Dyer Final Project – A Data-Driven Analysis of the 2015-16 Fantasy Football Season

Introduction

The fantasy football industry had a humble beginning. After starting out as a game played by few, with pencil and paper, fantasy football has evolved into a $70 billion behemoth. Fantasy football can be played on numerous websites with a large number of different formats, and tens of thousands of leagues feature any number of teams throughout the Internet.

For the casual or even diehard fan, putting together a championship winning fantasy football team is far from straight-forward. An infinite number of variables are at play, from the second you draft your team to the final substitution you make before the championship weekend. As you’ll find out, mock drafts and preseason predictions can only get you so far.

This article will provide an in-depth, data-driven analysis of the 2015-16 fantasy football season. All of the data used in this article is from ESPN.com, and all of the numbers used are for leagues that follow ESPN’s default league setup and scoring system. Unless otherwise noted, the data used and represented comes from ESPN’s top 200 fantasy players for the past season.

Who should you draft with your first pick? If you already have two running backs, when should you pick a third? When should you draft your starting quarterback? Who is Gary Barnidge, and what makes him so special? These questions plague nearly every fantasy football manager, and can create headaches when it comes to piecing together your championship winning team. But after reading this article you will be more than prepared to take home the crown next season.

The Draft

Perhaps the most crucial aspect of the fantasy football season is the draft. The internet provides users with countless mock drafts and predictions from a myriad of writers and websites. For the most part, player projections are pretty much shots in the dark, but we will dive into that later on. Right now, we’re focused on analyzing the fantasy football draft from a data stand point.

The following chart will offer insight into how ESPN’s fantasy football drafts played out. The points from the top 200 players during the 2015-16 season are plotted against the average place that they were selected in the drafts of all ESPN fantasy football managers.

We’re able to identify several notable and interesting trends looking at this information.

  • First, the outliers. The two primary outliers that immediately jump off the screen are Cleveland Browns’ tight end Gary Barnidge (149 fantasy points), who was selected, on average, 170th in ESPN drafts, and Carolina Panthers’ quarterback Cam Newton, who had the most fantasy points of any available player with 373. Strange, you might wonder, that Newton’s average selection in ESPN fantasy drafts was 89th. We’ll analyze that later, but for now, on the the topic of quarterbacks…
  • It is interesting to note the trend of points versus draft position when it comes to quarterbacks; several quarterbacks are drafted fairly early on average (those being Aaron Rodgers, Andrew Luck, Tom Brady, and Russell Wilson), but the majority of the top quarterbacks are drafted anywhere between 90th and 120th on average. Why draft a quarterback, who is going to most likely give you the most fantasy points, in such a deep round? The simple answer is a) the quarterback position is generally a pretty deep one, and b) taking a shot at a running back or wide receiver early is probably the smarter option, mostly because you start two of them every week. We’ll look a little closer at this question later, but while we’re on the topic of running backs and wide receivers…
  • It is clear to see the general decline in points for running backs and wide receivers as drafts progress. You can distinctly see that running backs are the most commonly selected position early in drafts, followed closely by wide receivers. Why this trend, you might ask? You can get a fairly firm grasp of why simply by looking at the graph; despite having more available, the wide receiver and running back positions are not nearly as deep as the quarterback position. Think about this: just six wide receivers and only two running backs had more than 200 points during the 2015-16 season, while the average quarterback put up 224.2 points. That statistic alone can explain the influx of running backs and wide receivers taken early in drafts. There are few surefire picks at those positions, so your best bet for selecting a reliable one is to draft them early and often.
  • The tight end position is one of the more interesting ones when it comes to the draft. Tight ends put up the lowest average points (87.5) of any position, but having a good one could make or break your season. Other than New England Patriots’ tight end and freak of nature Rob Gronkowski (176 fantasy points), no tight ends were taken (on average) before the 30th pick and just four broke the 100-point mark last season, which reveals how hit-or-miss the position is when it comes to the draft. Outside of Gronkowski the majority of tight ends are taken between the 100th and 130th pick in the draft, and put up relatively similar numbers.
  • Finally, we’ll take a loot at kickers and defense/special teams units. Kickers are similar to quarterbacks in the sense that the field is fairly deep, and most kickers are fairly reliable. Aside from Philadelphia Eagles’ kicker Cody Parkey (who was suffered a season-ending injury early in the season) and Atlanta Falcons kicker Matt Bryant (who had just one field goal attempt after Week 10), no kicker in the top 200 had fewer than 100 points. Fantasy managers will generally carry just one kicker, and most kickers are pretty dependable, so taking a shot on one late in the draft makes sense. Defense/special teams units follow a fairly similar drafting trend as kickers do, despite being a very different fantasy position. The Seattle Seahawks defensive unit was the highest selected at the position, being taken 48th overall on average. Outside of that, however, most units are taken anywhere between 100th and 130th overall (sound familiar?). My guess: it’s much harder to predict how a teams’ defense will perform given the number of variable – injuries, strength of schedule, and the number of players involved – so, taking a defense/special teams unit early in the draft could easily backfire.

The Shakedown

So now that we have looked at how fantasy drafts generally shake out, let’s take a deeper look at the why this is. Seeing how the individual positions stack up against each other as a whole can help paint a better picture of how and why players are drafted where they are. The following graph charts the average number of points for every fantasy position (note: because of the lack of kickers and defense/special teams units in the top 200, the numbers for those positions are for all players, not just from the top 200 like the other four positions).

As was mentioned earlier, quarterbacks lead the field by a country mile. This helps to illustrate just how much deeper the quarterback position is, and further explains why quarterbacks are generally taken in later rounds of the draft; when the average quarterback puts up greater than 100 points more than the average wide receiver or running back, taking one early in the draft could be unnecessary.

Moving down the list to running backs, it’s easy to see that the position is not nearly as deep as quarterback. This explains why running backs are selected earlier in drafts compared to other positions – top-tier running backs are tough to find, and the position simply isn’t reliable. Just two running backs put up more than 200 points last season (Adrian Peterson of the Minnesota Vikings and Devonta Freeman of the Atlanta Falcons).

The same can be said of wide receivers to a degree. Out of the top 200 fantasy players that we are analyzing, wide receivers comprise the majority of the bunch, followed closely by running backs. The explanation for that is simple: starting lineups in standard leagues feature two running backs and wide receivers. Wide receivers put up, on average, roughly 30 more points than running backs, which could explain why running backs are a more valuable commodity (we’ll take a deeper look at this later on).

Finally, tight ends, defense/special teams, and kickers all put up fairly similar numbers, and similar numbers compared to running backs and wide receivers. This leads us to an interesting question: what makes certain positions more valuable than others when the average player at each position puts up fairly similar statistics? And why would it be pertinent to draft a certain position before another one? This certainly isn’t a straightforward question, and most fantasy football managers have their own strategy when it comes to drafting, but looking at preseason projections could provide greater insight into the conundrum.

Why? 

Drafting a solid fantasy football team is a difficult task. For the most part, the team you start the season with is hardly ever the team you finish with. Much of this can be attributed (or blamed, depending on how you look at it) on preseason predictions.

Unless you are already a fantasy football guru, equipped with your own predictions and metrics, ESPN’ preseason projections and rankings are the only tool managers have when it comes to the draft. Unfortunately, these predictions are hardly a science, despite how knowledgeable ESPN’s experts may be. Let’s take a look at the quarterback position for example. In the following chart, the gray bars represent a quarterback’s projected fantasy points for the 2015-16 season according to ESPN, while the blue bars represent the actual number of points they accumulated.

As you can see, the only consistency when it comes to predictions versus outcomes is the inconsistency. In some cases, the predictions are dead-on, a la Drew Brees (2.9-point difference between projected and actual points) and Eli Manning (8.2-point difference). However, in other cases the predictions were not nearly as accurate, examples being Cam Newton (100.4-point difference) and Carson Palmer (52.5-point difference).

So does this mean that we should abandon ESPN’s preseason projections when it’s time to draft a fantasy team? Of course not. But it is clear that those “expert” predictions may not be as accurate as they are made out to be.

A similar analysis can be made of the other positions, leading us to the conclusion that, although preseason predictions can be a valuable tool, they are by no means a science. To illustrate this point further simply look at the following graph, which depicts the average difference between ESPN’s projected points and actual points for each position.

Projected Points vs. Actual Points Differential for 2015-’16 Season (ESPN)
Create bar charts

It is interesting to note how off ESPN’s quarterback predictions were for last season. What does this say about ESPN and quarterbacks in general? I’m not entirely sure. However, what this chart does show is that, although these predictions are helpful when it comes to drafting a team, they definitely are not exact, which just goes to show how difficult it is to draft a team that plays the way that you expect it to.

What Does it all Mean?

So, at the end of the day, what does all of this mean to you, the fantasy football manager? Looking at the data and visuals provided you can draw several conclusions when it comes to drafting players, projections, and the like, but here are several key takeaways from the information provided:

  • Take a shot at a top-tier running back or wide receiver early in the draft. Few players at those positions are particularly reliable, so if you have a chance to take star, do it.
  • Quarterbacks are pretty reliable and the talent pool is deep, so you can wait to take one in a later round.
  • Tight ends are pretty hit or miss, and outside of Rob Gronkowski cannot compete when it comes to points with running backs and wide receivers. You’ll have to do a little research and have a fair amount of luck to draft a solid one.
  • Kickers are fairly dispensable, and are practically a dime a dozen.
  • Defense/Special Teams units are going to be difficult to predict. If you want to get really in-depth, look at the strength of schedule for various defensive units.

Like I said earlier, rarely will your fantasy team look the same at the end of the season as it did in the beginning. Injuries, depth charts, strength of schedules, and a myriad of other variables factor into a football season, and all can impact the fantasy production of your team’s players. So do your research, take a few chances, and cross your fingers.

If you want to see a glimpse of the statistics that drove the data for this article, the table below allows you to sort through the top 50 fantasy players and the relevant statistics used.

Sources:

  • http://games.espn.go.com/ffl/livedraftresults
  • http://games.espn.go.com/ffl/tools/projections?display=alt
  • http://games.espn.go.com/ffl/leaders