Consider Trashing These Stats When Placing Your NFL Picks

Jason Lake

Friday, July 15, 2016 9:16 PM UTC

Friday, Jul. 15, 2016 9:16 PM UTC

It's always a good idea to look at the numbers when you're putting together your NFL picks, but you'll do a much better job if you look at the right numbers.

A little learning is a dangerous thing;
drink deep, or taste not the Pierian spring:
there shallow draughts intoxicate the brain,
and drinking largely sobers us again.

– Alexander Pope, An Essay on Criticism (1709)

Most people who bet on football don't really put too much thought in it. They'll just bet a few bucks on their favorite team, for funsies. These recreational bettors are the lifeblood of the NFL. They should be cherished, not ridiculed for being “square.”

But what about the rest of us? We might think of ourselves as informed bettors, poring over stat sheets, analyzing the data, and making smart decisions with our NFL picks. Unfortunately, there's a lot of bad data out there. Numbers can be useless, or even downright misleading. Even the smartest of us can get tripped up by bad stats.

A No-Win Situation
That includes you, Stephen J. Dubner. There's a “Football Freaknomics” video that talks about how poor a statistic passer rating is. It's definitely not the best stat in the toolbox, but for their argument, Dubner points at Daunte Culpepper's 87.8 career passer rating – higher than Dan Marino's (86.4), higher than Brett Favre's (86.0). Then Dubner drops the hammer: Culpepper was only 41-59 as a starting quarterback. Dubner even giggles as he says “41-59.”

Sweet baby corn, this is wrong in so many ways. One: Wins and losses are not very reliable indicators of future performance. Bad teams win football games all the time, and good teams lose. It's much better to look at efficiency-based stats like the DVOA (Defense-adjusted Value Over Average) rankings at Football Outsiders, but if you need something quick and dirty, point differential says a lot more about a team's quality than wins and losses.

Daunte's Inferno
Two: Football is a team sport. If win-loss records are bad, then QB win-loss records are awful. Culpepper played for six years on a Minnesota Vikings team that had one of the worst defenses in the NFL. That includes his outstanding 2000 rookie campaign, when he led the league with 33 touchdown passes, and the Vikings went 11-5. It also includes his incredible 2004 season, when Culpepper led the NFL with 4,717 yards passing. The Vikings went 8-8.

Three: Straight-up comparisons of stats from different eras is ridiculous. Football has turned into a pass-heavy enterprise in the 21st Century; seven of the Top 10 NFL quarterbacks in career passer rating are active going into the 2016 season. Only one of the quarterbacks in the Top 100 played before 1950: Otto Graham (86.6 passer rating), at No. 20.

I could go on, but I'm out of rant juice. Here's the point: Watch out for win-loss records, and don't put too much stock in counting stats like yards per game or touchdowns per season. These numbers will lead you astray. Use efficiency-based stats instead if you want to beat the NFL odds. And whatever you do, never giggle at someone's record.


Read more on NFL Betting Here: The Betting Bible To Cash In NFL Picks

comment here