Draft Strategy

QBs that Win MFL10s

I took a stab at predicting MFL10 QB ADP and highlighted the weak relationship between win rate and how late teams waited on quarterback. No quarterback drafted outside the top-153 picks had a positive win rate (anything over 10 percent) but the overall relationship between QB ADP and team success was still relatively weak.

2016 QB Overall ADP and Win Rates

Teams that drafted Cam Newton got wrecked in 2016, winning only 4.7 percent of their leagues, although other early QBs tended to do well. But just how well did they do? Even if we remove Newton’s data point, these two stats — QB ADP and win rate — only result in an R-squared value of 0.406.1

Aim for the Big Guns

If ADP is not the end-all, be-all of picking QBs to win your MFL10s, what else might be? Last year, I found that the number of weeks a QB finished in the top 12 correlated strongly with the win rate for MFL10 teams that drafted him, so I thought it would be worth testing to see if 2016 showed the same connection.

2016 QB Top-12 Games and Win Rates

Just like 2015 drafts, 2016 MFL10s showed a strong correlation between top-12 games and win rate. Here, including Newton, we have an r-squared value of 0.64, compared to 0.24 between ADP and win rate. As reference points, QB points per game and total points scored out at 0.48 and 0.54, both worse than the number of top-12 weeks, just like they were in 2015 as well.

If we look more closely at the data, we can also see that every single positive win-rate QB had at least seven games in the top-12 last season. In other words, seven top-12 weeks was a necessary, but not sufficient, condition for achieving a positive win rate at QB this year. Over the last two seasons, 16 of 18 quarterbacks with win rates over 10 percent had at least seven top-12 weeks.2

The 2016 season, it seems, only reinforced the lesson I learned last year:

My takeaway is this: if you intend to wait on QB, throw a dart or two at high-upside players. While low-ceiling, reliable options may be all the talk, QBs who consistently hit the weekly top-12 actually correlate more strongly to winning teams.

“Sharp” MFL10 drafters are likely to praise QBs like Joe Flacco and Alex Smith yet again this year, but I will keep targeting the guys with more explosive weekly upside, like Andy Dalton. Dalton has never had fewer than six top-12 weeks in each of his last three seasons, more than Flacco or Smith have had in any of those three seasons.

Here is the full table of top-12 MFL10 QB weeks over the last three seasons:

 Average # of Top-12 Weeks201620152014
Dak Prescott1010
Aaron Rodgers9.3310810
Andrew Luck910413
Drew Brees8.679710
Russell Wilson8.337810
Tom Brady87107
Ben Roethlisberger7.67878
Andy Dalton7.33796
Marcus Mariota786
Matt Ryan6.671037
Cam Newton6.67695
Matthew Stafford6.33766
Philip Rivers6.33487
Kirk Cousins6873
Derek Carr6873
Blake Bortles66102
Eli Manning6576
Tyrod Taylor657
Peyton Manning6111
Jameis Winston5.574
Ryan Tannehill5537
Carson Palmer5573
Kyle Orton55
Alex Smith4.67554
Joe Flacco4.67455
Jay Cutler4.67077
Colin Kaepernick4633
Ryan Fitzpatrick4174
Tony Romo417
Teddy Bridgewater435
Sam Bradford3.543
Brian Hoyer3342
Trevor Siemian33
Blaine Gabbert2.523
Nick Foles2.523
Josh McCown2.33133
Carson Wentz22
Matt Barkley22
Case Keenum22
Shaun Hill22
Geno Smith22
Drew Stanton22
Austin Davis22
Zach Mettenberger22
Mark Sanchez1.503
Robert Griffin III111
Matt Moore11
Brock Osweiler11
Cody Kessler11
Jimmy Garoppolo11
Jake Locker11
Colt McCoy11
Michael Vick11
Chad Henne11
Matt Cassel11
Johnny Manziel11
Brandon Weeden11

Subscribe for a constant stream of league-beating articles available only with a Premium Pass.

  1. 0.406 is much higher than last year’s 0.09.  That is consistent with my claim that 2015 was an aberration in the number of successful late-round QBs.  (back)
  2. Brian Hoyer and Zach Mettenberger were the 2015 exceptions, with four and zero top-12 weeks, respectively.  (back)
By Joshua Lake | @LakeTwoQBs | Archive

Comments   Add comment