Guessing Games
A fantasy football enthusiast considers modes of prediction and the fun of imagination.
By Caleb Gordon
I’ve been playing fantasy football for a long time. Even while I played high school football, the mental and conceptual side of the game appealed to me just as much as physical participation. Football is an extremely complex game, and even at its lower levels requires numerous auxiliary staff to gather and apply information, develop play designs, and make decisions based on the talent and tendencies of both their own players and their opponents. While there aren’t many ways to continue playing the game once school teams are no longer an option, and coaching can sometimes require even more time and effort than physical play, fantasy football offers a way of engaging with the game which feels continuous with its strategic essence without the demands that come with being part of an actual team.
For those of you who don’t know how fantasy football works: a group of people (usually numbering between about ten and twelve) take turns selecting NFL players to form their own “teams,” which they then manage over the course of the season by continuing to add and subtract players and choose which of their players they want to represent their team each week. Teams usually feature a “lineup” of all the offensive positions which record statistics related to the movement of the ball. You’ll have a quarterback or two, several running backs and wide receivers, a tight end or two, often a kicker, and to round it out you often choose a team’s entire defense to represent your own squad. If the positions don’t mean anything to you, that’s ok – it’s not hugely important. The point is that you gather a collection of players which basically mirrors the form of the teams which play each other every week. Those players are assigned points based on the official statistics they compile, and so the competitive aspect of fantasy football is in trying to collect players who will be statistically prolific over the course of the season while also choosing the right players for each week’s matchup.
A lot can change from week to week. Some players start every week because they have established reputations and consistent performance. Other players might be expected to score a lot of points in one week, while not being expected to maintain the same level of production for others due to changing circumstances. And sometimes fantasy managers have to take a shot in the dark with an unknown player due to the unavailability of their preferred players. In one sense, fantasy football involves a lot of math (How many opportunities are available? How likely is a player to be granted opportunities? What is he likely to do with them?). However, a lot of that math – nearly all of it, really – is speculative. And therein lies the fun: the thoughtful manager of a fantasy team will have an idea about the range of statistical production they expect from their lineup, but each week as the games play out, managers experience their teams as a mixture of both met and unmet hope. You have to believe in a player enough to start him instead of someone else, but it’s hard not to be conscious of your own imagination. You want him to have a good game; you can see it happening, even if it seems unlikely. Every week you’ll be both right and wrong, inevitably, about various players. You hope to be more right than wrong, as a good manager will be. And if the fun is the weekly revelation, the skill is in that balance.
Just like an NFL general manager, someone managing a fantasy football team wins by combining an eye for talent, faith in development, and a large helping of luck. In many ways it’s a competition of insight, verging almost on prophecy.
Fantasy leagues are a unique way of interacting with sports. There are two main kinds: leagues which form entirely new teams each year, and “dynasty” leagues where you can keep your entire team as long as you want. Both ways of playing require you to predict how your players will perform over the course of each season, but whereas standard leagues only require you to predict one season at a time, dynasty leagues require you to think years in advance: you have to think about how players are going to develop, how good they can become, what kinds of opportunities they are likely to receive, and how long they can remain effective or in favor. One of the most appealing elements of these leagues is how closely you can role-play the decision-making aspect of pro football. Just like an NFL general manager, someone managing a fantasy football team wins by combining an eye for talent, faith in development, and a large helping of luck. In many ways it’s a competition of insight, verging almost on prophecy. A depth of knowledge helps, but its depth usually cannot be merely factual; you are not just projecting the development of individuals, but the development of entire teams. You need to know how the individual factors into his own competitive context. You are projecting the strategies employed by coaches, contingencies which lead to opportunities, consistency over time, and sometimes deciding whether to take a chance on explosive potential while accepting the risk of getting nothing at all. Sometimes you just have a feeling. But at the end of the process, it always is a game about judgement, imagination, and prediction. If you are good at fantasy football, it usually means you are good at predicting the future.
There is enough skill involved that it is easy to be bad but harder to be good. Luck is a major element, but success over time indicates a reliable eye for identifying rising talent, accurate assessment of your own league’s economy, and understanding of the sport’s off-field dynamics such as salary considerations and strategy trends.
Responses to my article revealed other ways of playing the same game.
One of the most interesting – and impactful – ways that fantasy football has changed since I started playing is the exploding popularity of computer modelling and “advanced statistics.” As the internet has matured, we’ve also seen exploding participation in various online communities who exist to discuss and argue about player projections and values. Much of this conversation is now dominated by references to computer models and advanced statistics. There is certainly a place for this in the role-playing aspect of managing a team; real teams use statistical models, too. Is it more important to be bigger or faster? How consistently does early success portend future excellence? Such questions can generate fairly consistent recommendations when the data is collated and mapped. But I’m still surprised when I read people who make arguments – often very detailed arguments – about players they’ve never watched. Personally, a lot of the enjoyment I get from fantasy football is related to my enjoyment of watching the game. I spend a lot of time watching players at lower levels and trying to determine what I think of their professional prospects. Which traits will carry over as the competition intensifies and the quality increases? Which individuals have the drive, intelligence, and toughness to succeed as the sport becomes more difficult? Who has room to develop and grow – and how much?
I moonlight pseudonymously as a fantasy football writer on various internet forums, and in a recent article I presented a variety of points to suggest that a relatively obscure player might exceed a lot of expectations. I found myself surprised at a number of commenters who seemed to miss the point of the article entirely. Essentially, I had offered an attempt at identifying an exception to the rules of mathematical modelling; I had presented a collection of reasons why a player whom the models dislike could end up being a valuable contributor anyway. “Nah,” said a number of commenters, “The models don’t like him.” These responses made me laugh a bit – I know the models don’t like him. That’s why I wrote the article. But I also found these responses strange in the way that they revealed other ways of playing the same game; these responses revealed other sources of enjoyment (I assume) that have little to do with those managers’ imaginative engagement with the players who they choose for their team. These players, seemingly, primarily exist as numbers as far as they are concerned. I can’t help but wonder to myself, But doesn’t your enjoyment of fantasy football stem from your love of the game itself?
Some of this may be explained by the rise of fantasy football as a platform for low-to-medium-stakes gambling. Managers often pay fees for each league they’re in, and at the end of the season will collect winnings if they finish at or near the top of their league(s). Many people manage multiple teams – some, I’ve heard, even hundreds – and for those individuals, the maximum fun is, I expect, winning (money). When you play this way, the priority of mathematical modelling makes more sense: the whole point is to maximize your chances at success, and when you are not personally invested in your team there is little reason to take chances on outliers.
I’m content for my team to be worse if that’s a function of my preference to play the game in a more imaginative way.
I only manage one team. I’ve toyed with the idea of adding another, but I decided I actually prefer keeping all of my eggs in one basket. I like that my failures cannot be offset by additional chances, that every year I fail to win is another year I have to wait, and that when I guess right on a player that others have ignored, that extension of faith is realized through my entire experience of the game. One big hit is not just a local event, a potentially interesting piece of statistical noise, but the continuing evolution of my team. I believed in that player, and when that belief becomes justified – that’s fun.
I think part of what I’m actually saying is that I’m content for my team to be worse if that’s a function of my preference to play the game in a more imaginative way. I don’t just want to make the statistically most-defensible choices; I like to find my own players and attempt to identify my own exceptions. Part of this is because I assume the managers I’m competing against are also relying on computer modelling. To the degree that I believe the models will always have exceptions, you could probably say that I’m trying to find my own competitive edge by cutting across the grain and pitting my own intuition against the best projections that computers can generate. I suppose I have a fairly deep conviction that the exceptions to the models aren’t random – that they can be spotted. The exceptions are not out of the blue, without any evidence at all, but signified in ways that the models cannot capture and use. And – to a degree – I believe that I can spot them. Whether this is true is difficult to say. My teams tend to be better than average, but it’s been awhile since I last won a championship.
Football games are played by humans, and humans are, of course, not predictable. But also of course, they are predictable. It’s that blurriness that makes fantasy football so interesting to me. I know that I don’t know. And yet . . . I think I can see something up ahead. It’s not always what the maps are pointing to, but the maps are based on the past–they change as we arrive.
Caleb Gordon is a PhD candidate in theology at the University of Manchester. He is originally from Alaska, but currently resides in the city of Durham (UK). When he’s not exploring football-related esoterica, he researches and writes on the confluence of theology, aesthetics, and environmental ethics. He also teaches at Lindisfarne College of Theology.