probability of bank runs and a probability of the economy crashing then the probability of both bank runs and the economy crashing isn't higher than either the probability of bank runs or the probability of the economy crashing separately. Similarly if I think its likely a stock will rise this week and also that the stock may fall within the month then the probability of it rising next week and crashing next month cant be higher than it rising next week or crashing next month separately.
2) Assumption of regression to the mean.
For any given population P there will be a mean on any given measure. For some sequences of events, like for example coin tosses, an assumption of regression to the mean may well be valid - as a long term assumption ( ignoring shorter term runs). But for trading on price the concept of a mean can have no real meaning other than on perhaps very short term timescales. Price is auto correlated to a degree but also has a fractal nature and also shows clear short, medium and longer term trends and historically a general bias in the upward direction with flat or bearish interludes. Therefore great care is required in making any assumptions about "reversion to the mean" re price.
3) It's bound to happen soon because it hasn't happened for a while.
You could be a bear or a bull I guess and fall into this fallacious way of thinking. Its somewhat related to the previous one I think as its based on assumptions about the statistics of events. But care needs to be taken because events are "lumpy" and may not accord with statistical expectations for various reasons eg changed market conditions etc etc. A classic example of how to reasons fallaciously in this way is the suggestion that one day late in December Charles Dickens announced that he couldn't travel by train anymore that year, "on the grounds that the average annual quota of railroad accidents in Britain had not been filled and therefore further disasters were obviously imminent." Or maybe it was just a very dry year Charles? Follow the price, note the trend and don't rigidly impose your own expectations, even if informed by such stats, on your decisions about where the market is likely going.
4) Cum Hoc, Ergo Propter Hoc and Post Hoc, Ergo Propter Hoc
Fallacies to do with correlation and causation. They are easy to fall into. Just because two things appear correlated in the way they move in time that does not mean there is a causal relationship between them. There has to be a plausible realistic connection to explain any assumed correlation. Also the causation my be distant in the sense that some other factors are influencing both of the things that appear correlated and they are not directly themselves causally related. I suppose a daft example of this would be to assume that because two stock prices move very similarly that one influences the other to follow in some way. Obviously in some cases that might be so, say where one business depends very closely on another. More likely is that both are impacted by some other economic factor. The important point is to examine potential correlations and causal connections with a critical eye.
5) Bandwagon Fallacy / Argumentum ad Populum
Basically arguing a position or holding a position largely because its popular or most people are holding that position. A position - whether in a logical argument or in regard to the logic behind a traded position should stand or fall on its own merits - not whether everyone agrees or doesn't with your conclusion. When BTC was at its peak the popular majority view was that BTC was going to the moon. But that view as it turned out was wrong and it crashed. Be careful not to assume positions simply because they are popular. In fact there may sometimes be grounds to believe that by the time positions become very popular they may be peaking and may reverse. Let the logic of taking a position drive your decisions not what the popular view is.
Additional ideas and comments welcome
Here are some initial thoughts on a few.
1) Fallacy of conjunctive probability.
Literature shows that humans ( and rats btw) tend to judge the probability of two events occurring together as more likely than those events occurring separately. The classic example is the Jane fallacy.
The probability of two distinct types of events occurring together though is never greater than the probability of them occurring separately.
So for example if there is a probability of bank runs and a probability of the economy crashing then the probability of both bank runs and the economy crashing isn't higher than either the probability of bank runs or the probability of the economy crashing separately. Similarly if I think its likely a stock will rise this week and also that the stock may fall within the month then the probability of it rising next week and crashing next month cant be higher than it rising next week or crashing next month separately.
2) Assumption of regression to the mean.
For any given population P there will be a mean on any given measure. For some sequences of events, like for example coin tosses, an assumption of regression to the mean may well be valid - as a long term assumption ( ignoring shorter term runs). But for trading on price the concept of a mean can have no real meaning other than on perhaps very short term timescales. Price is auto correlated to a degree but also has a fractal nature and also shows clear short, medium and longer term trends and historically a general bias in the upward direction with flat or bearish interludes. Therefore great care is required in making any assumptions about "reversion to the mean" re price.
3) It's bound to happen soon because it hasn't happened for a while.
You could be a bear or a bull I guess and fall into this fallacious way of thinking. Its somewhat related to the previous one I think as its based on assumptions about the statistics of events. But care needs to be taken because events are "lumpy" and may not accord with statistical expectations for various reasons eg changed market conditions etc etc. A classic example of how to reasons fallaciously in this way is the suggestion that one day late in December Charles Dickens announced that he couldn't travel by train anymore that year, "on the grounds that the average annual quota of railroad accidents in Britain had not been filled and therefore further disasters were obviously imminent." Or maybe it was just a very dry year Charles? Follow the price, note the trend and don't rigidly impose your own expectations, even if informed by such stats, on your decisions about where the market is likely going.
4) Cum Hoc, Ergo Propter Hoc and Post Hoc, Ergo Propter Hoc
Fallacies to do with correlation and causation. They are easy to fall into. Just because two things appear correlated in the way they move in time that does not mean there is a causal relationship between them. There has to be a plausible realistic connection to explain any assumed correlation. Also the causation my be distant in the sense that some other factors are influencing both of the things that appear correlated and they are not directly themselves causally related. I suppose a daft example of this would be to assume that because two stock prices move very similarly that one influences the other to follow in some way. Obviously in some cases that might be so, say where one business depends very closely on another. More likely is that both are impacted by some other economic factor. The important point is to examine potential correlations and causal connections with a critical eye.
5) Bandwagon Fallacy / Argumentum ad Populum
Basically arguing a position or holding a position largely because its popular or most people are holding that position. A position - whether in a logical argument or in regard to the logic behind a traded position should stand or fall on its own merits - not whether everyone agrees or doesn't with your conclusion. When BTC was at its peak the popular majority view was that BTC was going to the moon. But that view as it turned out was wrong and it crashed. Be careful not to assume positions simply because they are popular. In fact there may sometimes be grounds to believe that by the time positions become very popular they may be peaking and may reverse. Let the logic of taking a position drive your decisions not what the popular view is.
Additional ideas and comments welcome
Disclaimer
The information and publications are not meant to be, and do not constitute, financial, investment, trading, or other types of advice or recommendations supplied or endorsed by TradingView. Read more in the Terms of Use.