Cognitive biases are errors in judgement that influence how people think and act. The human brain is incredible, but it’s not perfect. Unfortunately, it’s a lot harder for people to process information in a logical, objective manner than they would like, particularly when dealing with numbers and probabilities. Cognitive biases play a big role in daily fantasy sports contests, especially when it comes to GPP ownership. Whilst this article just gives a brief overview of some of the most common cognitive biases, understanding how these biases affect decision making can be extremely beneficial to DFS players. After reading the article, you will hopefully be more aware of the ways in which your opponents make errors in judgement and be less susceptible to similar errors yourself.
The ambiguity effect is a bias where a lack of information influences decision making. A simple example of this in DFS would be if you were choosing between a player that had completely average stats across the board and one that you had no data on. In this situation you would still be more likely to select the player you already knew about, even though the two players are essentially equivalent. Without any data on a player you should assume initially that they have average stats, and there is also a chance that they are above average. In this case you should be focusing on factors like projected ownership, salary and correlation rather than the player stats themselves. Given that DFS sites account for past performance when setting player salaries for upcoming contests, it’s likely that the unknown player will be cheaper than the known player. You’ll also typically see lower ownership on unknown players given that your opponents are susceptible to the ambiguity effect themselves.
Anchoring is the tendency to rely too heavily on initial information, using it to make subsequent decisions. In DFS you might look at a player’s salary and see that he’s much cheaper than he was in a previous week, so think he’s a fantastic bargain. This could lead you to disregard the fact that there are good reasons for the discount, e.g. he’s in a much worse situation this week or his performances never warranted the higher salary to begin with. Another example might be assigning too much weight to player transfer fees when assessing their talent. If a club brings in a new record signing, that doesn’t mean he is going to be a great fantasy option. In a lot of cases he probably isn’t worth the money the club spent on him anyway!
People tend to place greater weight to information that comes to mind easily over information that’s hard to remember or unknown. Recent fantasy performances are easy to remember, so people tend to rely on them a lot more than long term trends, which can be a huge mistake. If there’s no obvious role change or reason for an uptick in performance, it often doesn’t make sense to heavily adjust your expectations about a player. Another example in football relates to set piece takers; if you just learned that a player has claimed set piece duty for his team, you might focus on that piece of information instead of the bigger picture that he does little else for fantasy purposes.
Everyone has probably heard of this one already, but it’s still easy to be susceptible to it. The more people that start to believe something, the more other people also adopt that belief, even if there is little to no evidence to support it. This is very common in DFS, and can influence ownership significantly. For example, if a player is heavily recommended throughout the week his ownership will rise, regardless of whether he is actually a good option or not.
People that fall victim to the belief bias tend to judge the strength of an argument based on how much their prior beliefs align with its conclusion, rather than the supporting evidence behind that conclusion. If all your research and analysis suggest that a certain player is a great option, but you already dislike the player, you will be less willing to accept the evidence and might still be reluctant to put him into your lineups, and vice versa. Gut feelings can be valuable, but relying solely on them to make DFS lineups is a losing strategy more often than not.
Bias Blind Spot
Everyone tends to think they are less biased than others. After reading this article you should be more aware of some of the common cognitive biases than most DFS players, but try not to assume you will be less susceptible to these biases in the future.
People tend to disregard the errors or faults associated with choices that they made and overweight the errors or faults associated with the alternative options. This is very common in DFS, and emphasises the importance of properly reviewing lineups in an objective manner. Instead of just saying ‘my player was a better choice than his player, he got lucky’, try to understand all the reasons behind their choice and consider as much information as possible when assessing your decision making.
The tendency to focus on information that confirms what you already believe, and disregard evidence against that belief. In DFS you might still use a player that you liked at first glance, even if all your subsequent research shows that they aren’t actually a very good option. This effect can also lead you to still believe misinformation even after you learn new evidence, for example if you continue to use a player that was simply just on a hot streak at the start of the season.
The tendency to underweight new evidence in comparison to an existing belief. In daily fantasy football this can happen quite frequently. Team news is usually announced one hour before kick-off, which means you have limited time to reassess your lineups before the slate starts. The team news can significantly affect the players that you should be using in your lineups, but you will often be reluctant to change them anyway. It’s important to consider all relevant information when making fantasy teams, even if it means changing your lineup just before the slate starts. Another example is the failure to take advantage of late-swap; the current score of your own/your opponent’s lineups from earlier games is extremely important information to consider when deciding which late-game players to use.
If something is familiar, you might naturally develop a misguided preference for it over unfamiliar alternatives. An example of this in DFS might be entering the same contests each week even if you should be moving up in stakes, or aren’t actually very profitable in one or more of those contests. Another example might be choosing one player over another even if they are a worse option, just because you’ve used them in recent weeks and had success with it. Also, if you’re a fan of a certain team you might be more inclined to use players from that team in DFS, and vice versa.
People sometimes tend to react differently to identical information depending on how it is presented, i.e. if the information is presented in a negative way they will react more negatively to it, and vice versa. If someone tells you that ‘player A is only averaging 9 fantasy points per game’ whereas player B is averaging at least 9.5 fantasy points per game’, you might look at player B in a more favourable light than you should even though there isn’t a huge difference between the two.
The tendency to see something as having been predictable after it has happened, even if there was little previous evidence to help predict it. If a player has a huge fantasy day, it can be easy to be annoyed with yourself for not using that player even if it was hard to expect a great performance beforehand.
Hot-Hand (Foot) Fallacy
The belief that past success will lead to greater probability of further success in the future, even if the events are random. The most obvious example of this in football is a hot streak in front of goal; people tend to assume that the player is more likely to continue scoring because they are full of confidence or have somehow improved their finishing, when in reality streaks will naturally occur in small sample sizes even if they ‘appear’ to be non-random.
The tendency to perceive correlation between variables even when they are unrelated or only weakly correlated. An example of this in football might be the assumption that a player will score lots of goals because he shoots frequently, whereas in fact shot quality (i.e. xG/Sh) is much more important than shot volume.
Illusion of Control
The tendency for people to think they have much more control over an event than they actually do. Ultimately, daily fantasy football is high variance and no matter how much research you do it will be impossible to know exactly what is going to happen. Be careful not to let results influence your ability to think clearly here. It’s easy to feel like you had more ability than you did to predict something that ended up happening, and vice versa. Being tilted isn’t fun, and it’s much more likely to happen if you were expecting a great outcome before the slate had even started.
Illusion of Validity
This is similar to confirmation bias. If data shows a clear pattern you might be overconfident in your ability to make predictions, even if the pattern itself isn’t predictive of future events. When looking at data, try to keep in mind the correlation of each variable. If a number of correlated variables show a pattern it probably shouldn’t be that surprising, but if independent variables show a pattern this might be worth taking note of.
Illusory Truth Effect
The tendency to believe something to be true after hearing it enough times. Whilst DFS articles and podcasts can be helpful, it can also be dangerous to rely on them completely without doing your own research first. If you hear throughout the week that a player isn’t worth using, you might automatically ignore them without looking into the numbers behind it yourself.
More information doesn’t necessarily lead to more optimal decision making. It’s likely that you’ll tweak your lineups multiple times before a slate, but try not to overadjust based on tiny or irrelevant pieces of information. Consider how relevant the information actually is before making any rash changes.
It’s important to develop a consistent process for DFS research and lineup construction, but always keep in mind alternative approaches. As Maslow said, “I suppose it is tempting, if the only tool you have is a hammer, to treat everything as if it were a nail”. Studying lineups from other good players can help with this if you attempt to understand the thought process behind their decisions.
It’s much easier to remember the slates where things went badly to the ones where you ended up having a very profitable day. Therefore, it’s important to constantly review your process and results; zooming out for a more long-term view can help you to avoid becoming overly negative when on a downswing. It’s also easy to get put off using certain players if they have a poor fantasy score on one or more occasions. Past results can be helpful to predict future performance, but it’s important to remember that bad past results won’t necessarily translate to bad results in the future.
Neglect of Probability
A data-heavy approach to DFS makes a lot of sense, but unfortunately it can be far too easy to override the numbers when building DFS lineups. A simple example of this might be using striker A instead of B, even though the numbers say that player B has a greater chance to score. Always try to think in terms of probabilities and the range of outcomes if you get stuck on a decision.
This is a particular instance of confirmation bias where you subconsciously look for information that confirms your initial thoughts on a slate and ignore information that argues against them. Whilst initial thoughts can be valuable, it’s important to not get too attached to prior beliefs after you have done more detailed analysis.
The opposite to negativity bias; believing that you are more likely to experience good outcomes than you actually are. Try not to get overconfident when results are going well, and always keep probabilities in mind when constructing lineups rather than being overconfident that a player will perform well. If you find it easy to build lineups for a slate you might be more likely to believe that you are going to win money that day, and vice versa.
The tendency to overweight recent results compared to the longer-term picture. This is a big reason why DFS tournaments can be hugely profitable. If you’re aware of how recency bias will affect GPP ownership, you should be able to leverage that information to your advantage. Remember that a recent run of great fantasy performances isn’t enough to justify using a player, but many people will treat it as though it is. If a player’s performance level seems unsustainable, you should be looking to fade that player in GPPs to get some leverage on the field. Poor recent performances can often be even more valuable, as these will often result in a player being drastically underowned.
There are many more cognitive biases and logical fallacies that weren’t mentioned here, but hopefully being aware of some of the most common ones will make you think twice before jumping to decisions when building lineups. The key takeaway here is to adopt a data driven approach and look at the numbers before confidently locking anyone in to your fantasy teams.
Thanks for reading! If you enjoyed the article, please don’t forget to share it on social media!