This content, written by Colin Zima, was initially posted in Looker Blog on Aug 30, 2017. The content is subject to limited support.
pegged the fantasy football market at over $70 billion dollars, with nearly building teams. Because the game becomes more popular every year, and there are lots of Looker fans and leagues, we decided to analyze the data (because that’s what we do here).
We have a quarterly “Use the Product” day at Looker, where anyone can use the day to do some hacking on a dataset of their choice. For this quarter’s event, I decided to get my hands on some Fantasy Data in anticipation of this year’s draft.
Looking at the data, we learned a lot more than just drafting strategy. This is the first post of a series, where we will dive into a dataset from FantasyData.com, and see what we can – and can’t – learn from the past to use in the new season.
Let’s start by getting our bearings and taking a look at our data set. We’ve got some stats from 2008 through 2016 to look at player and team performance, including draft data from the past few seasons. There are a bunch of different ways to slice this, but we focused mainly on individual players and positions.
Top performing players
First let’s look at an overview of top single season performers from the last 8 years.
At the top of the ranks we have Peyton Manning’s 2013 season. To put that glorious performance into context, not only did it put him at the top of the list, but it did so by a full 14 points. It’s also important to note that he doesn’t appear on the list again, which tells us that one season’s performance is not a guarantee of continued success.
Unsurprisingly, there are more quarterbacks on the list of top single season performers than any other position (8), but what is surprising is the amount of repeats in this list. Aaron Rodgers leads with five appearances followed by Tom Brady and Drew Brees who both have three seasons making the list. When evaluating players to draft, sustained high level performance is a much better predictor of future success.
- While Aaron Rodgers makes the most appearances in the Top Performers list, Drew Brees is actually THE single top point-earning player when we look at performance overall (for our dataset that is from 2008 to 2016).
Comparing the two, Aaron Rodger’s 2013 injury is what lost him the lead.
Now, let’s take a look at the top performing player at each position year-by-year.
When looking for repeat performers by position, we saw multiple repeat performers at quarterback, tight end, and wide receiver, which again emphasizes that past performance is an indicator of how they will play in future seasons.
However, it is interesting to note that among running backs, there is no repeat top performer. Considering the high injury risk, this is not surprising, especially as more teams move towards pass-heavy offenses.
The draft: analyzing patterns by position
Before we look at which positions should be drafted first, let’s take a peek at last year’s draft order by position. Below we see the Player Average Draft Position over the Draft Pick Number:
Looking at the first two rounds, we can see a couple things:
The Gronk hype was real last year and almost single-handedly dragged tight ends into the first two rounds.
Over 90% of drafters chose to stick with some combination of running back and wide receiver. Not surprising, but it definitely shows that if it’s a specific RB or WR that you want, get him in your first round.
So now that we have an overview, who was making the right decisions? Is scrambling for premium running backs and wide receivers the right play? Or did the outliers that went with early quarterbacks or tight ends get rewarded. Let’s take a look at output.
What history tells us about drafting
- Prepare for injuries with running backs: Looking at the chart, there is a bit more noise than we might have expected. Right off the bat, we noticed some massive disappointments amongst the running backs – often resulting from significant injuries:
Where Todd Gurley was just an underperformer at full strength, Adrian Peterson, Eddie Lacey, and Jamaal Charles all suffered season-ending injuries. Looking at previous seasons, we found similar patterns with top running backs.
Here’s a look at 2014:
Adrian Peterson missed time due to off-field issues, Doug Martin and Giovani Bernard suffered from injuries, and a healthy Montee Ball failed to live up to expectations. Running backs are risky – we already knew that – but take injury chances seriously and remember #4 in this list.
- Quarterbacks are predictable. Turning to quarterbacks, we saw a different pattern. The data demonstrates the impact of the NFL putting a strong focus on protecting quarterbacks, and the result is much more predictability out of fantasy QBs.
We’re missing some data from 2015, but taking a look at 2014 and 2016, we see some pretty consistent success in the top five, other than a little outperformance from Aaron Rodgers in 2016 and a bit of disappointment with Cam Newton. The only real outlier was Nick Foles’ 129-point 2014 campaign, the only significant injury in the group.
- The drop-off rates here in the top six quarterbacks are pretty gentle, indicating that stretching for the superstars may be misplaced if you can swing the 5th or 6th QB in the draft. This year (below), you may want to look to Derek Carr and Russell Wilson based on early drafts (the bold pick may be Andrew Luck, but beware the injury issues).
- Wide receivers provide flexibility. Next let’s look at wide receiver production. The operative word here is depth. Here are the outputs of the top 30 picks in 2014 and 2016:
Other than a Keenan Allen injury (orange, bottom left) and the Josh Gordon off-the-field issues (orange, bottom right), you are getting what you pay for with wide receivers. A nice gentle decay means you can fight for the studs early or backload with the middlers. Wide receivers are going to give you the most optionality in structuring your team. So our recommendation would be to work your wide receiving corp at the margin, filling gaps when you don’t have great options at other positions.
- The tight end drop-off. And finally, let’s take a look at tight ends. With tight ends, it’s all about the outliers. This chart explains why Gronk was going in the first round in 2016:
Gronk and Jimmy Graham scored 2-10 points per game more than a typical top tight end in their best years. Compare that to the gentler curves with quarterbacks and wide receivers and you can understand why a full strength Gronk (or Jimmy Graham in New Orleans) could command a premium.
Looking at this season’s early draft numbers, Gronk is definitely worth the 20th overall ADP right now if he can stay healthy. You are looking at a pretty steady drop-off after that (with some long shot potential if Jordan Reed can stay healthy at 50th overall).
The data confirms what we all know to be true: the players most likely to be injured are your riskiest bets. But it also shows that you can counterweight that risk with some reliable picks.
What can we learn from the biggests busts of 2016?
Last, we’ll take a look at the biggest busts of 2016 to see if there’s anything we can learn. Here’s a quick overview of the draft class of 2016, with projections on the x-axis and outperformance vs. projections along the y-axis:
- Projections Outperformance: Outperformance, is pretty well centered on the 0, meaning Yahoo estimates are a pretty good ballpark estimate. That said, we have some pretty significant divergence, especially on the higher estimates. Digging in to Yahoo’s blue chips (the right side of our chart) we see some themes we covered above:
Be aware of the high injury risk associated with the running back position (as well as high-risk players at other positions such as Keenan Allen and Rob Gronkowski).
Blue chip quarterbacks are giving you about what you expect, even though Cam Newton had one of the worst divergences in recent history.
On the upside, young, workhorse running backs are where the value is. Although most people may not have a shot to draft David Johnson, who was first overall in almost 50% of leagues.
Looking at the big outperformers out of nowhere, we have a few themes, and a lot we can ignore. Here’s everyone who beat their pre-season projection by more than 100 points:
First, we have lots of unexpected starting quarterbacks. Not a ton of action we can take there, since these still aren’t starter-level outputs, save Dak Prescott. Might be worth a flyer on some of the new rookies, especially if they have a strong chance to win the starting role.
Taking a look at the other outperformers, we have young wide receivers that stepped into big roles with injuries or shifts in the offense – Adam Thielen is the old man of the bunch at 26, with Tyreek Hill, Cameron Meredith, Tyrell Williams, and Davonte Adams all under 25.
We have Pitta’s return from injury (the only veteran on here), David Johnson’s monster season, and Jordan Howard emerging from the running back committee in Chicago (sometimes it’s worth taking a flier on a rookie like Joe Mixon in a running back by committee on the chance he wins the starting job outright).
And for fun, before you leave, let’s look at the worst efforts ...because sometimes being benched does you more good than when you play.
- Remarkably, Marcus Thigpen’s 2015 takes the cake at -4.10 points, spread over three individually negative games (beware the kick/punt returner, no points except for touchdowns and turnovers):
In our next post, we’ll dig further into this dataset and see if there are any more concrete tips we can get for choosing a draft pick. Stay in the know by subscribing to the blog in the box below.
If you want to learn more about Looker’s complete data platform and how it can help you get this level of insight from your business, sign up for a demo .