What Can Standing Ovations Teach Us About Fantasy Sports?
For starters, we can learn valuable lessons about the power of perception. They can improve our understanding of peer effects and help us predict the behavior of others. Most importantly a standing ovation can be used as a simple model to think more logically about the decisions we make as fantasy owners.
Let’s pretend for a minute that you’re going to see an opera. It doesn’t matter whether or not you have an appreciation for cultural arts – you have nothing else on your agenda and you got the tickets for free (work with me here).
(Besides the fact that it’s finally over)…Did the production exceed your expectations? How many people are applauding around you? Are you with a group of your peers that decided to give a standing ovation?
Think about it.
Depending on the individual, there could be any number of variables that go into making such a simple decision.
The problem is that everyone attending the opera has the ability to make their own choice, however there aren’t any definitive guidelines on whether standing is appropriate. Therefore, on some level, each patron’s decision making process is being influenced by the behavior of the people around them.
This is referred to as the peer effect. Fantasy owners feel it when they make decisions based on the informational influence of others. The information they get could come from anywhere – ESPN, PigskinBoss.com, or their local butcher; the point is that they’re using input signals from a network of peers to establish their own threshold for taking a particular action.
“In general, the standing ovation model is applied to situations where someone makes a binary decision (deciding between two outcomes) after being socially influenced by people of different degrees of sophistication who have received information through a network (information is transmitted through a network).“
The Standing Ovation Model applies perfectly to fantasy sports because you’re constantly making decisions under uncertainty. The better you are at acquiring information, interpreting it, and adjusting your strategy, the more likely you are to win your league.
Think about the last time you were organizing your waiver priorities or considering which players to claim through free agency. What criteria did you base your decisions on?
Did you consider the needs of your opponents? How likely were they to go after the same players? Where were they relative to your position on the waiver priority list? How many players of equal caliber were available?
The Standing Ovation Model gives fantasy owners a way of analyzing the answers to these questions. It provides them with a simple set of rules for thinking more logically about each roster move they make.
For example, let’s suppose we’re considering the possibility of adding a new player to our fantasy team. Player A is currently on our roster and Player B is someone we’re looking to claim off waivers or through free agency.
In order to apply the model to this scenario, we’ll first need to make four basic assumptions.
#1 – In terms of total points, if the expected value of Player B (EVb) is greater than the expected value of Player A (EVa), then the acquisition of Player B will have a net gain in expected value or Quality (Q). Simply subtract the total points you expect to earn from Player A from EVb to calculate Q.
Q = EVb – EVa
#2 – Each owner in your league will make their decision to acquire Player B based on a Signal S,
where S = Q + e
e = error, which is different for each owner (I’ll explain this in more detail later on).
In other words, each owner responds to a different signal because owners aren’t basing their choices solely on the expected value or quality of the players.
#3 – Initially, you will acquire Player B if S > Threshold (e.g. the minimum number of total points you must receive in order to justify the acquisition of Player B).
#4 – Then you acquire Player B if more than X% of owners are also interested in the same player.
From these four assumptions we see that if Player B has a higher expected value than Player A, resulting in a higher Quality (Q), more owners will attempt to pick him up. Also, if there is a lower common Threshold (T) amongst owners in your league more teams will look to acquire Player B.
In the Standing Ovation Model, peer effects show that making X% smaller will increase the likelihood of the event because when X% is small, not many people have to stand to create an ovation. X% would be small when people are insecure and would just stand because a few others stood.
In fantasy sports establishing the value of X% becomes a function of supply vs. demand. If there are fewer players of equal caliber to Player B to satisfy the demand of the owners in your fantasy league, X% becomes much smaller.
In either model X% is largely dependent on the informational influence of our peers and our perception of quality. Often times fantasy owners set a low X% because they overestimate the ability of Player B or don’t understand the behavior of their opponents. This increases the likelihood that they act irrationally when making roster adjustments.
This concept goes hand in hand with the last result of our four assumptions. When informational influence leads us to believe that Player B’s production will exceed his expected value (EVb), we’ve increased the variance in e (the error term) and thus we’re more likely to acquire Player B based on the Signal (S).
In our example of the opera, variance can be introduced in several ways. We may see or hear inaccurate reviews before the show or find ourselves sitting next to someone who looks like they know as much about the opera as John Madden knows about football. The opinions and behavior of these so-called experts influences our decision to stand and applaud.
Basically the error term accounts for any possible outside influences that may have changed our perception of the actual event. The variance in e is greater when there is a wide range of diversity in the crowd and the event is multi-dimensional and complicated because people won’t know whether to stand or not.
What’s important to realize is that the Quality, Q, doesn’t always have to be greater than the Threshold, T, for people to initiate the Signal, S.
In the fantasy sports version of this model, variance is function of your perception of Player B’s value relative to their actual output.
Remember that you are dealing with subjective probabilities. How accurately you gauge factors such as injury risk and strength of schedule will determine how much variance you’re actually introducing into the equation.
The goal of applying the Standing Ovation Model to fantasy sports is to become more normative (better at deciding) and more positive (better at predicting the behavior of others). In order to do that we’ll have to assign values to each variable in the formula.
Quality: Q = EVb – EVa
Although subjective, this variable is the easiest to quantify. It gets even easier when you lose Player A to a season ending injury. Simply look at the projected point totals for each player over a given time period and subtract the difference.
The values of EVb and EVa are the same for each owner in a given league or network, where as their threshold and the error will vary depending on their perception of the information they receive.
This is the expected value (in total points) you need to receive in order to justify replacing Player A with Player B. There are a number of factors that you must consider when establishing your threshold. Need, risk tolerance, and the supply of available players, etc. will determine when the Signal, S, is acted upon.
Anytime a fantasy owner’s Threshold (T) is low, they will be more inclined to acquire Player B regardless of how many other owners are interested in that player.
Earlier we acknowledged that the values of EVb and EVa were the same for every owner in a given fantasy league. Essentially e accounts for the probabilities of downside loss and upside profit associated with each option we are considering.
e also accounts for the accuracy of the information individual owners receive from their respective networks. Each owner’s ability to scout for talent, asses injury risk, and predict future outcomes, etc. will establish the level of variance.
Supposing Player B is perceived to have high probability for upside profit, it may be the case that this player is acquired even though he has a Quality (Q) lower than the owner’s Threshold (T). Inaccurate information leads to a higher values of e and trigger the Signal (S) much sooner.
Variance in informational influence is what causes us to label our poor performers as “busts” and our undrafted heroes as “sleepers”.
In reality these terms just describe the unrealistic expectations we set for these players based on the opinions of our peer network.
Fantasy owners will acquire Player B if more than X% of owners are also interested in the same player.
The value you assign to X% depends on the gaps you need to fill on your own roster, the needs of your opponents, and the availability of high quality talent. Here understanding your opponent’s tendencies becomes critical to your success as a fantasy owner.
The standing ovation model can be used to determine whether or not you should use your waiver priority or gamble on obtaining Player B through free agency.
If you want to beat other owners to the punch, remember to lower the value of X% when you perceive the Quality (Q) of Player B to be very high compared to the quality of your existing players.
The End Game
Your goal as a fantasy owner should always be to minimize risk and maximize reward by combining different players into one optimal asset. You accomplish this goal by objectively processing information, adapting to change, and acting decisively. Therefore your ability to understand peer effects and the power of informational influence is critical to your success in fantasy sports.
The next time you have an important decision to make as a fantasy owner, try including the Standing Ovation Model in your thought process. I guarantee you’ll think more logically.