Daily Fantasy Football: Identifying Trends in Big Games From Quarterbacks

I feel pretty safe in asserting that we all have crushes within the NFL, correct? The guys over whom we obsess, almost to an unhealthy degree? You know you've got one. Mine's Marcus Mariota.

When Mariota was playing college ball at Oregon, I had a spreadsheet dedicated to his junior-year stats, just so I could quickly calculate his Adjusted Yards per Attempt at the end of each game. His Sports Reference page was my most-visited site that didn't include a corgi jumping in water. It's a little stalkerish, but I'm cool with it.

Thus when Mariota transitioned to the NFL, I looked for any excuse I could get to use him in daily fantasy sports. He's facing a fellow Heisman winner in Johnny Manziel? Great! The team's wearing sick baby blue uniforms? Even better!

Now, using Mariota in his rookie season wasn't always a bad thing. He scored at least 20 points on FanDuel in 6 of the 10 injury-free games he played, and he even had two games above 30 points. Those are tourney-winning games if you plug a guy at his price into your lineup.

Did I do so for either of his 30-point games? Not a chance, old chap.

Turns out that using uniform color to select your quarterbacks isn't what some would classify as being "optimal." Who knew?

Out of desperation to avoid missing out on any more of Mariota's future high-upside games, I decided to take a look at other big-time fantasy games out of quarterbacks. If we could develop a blueprint, then spotting these games in advance would be easier, allowing me to allot more time to watching dog videos.

As such, I went through the 157 games from 2013 to 2015 in which a quarterback scored at least 27 points on FanDuel. Why 27 points? This would be three-times value for even the most expensive quarterbacks each week, making them solid tournament options when they hit that mark. If a lower-priced guy hits it, then that's even better.

For each game, I recorded whether the player was at home or on the road, whether the game was beneath a roof (either a dome or a closed retractable roof), whether the game was against a divisional opponent, the spread, the over/under, the implied team total, the wind speed, the temperature, the quarterback's efficiency ranking, and the opposing defense's ranking against the pass. If a game was played with a roof, the temperature was recorded as being 70 degrees with a wind speed of zero.

Both the quarterback's ranking and the opposing defense's ranking were based on numberFire's Net Expected Points (NEP). This is the metric we use to measure the efficiency of both teams and individuals, with the team totals being adjusted based on strength of schedule.

Here's how NEP works. Prior to each play, there's an expected number of points that the offense will score on its current drive. If the quarterback completes a three-yard pass on 3rd and 2, those expected points will increase, resulting in positive NEP. However, if he completes a three-yard pass on 3rd and 4, those expected points will go down, resulting in negative NEP. The sum of these fluctuations in expected points is NEP.

By using these tools, we can get a bit closer to finding which quarterbacks are about to go off. What does our blueprint look like for big-time fantasy games?