This is the 151st installment of The Labyrinthian, a series dedicated to exploring random fields of knowledge in order to give you unordinary theoretical, philosophical, strategic, and/or often rambling guidance on daily fantasy sports. Consult the introductory piece to the series for further explanation.
As I’m writing this sentence we’re just under three months away from the first game of the 2017 NFL regular season — so there’s clearly a pressing need for us to start covering football.
Who cares if right now the most noteworthy NFL news item is that a whole bunch of players voted Cowboys quarterback Dak Prescott and running back Ezekiel Elliott as top-15 players in the league? That’s what we’re going to cover!
You’re welcome!
The Top 100 Players
In the spring of 2011, the NFL was in the middle of an unsavory dispute with the NFL Players Association that resulted in an 18-week, 4-day work stoppage. During this time — to give the impression that it gives a sh*t about its players (as well as to make money during the offseason and keep people addicted to football) — the NFL launched the first season of The Top 100 Players.
The 2011 NFL Draft ended on the afternoon of Sunday, April 30. Literally that night — just a few hours later — the first episode of The Top 100 Players of 2011 ran on the NFL Network. Say what you want about the NFL, but those punks at least don’t mind being transparent about their greediness.
The premise of the show was simple. It would cover the 100 best players in the NFL . . . as ranked by the players themselves! This makes some sense from the league’s perspective: If you want to limit as much as possible the ability of players to determine their true valuations within a free market, the least you can do is let them haphazardly evaluate each other and then exploit use their work to create a television show that provides no tangible information but does generate extra money for the league, amirite Roger?
Plus, who could possibly be better at ranking NFL players than the players themselves?
It turns out there are probably lots of people who can rank NFL players better than players.
The Top 100 Players of 2017
Sadly but unsurprisingly, this abomination of a show has endured into its seventh season, and this coming Monday (June 26) at 8 p.m. ET the exact ranking of the top 10 players will be revealed. In the words of Terrell Owens, “Get your popcorn ready.”
Because the NFL can’t help but perpetually tease, it has already released the composition (but not the order) of the top 10 players:
- Tom Brady: If anyone wants to bet that last year’s Super Bowl MVP (and maybe the greatest quarterback in NFL history) won’t be the No. 1 player on this list, let me know on Twitter. I’m only half-joking. Brady’s the early favorite to win the 2017 NFL MVP award. The prop markets currently have him at +400.
- Aaron Rodgers: Last year’s No. 6 player, Rodgers tends to be in the top five when wide receiver Jordy Nelson is healthy and scoring double-digit touchdowns. A few days ago Rodgers was +1,000 to be the 2017 NFL MVP. Now he’s +700.
- Matt Ryan: This is his first time in the top 10 — and also the top 100. Ryan isn’t a top-10 superstar, but he’s definitely been a top-100 player for the last seven years. Perhaps no player exemplifies the ridiculousness of this list more than Ryan. The prop markets originally opened with him at +1,200 to repeat as NFL MVP. That line has quickly moved to a more reasonable +1,600.
- Ezekiel Elliott: Sure, he ran behind one of the best offensive lines in the league and benefited from maybe the best rookie quarterback of all time and got more carries than any other running back and plays a short-lived position that’s more dependent on opportunity than ability — but let’s just say that Elliott’s already a top-10 player. Also, I get supreme pleasure every time I see his name misspelled on NFL.com — and it happens a lot. Zeke is the most-favored non-quarterback to win the 2017 NFL MVP at +2,000. That line opened at +1,200. Over the last decade the only non-quarterback to win the award has been Adrian Peterson, who was the No. 5 player last year and is now No. 98.
- Le’Veon Bell: Bell is undoubtedly a more complete player than Elliott, and in best ball drafts Bell is typically selected ahead of Elliott — and yet this is only his first appearance in the top 10 . . . after being an All-Pro in 2014 and a stud in 2015 (albeit in only six games). Sure, that makes sense.
- Antonio Brown: This is Brown’s third consecutive year in the top 10. Meanwhile, Steelers quarterback Ben Roethlisberger has never broken the top 20 — even though he’s won two Super Bowls without Brown (and Bell, for that matter). Something tells me that Roethlisberger is more important to the Steelers offense than Brown is.
- Odell Beckham: He’s opened his career with three seasons of at least 1,000 yards and 10 TDs receiving. He’s basically a shorter, less athletic, less productive, maybe more diva-ish young Randy Moss.
- Julio Jones: Since his second year in the league, Julio has averaged just over 100 yards receiving per game. No one comes close to matching his five-year record of yardage accumulation. He’s probably the NFL’s best wide receiver.
- Von Miller: In his six-year career, Miller has 73.5 sacks, 272 tackles, 73 assists, 20 forced fumbles, five fumble recoveries, and 13 passes defended in 88 games. In his six-year career, J.J. Watt — last year’s No. 3 player and 2015’s top overall player — has 76.0 sacks, 299 tackles, 80 assists, 15 forced fumbles, 13 fumble recoveries, and 45 passes defended in 83 games. But, yes, Miller is a top-10 player and Watt is now No. 35.
- Khalil Mack: With an average of 13.0 sacks and 55.5 tackles per season over the last two years, Mack is almost as good as Watt when he’s 80 percent healthy.
Why am I talking about The Top 100 Players of 2017 if I think the rankings suck and the process is bogus?
- I’m obsessed with football.
- The rankings and process might have some relevance to fantasy sports.
Let’s dig in.
The Problems with The Top 100 Players
There are a number of evaluative problems with The Top 100 Players. Some of these might be unique to the project of having players rank players — but they’re all mistakes that anyone can make when assessing something — and that includes DFS.
Let’s consider them.
Recency Bias
Cam Newton this year is ranked as the No. 44 player in the NFL. Last year, after a 2015 MVP campaign that culminated in a Super Bowl appearance, Newton was No. 1.
Both of those rankings are examples of recency bias.
In 2016, Newton wasn’t the NFL’s best player, but he’s not only the 44th-best player now. The quarterback position is the most important in football. Newton has started 16 games in four of his six NFL seasons. In those four full seasons, he’s been a top-five fantasy quarterback each year. He ran intensely hot in 2015 — leading the position with 25.94 DraftKings points per game and a +7.03 Plus/Minus — so we shouldn’t expect him to have a season like that again, but he’s a 28-year-old signal caller one year removed from an All-Pro campaign.
He’s at least a top-30 player. The only way he isn’t is if you discount the first five years of his career and look primarily at 2016 — and that’s what the players did. Even then, the players were probably still too low on Newton.
Overvaluation of Winning
Winning is everything. Newton’s drop in the ranks (I argue) is just as attributable to his team’s failure to win as it is to his failure to produce. In 2015, the Panthers were 15-1. In 2016, they were 6-10. If they had been a 10-6 team that managed to win a playoff game, they basically would’ve been the Seahawks, and Newton would’ve gotten some respect as a guy who managed to “tough it out” and “lead his team to victory” even though he wasn’t at his best. He probably would’ve been ranked around Roethlisberger and Russell Wilson, who are Nos. 22 and 24.
While the quarterback position is important, Newton’s lackluster season isn’t the reason the Panthers sucked in 2016. Stud middle linebacker Luke Kuechly missed six games to an injury, and the team entered the season without its two primary cornerbacks from the previous year. Newton’s play didn’t help his team, but the Panthers likely would’ve struggled even if Newton had played better.
In a couple of ways fantasy players can often become too fixated on winning:
- NFL players on losing and/or underdog teams are owned less than they should be in guaranteed prize pools. Such players are often rosterable at steep ownership discounts. During the upcoming season, Pro subscribers will be able to check out ownership patterns for themselves via our DFS Ownership Dashboard and Trends tool.
- Even good DFS players can go months without a big win. What’s important in the short term isn’t whether you’re producing; it’s how you’re playing. If you’re playing well, over the long term you’ll be a productive DFS player.
Undervaluation of Consistency
Drew Brees has led the NFL in yards passing in five of the last six seasons. He’s passed for more than 5,000 yards in a season five times in the last nine years. No one else in league history has done it more than once. He currently has the NFL record for career completion percentage. Over the last four years he’s completed just over 69.0 percent of his passes.
In the words of BBC’s Sherlock Holmes, “Boring.”
From 2011 to 2014, Brees on average was ranked No. 7. Since then — and Brees has led the NFL in passing in each of the last three years — he’s been 30th, 30th, and 16th.
While the Saints’ three consecutive 7-9 records are partially to blame for his drop in the rankings, Brees has also become too familiar to his peers. They’re used to his greatness. His aerial abilities no longer impress them, and so they have undervalued his consistent prodigiousness.
It’s possible in DFS to undervalue players we find boring. When was the last time you were excited by the idea of rostering Frank Gore? Yet last season Gore had a +4.26 FanDuel Plus/Minus and strong 75.0 percent Consistency Rating. For the last decade, Gore has been at worst a steady RB2, averaging over 1,350 yards from scrimmage and just under eight TDs per season. He’s probably been the most undervalued NFL player of the last 10 years.
Small Sample
For DFS players, recency bias in any sport is a problem, but it’s especially a problem in the NFL.
In MLB, if we focus on how a guy performed in 2016 but ignore his career data, that’s not ideal, but at least we’re still looking at 162 games. In NFL, if we focus on one season’s worth of data, we’re looking at just 16 games. That’s a pitifully small sample that could very well be unrepresentative.
In 2015, DeAndre Hopkins became one of only 25 wide receivers in NFL history to have at least 1,500 yards receiving in a season. Of course, his production wasn’t fluky. The year before — while playing with the unfortunate foursome of Ryan Fitzpatrick, Case Keenum, Ryan Mallett, and Tom Savage at quarterback — Hopkins had 1,200 yards in just his second season. A first-round pick in the 2013 draft, Hopkins is clearly talented. After his third season, only Moss, OBJ, A.J. Green, Torry Holt, and Jerry Rice had more yards receiving through their first three years than Hopkins.
After his third-year breakout, he was ranked the No. 19 player in the league. Now, though, after his Brock Osweiler-impacted 2016 campaign, he isn’t even in the top 100. On the one hand, his fourth NFL season wasn’t good. He had only 954 yards and four TDs. On the other hand, he still has more career receiving yards than any other receiver has had in his first four years with the exceptions of Moss, Holt, Rice, Green, Anquan Boldin, and Larry Fitzgerald.
All apologies to Fitz (No. 45), Jarvis Landry (No. 42), and even the volatile Tyreek Hill (No. 36), but right now Hopkins is a better wide receiver than any of them is.
When NFL players failed to rank Hopkins in the top 100, they weren’t victims of just recency bias. They were victims of a small sample. They prioritized 16 games unrepresentative of who Hopkins has been and most likely will be in the future.
Relative Value
Perhaps the most important evaluative problem many people have is their metaphorical inability to compare apples to oranges and to know which fruits are more important to various types of fruit salads.
It’s also a real problem that some people have. There was a brief period in the 2000s when my mom thought it was ‘exotic’ to put avocado in fruit salad. If “exotic” is interchangeable with “horrible,” sure. If not having avocado in fruit salad is wrong, I don’t want to be right. The same goes for tomatoes. What the f*ck, world?
Anyway, people are generally not good at comparing different types of component parts and knowing which parts are more important to the whole. I touch on this idea in my recent article on DFS decision fatigue. It’s not as if NFL players ranked a bunch of kickers in the top 100, so they’re not morons. At the same time, you’d have to think that maybe Stephen Gostkowski — who in his 11-year career has converted 93.9 percent of his 33 field goal attempts in the playoffs and 87.1 percent of his 348 regular season attempts — is better at doing his job than LeGarrette Blount (No. 80) is at doing his.
In general, though, there are spots where players focused too much on positional ability to the exclusion of positional value, which is an important part of the “Is this guy one of the best players in the NFL?” equation. For instance, Marcus Mariota is No. 50, Andrew Luck is No. 51, and Jameis Winston is No. 57. We can quibble with those rankings if we want — I think Luck should be ahead of the two second-year quarterbacks; the prop markets agree with me, by the way, as Luck is +2,500 to be the 2017 NFL MVP whereas Mariota is +3,300 and Winston is +6,600 — but we can all probably agree that those three quarterbacks have the potential to win MVP awards and lead their teams to Super Bowls within the next five years. They are significantly more important than defensive linemen Gerald McCoy (No. 52), Ndamukong Suh (No. 55), and Cliff Avril (No. 56).
In 2014 Luck led the NFL with 40 TDs passing. McCoy, Suh, and Avril have never come close to having that kind of impact on the defensive side of the ball. They are good — among the best players at their positions — but McCoy, Suh, and Avril are not as good as Mariota, Luck, and Winston, even if they’re all technically better at what they do.
Winston is the No. 13 quarterback in the rankings. I’m not a fan of Winston as a person — but whoever is the 13th-best NFL quarterback is better than three defensive linemen ranked in the 50s — as are Kirk Cousins (No. 70) and probably Philip Rivers (No. 73), who are 14th and 15th at the position.
The NFL is a passing league, and it’s hard to have a good passing game without a good quarterback — and it’s hard to play quarterback. As much as the NFL rewards top-tier passers, its players discount mid-tier passers.
When constructing lineups, DFS players can make the same sort of mistake.
Final Thoughts
There’s no great way to end this piece.
[Insert here 5,000-word rant on Peterson’s ridiculous +5,000 odds to win the 2017 NFL MVP.]——
The Labyrinthian: 2017.56, 151
Previous installments can be accessed via my author page or the series archive.
