Let’s say you are trying to make a tough decision, you know like everyone did in their life. You've got loads of information at your fingertips, but how do you know what's most important? Should you spend hours analyzing every detail, using all the information, flooding your brain with the information or should you trust your gut and take a leap of faith?
It turns out this is a classic problem that experts have been studying for years. Their findings might surprise you.
You might think that more information is always better, I once felt the same. But that's not necessarily true. In fact, having too much information can actually lead to worse decisions and overconfidence in your abilities or simply just make your head hurt.
Let's look at a study where 25 experienced bookmakers were asked to predict the top five horses in 45 races. The bookmakers were given a list of 88 variables commonly found on a past performance chart of a racehorse, and they had to rank the importance of each one. Then, they were given past data on the races in increments of 5, 10, 20, and 40 variables, which they had previously selected as the most important.
What did the study find? Well, when the bookmakers only had five pieces of information, their accuracy and confidence were closely related. But as they received more information, their accuracy plateaued, and their confidence skyrocketed.
With 40 pieces of information, the bookmakers' confidence was over 30%, even though their accuracy remained the same. In other words, more information doesn't lead to more accuracy; it just leads to higher overconfidence.
A similar study looked at the ability of college football fans to predict the outcomes of 15 NCAA games. Participants had to demonstrate their knowledge of football before the study, and they were given a range of statistics, such as fumbles, turnover margin, and yards gained, to help them make their predictions.
The computer model was given the same data to see if more information would lead to better predictions.
So how did it go? The computer model's accuracy increased as more information was added, but the human experts' accuracy did not improve with more information. In fact, their accuracy remained about the same, regardless of whether they had six or 30 pieces of information. But just like the bookmakers, their confidence increased with the amount of information available, even though it didn't actually make them more accurate.
Related to stock analysis, a study was conducted where financial analysts were given the task to forecast fourth-quarter earnings in 45 cases. The information was presented in three different formats.
The first format consisted of the past three quarters of EPS, net sales, and stock price, which is the baseline data.
The second format included baseline data plus redundant or irrelevant information
The third format included baseline data plus non redundant information that should have improved forecasting ability, such as the fact that the dividend was increased.
The analysts were asked to provide their forecast and their confidence in their forecast.
Interestingly, both the redundant and nonredundant information significantly increased the forecast error, meaning that more information did not lead to better accuracy.
However, the analysts' self-reported confidence ratings for each of their forecasts increased significantly with the amount of information available. This suggests that more information did not help the analysts make better forecasts, but it did make them more overconfident in their predictions.
So what does all this mean? Well, it suggests that sometimes, less is more. When it comes to decision-making in trading or investing, it's important to consider the quality of the information you have, not just the quantity.
This reminds me of Joel Greenblatt, a prominent American investor and hedge fund manager, who has shown that when it comes to picking stocks, less is often more.
In fact, Greenblatt's strategy is refreshingly simple: he focuses on only two metrics - return on capital employed (ROCE) and earnings yield - to identify undervalued companies that have the potential to deliver strong returns.
While this may seem like an overly simplistic approach in today's world of big data and complex algorithms, Greenblatt's track record speaks for itself.
His investment firm, Gotham Capital, reportedly generated an average annualized return of 40% from 1985 to 2005, a remarkable feat that many attribute to his disciplined use of these two key metrics.
In a world where we are bombarded with endless amounts of data and information, it's refreshing to see that sometimes the simplest approach can be the most effective.
It turns out this is a classic problem that experts have been studying for years. Their findings might surprise you.
You might think that more information is always better, I once felt the same. But that's not necessarily true. In fact, having too much information can actually lead to worse decisions and overconfidence in your abilities or simply just make your head hurt.
Let's look at a study where 25 experienced bookmakers were asked to predict the top five horses in 45 races. The bookmakers were given a list of 88 variables commonly found on a past performance chart of a racehorse, and they had to rank the importance of each one. Then, they were given past data on the races in increments of 5, 10, 20, and 40 variables, which they had previously selected as the most important.
What did the study find? Well, when the bookmakers only had five pieces of information, their accuracy and confidence were closely related. But as they received more information, their accuracy plateaued, and their confidence skyrocketed.
With 40 pieces of information, the bookmakers' confidence was over 30%, even though their accuracy remained the same. In other words, more information doesn't lead to more accuracy; it just leads to higher overconfidence.
A similar study looked at the ability of college football fans to predict the outcomes of 15 NCAA games. Participants had to demonstrate their knowledge of football before the study, and they were given a range of statistics, such as fumbles, turnover margin, and yards gained, to help them make their predictions.
The computer model was given the same data to see if more information would lead to better predictions.
So how did it go? The computer model's accuracy increased as more information was added, but the human experts' accuracy did not improve with more information. In fact, their accuracy remained about the same, regardless of whether they had six or 30 pieces of information. But just like the bookmakers, their confidence increased with the amount of information available, even though it didn't actually make them more accurate.
Related to stock analysis, a study was conducted where financial analysts were given the task to forecast fourth-quarter earnings in 45 cases. The information was presented in three different formats.
The first format consisted of the past three quarters of EPS, net sales, and stock price, which is the baseline data.
The second format included baseline data plus redundant or irrelevant information
The third format included baseline data plus non redundant information that should have improved forecasting ability, such as the fact that the dividend was increased.
The analysts were asked to provide their forecast and their confidence in their forecast.
Interestingly, both the redundant and nonredundant information significantly increased the forecast error, meaning that more information did not lead to better accuracy.
However, the analysts' self-reported confidence ratings for each of their forecasts increased significantly with the amount of information available. This suggests that more information did not help the analysts make better forecasts, but it did make them more overconfident in their predictions.
So what does all this mean? Well, it suggests that sometimes, less is more. When it comes to decision-making in trading or investing, it's important to consider the quality of the information you have, not just the quantity.
This reminds me of Joel Greenblatt, a prominent American investor and hedge fund manager, who has shown that when it comes to picking stocks, less is often more.
In fact, Greenblatt's strategy is refreshingly simple: he focuses on only two metrics - return on capital employed (ROCE) and earnings yield - to identify undervalued companies that have the potential to deliver strong returns.
While this may seem like an overly simplistic approach in today's world of big data and complex algorithms, Greenblatt's track record speaks for itself.
His investment firm, Gotham Capital, reportedly generated an average annualized return of 40% from 1985 to 2005, a remarkable feat that many attribute to his disciplined use of these two key metrics.
In a world where we are bombarded with endless amounts of data and information, it's refreshing to see that sometimes the simplest approach can be the most effective.
相關出版品
免責聲明
這些資訊和出版物並不意味著也不構成TradingView提供或認可的金融、投資、交易或其他類型的意見或建議。請在使用條款閱讀更多資訊。