My favorite pastime right now1 is checking out various scenarios in the Game Splits App. My most recent experiment involved
peyote looking at the performance of quarterbacks when their team is favored (negative point spread) vs. when their team is the underdog. I liked this post by Rich Hribar, but wanted to use a slightly different approach. To accomplish this, I set the date range filter for the past five seasons, in order to get a big enough sample of games, without giving too much credit to ancient history. To keep the exercise simple, I didn’t make any other adjustments; I didn’t remove any games or filter for the presence or absence of other players, etc. In other words, this was just a quick exercise to see if anything interesting popped up.2
|QB||FPG Fav||FPG Dog||Diff||Games Fave||Games Dog||Total G||Pct Fave|
The table should be pretty self-explanatory. I didn’t include any of the actual passing stats because, you know, you can get those yourself. For the astute observer, you’ll notice EJ Manuel, Mike Glennon, and hypothetical Browns’ starter Brian Hoyer are missing. None of these three have really played enough games to analyze. Some quick thoughts.
- If you sort by the largest point differentials when favored, Matt Cassel is second. Perhaps this is a small indicator that, if the Vikings defense improves as expected, and Minnesota is favored more often, that Cassel (if he starts) may have a decent season. The transitive property would then suggest that perhaps the Vikes other skill players, like Greg Jennings, could have good seasons.
- If only Matthew Stafford could be favored all the time. He has the biggest point differential of any QB, scoring over seven fantasy points/game more when the Lions are favored. You should already be trying to acquire him everywhere. If some combination of a better set of receivers, a new offensive scheme, or an improved defense (the Lions do have the ninth-most draft capital invested in their defense) leads the Lions to be favored more often . . . good times.
- The other interesting thing about the top end of this list (or the left side of the chart below) is the company Stafford is keeping. Cassel seems to be an outlier, but it’s probably not a coincidence that the above average cohort includes Drew Brees, Robert Griffin III, Russell Wilson, Aaron Rodgers, Peyton Manning, and Tom Brady. It may not mean a lot, but it’s certainly good company.
- When I look at the set of QBs who perform significantly better when favored, my first thought is “they’re doing what they’re supposed to do: taking advantage of weaker competition.”
- Most of the QBs are pretty consistent, falling within a point or two of the same FPG average whether they’re favored or not. The positive outlook there is consistency. You know what you’re getting. That may not sound sexy, but there are definitely scenarios when a predictable asset is highly desirable.
- If Late Round QB is your thing, check out Carson Palmer, whose performances (albeit not always high scoring) appear to be immune to point spreads.
- Another such player to consider is Tony Romo. His ADP suggests he’s a bargain. Yes, the Dallas defense will likely be awful, but historically being an underdog hasn’t had much impact on Romo.
- At the opposite end of the spectrum, Andrew Luck actually performs better when he’s an underdog. I suppose the reason is something like “he needs to sling it around to keep them in the game.” Since the Colts have done very little to bolster their defense, Look for Luck to be a frequent underdog, which could make him the top fantasy QB of 2014.
- On a raw FPG basis, the best QBs when their team is the underdog? Cam Newton, Luck, Brees, Rodgers, Manning, and Philip Rivers.
- Take a closer look at Newton. He has a perfect split of games favored/underdog, and the total number of games is big enough to feel pretty confident in what the splits tell us. Not only does he sport one of the highest FPG rates when favored, he sports the highest rate when an underdog. And the difference between the two rates is nominal. In other words, he looks like a high end, “start him and forget him” option. This consistency meshes nicely with this article from Fantasy Douche. Newton could have a better than expected season. He may even be the man of steel.
- Poor, poor Chad Henne. He’s been favored in less than 14 percent of his career games.
I think there are a few ways you can apply this information. The most obvious is probably in daily fantasy contests, where incorporating point spread information can help you make better decisions about which players to put in your lineup. Back to Stafford. For his career, he averages 22 FPG, which is a good number. But that jumps to 25.9 when favored. That’s the third-highest total overall, trailing only Rodgers and Brees. But when he’s the underdog it drops all the way to 18 FPG.
Similar logic applies to decisions in your seasonal leagues. The point spread can help point you toward the better starter for any given week. The Game Splits App makes a fine accompaniment to the excellent series of articles penned by Hribar, who explored the impact of game flow on the performance of players at the primary fantasy positions. The point spread is a good pregame indicator of which teams’ offense might get “off script.” Use the App in conjunction with the tables in Hribars’ articles to identify players to start/sit. To get you started, here’s the QB article you need to read.