The Most American Day Ever — Except for the Other Most American Day Ever
Com’on, ‘Murica. It’s time to stop debating with your friends ‘friends’ about the election and start doing something useful — like eating a sh*tload of food, drinking an *ssload of alcohol, and watching God’s favorite NFL franchise (and five rando teams) play some f*cking football on a weekday.
All we’re missing is . . .
[Insert here any photo of a random woman sitting on top of a horse, wearing a skimpy US flag bikini, and holding a massive gun in her right hand and the Bible in her left.]It’s Thanksgiving. Let’s talk some football.
But Before We Talk About Football . . .
In October, I did a couple of pieces on the London NFL games. I was wondering if there were some trends that might be exploitable for the purposes of daily fantasy sports.
What follows is a lot of information that might have no applicability whatsoever.
You’re welcome.
Oh, by the way, d*mn right I’m counting this as a Labyrinthian. Something as meandering as this piece deserves to be included in the series.
From Dallas to Detroit to Wherever
The first Thanksgiving Day football game (according to myth and/or Wikipedia) was played between Yale and Princeton. Fairly quickly professional teams started having games on the holiday, since most people had the day off from work.
When the NFL started in 1920, six professional football games were played on Thanksgiving of that year, and at least one professional Thanksgiving game was played each year till 1941, when the observance was suspended because America was too busy saving the world.
With humankind indebted to it forever, America brought back the practice of Thanksgiving games in 1945, with the Detroit Lions hosting the Cleveland Rams. In typical fashion, the Lions lost. (Also interesting is that the Rams moved to Los Angeles the next year. Knowing that they started in Cleveland really helps me place their back-and-forth franchise moves in context, amirite LeBron?)
The Lions have hosted a game on Thanksgiving every year since then.
In 1952, the first iteration of the Dallas Texans ‘hosted’ a home Thanksgiving game in Akron, Ohio. (I could tell you how that happened, but I’m trying to force myself not to overwrite.) In 1960, the second iteration of the Dallas Texans (in the American Football League) hosted a Thanksgiving game. This time the game was actually played in Dallas at the Cotton Bowl, which the Texans shared with the cross-town Cowboys.
Less than six months after winning the AFL Championship, Texans owner Lamar Hunt moved the team to Missouri in May of 1963, changed the name of the franchise to “Kansas City Chiefs,” and then proceeded to complain for the rest of his life about how the Chiefs couldn’t get a Thanksgiving Day game on their schedule.
You know what? When you move a team from one city to another — especially a championship-winning team — the football gods don’t always give you want you want. Don’t mess with Texas.
In 1966, the Cowboys hosted the second-ever Thanksgiving game in Dallas. They beat the Cleveland Browns 26-14, with four Danny Villanueva field goals providing the margin of victory.
The Cowboys have hosted a Thanksgiving game every year since then — save for the two ‘lost’ years, when the NFL inexplicably let the St. Louis Cardinals host the event in 1975 and 1977. The Cards were beaten handily 32-14 by the Buffalo Bills in 1975 and 55-14 by the Miami Dolphins in 1977. They were also beaten 19-14 by the Cowboys when they hosted the Cards for the Thanksgiving game in 1976.
Since 1978, the Lions and Cowboys have had a monopoly on the Thanksgiving Day games, with the Lions always starting around 12:30 p.m. ET and the Cowboys 4:30 p.m. ET.
In 2006, the NFL got with the times and added a third Thanksgiving game.
Due to his historical importance to the league, his old age, and his constant nagging, Hunt on Nov. 23, 2006, was allowed to see his Chiefs host the NFL’s first prime time Thanksgiving game. He died less than a month later. Let that be a lesson to everyone: Sometimes when people get what they want, they die.
For the decade since then, the NFL has played three games on Thanksgiving.
What follows is everything about the Thanksgiving games that have been played since 2006.
Everything About the Thanksgiving Games
First, here are a couple of caveats.
- The sample is relatively small: It includes only 30 games. What that means is that, on the one hand, you shouldn’t give the data herein too much importance. On the other hand, those games are equivalent to 5.86 percent of the 512 games that would be played in an NFL season — and that’s not an insignificant percentage of games.
- The sample might be unrepresentative and/or skewed: 33.3 percent of the games feature the Lions of the last 10 years. Another 33.3 percent feature the Cowboys. On the one hand, it’s not ideal that the sample is that specific and narrow. On the other hand . . . the Lions and Cowboys are playing on Thanksgiving Day, so the sample isn’t really unrepresentative.
OK, what did I find?
Vegas Data
Over the last 10 years the average over/under has been 45.8 points. The spread has been 6.45 points.
(Side note: As I write this piece, our Vegas page shows that the three 2016 Thanksgiving games currently have an average over/under of 49.2 points with an average spread of 4.3 points. That cranberry relish has a little bit of a kick.)
Two-thirds of the time, the home team is the favorite, which is exactly what we see this year.
Historically, the over has been hit 53.33 percent of the time; the under, 46.67 percent. The favorite has covered the spread and reached its implied point total 66.67 percent of the time; the underdog, 33.33 and 30 percent.
Home/Road Splits
In the 20 games in which the home team has been favored, the over has been hit 45 percent; the under, 55 percent. The home favorite has covered 60 percent and reached its implied total 55 percent of the time. The road underdog has massively struggled, covering 40 percent of the time and reaching its implied total only 25 percent of the time. On average, these games have 45.0-point over/unders and 6.2-point spreads.
In the 10 games in which the road team has been favored — and, by the way, seven of those games feature the woeful Lions teams from 2006 to 2012 — the over has been hit 70 percent of the time; the under, 30 percent. The road favorite has dominated, covering 80 percent and reaching its implied total 90 percent of the time. The home underdog has covered only 20 percent of the time. At least it has put up something of a fight, hitting its implied total 40 percent of the time. On average, these games have 47.4-point over/unders and 6.95-point spreads.
Division/Non-Division Splits
Two of this week’s games — Vikings at Lions and Redskins at Cowboys — are division games, which can offer their own set of unique circumstances.
Over the last 10 years, 12 Thanksgiving games have been divisional matchups. Seven of those have featured the Lions and Cowboys. The average over/under has been 46.8 points with a five-point spread. The over/under has a 50/50 hit split. The favorites cover and reach their implied point total 58.33 percent of the time.
Here’s one intriguing sub-split: The divisional home favorites — which the Lions and Cowboys currently are — have covered only 37.50 percent of the time. Although they have reached their implied totals 50 percent of the time, they on average have an actual/implied differential of -3.31 points. Traditionally, home favorites haven’t done all that well when facing divisional opponents. For what it’s worth, the dogs haven’t done all that well either. They’ve covered 62.50 percent of the time but reached their implied totals in only 37.50 percent of games.
In the 18 non-divisional games, we see an average over/under of 45.1 points with a 7.4-point spread. The over is hit 55.56 percent of the time. The favorites cover and reach their implied point totals in 72.22 percent of games; the underdogs, 27.78 percent.
These games also have a notable sub-split: The non-divisional road favorites — which the Steelers currently are — have reached their implied point total 100 percent of the time across six games. The underdogs have also hit their implied totals 50 percent of the time. With all of this scoring, the over has easily been hit in five of six games. On average, these games have 45.9-point over/unders and 7.25-point spreads. The average totals surpass the over/unders by 8.92 points, and the non-divisional road favorites have outscored their implied totals by a robust 9.08 points. The home dogs in this situation have just about broken even, implied to score 19.33 points and scoring 19.17.
One more notable split: If we remove the 13 games with the Lions and Cowboys, we are left with five games. In these non-DET/DAL games, which have a lower average over/under (43.8 points), the under is hit 60 percent of the time. Whence this underperformance? The dogs. The favorite has covered 100 percent of the time and reached its implied total in three of the games with an actual/implied differential of +4.05 points. The dog has a 100 percent failure rate in terms of covering or reaching its implied total, with a differential of -7.45 points. The sample is small — but it doesn’t bode well for the Colts, who (per our FantasyLabs News feed) are presently expected to be without starting quarterback Andrew Luck.
Let’s Take a Quick Assessment
Right now, based on the data we’ve perused, here’s the impression we might have of the six teams playing on Thanksgiving:
– As road underdogs, the Vikings and Redskins look pretty undesirable. The best that you can say about them right now is that they are playing divisional opponents. Divisional road dogs haven’t been horrendous. At least they have an implied/actual differential of only -0.06 points.
– As home favorites, the Lions and Cowboys seem to be in a good spot. Historically, such teams have tended to reach their implied totals 1.5 times more often than they haven’t. However, divisional home favorites have historically sucked. So there’s reason to be guarded about this spot based on historical performance.
– The Steelers are road favorites. Historically, this is a great spot made all the better by the fact that the Steelers are playing against a team outside of the AFC North.
– The Colts are home underdogs. Historically, that’s a negative spot that isn’t made much better by the fact that the Colts are playing a team outside of their division.
On a macro level, we’re starting to get a picture of these teams.
The Particulars
Now I want to focus just a little on the particulars of these situations. I’m not going to look at these games on a micro level. Instead, I’m going to leave that to Bryan Mears, who every week writes the NFL Data Dive for the prime time slate. This week he’s writing a Thanksgiving Data Dive. Be sure to check it out on Wednesday.
Right here, I want to look at just a few more factors particular to the various teams.
Jim Caldwell
Caldwell has been head coach of the Lions since 2014. In the eight years before Caldwell became head coach, the Lions were Thanksgiving dogs each year except for one. They covered the spread twice before him.
With Caldwell, the Lions have destroyed their implied totals with a differential of +14.13 points. Their opponents have produced a differential of -5.38 points in that span. Of course, the sample is small, and those games came against the Eagles (2015) and Bears (2014), both of which were horrible: The Eagles allowed the fifth-most and the Bears the second-most points per game. Still, it’s impressive that the Lions were able to surpass their implied totals so handily.
On the one hand, it’s hard to see this and not get a little aroused. Per our Trends tool:
On the other hand, it’s also hard not to get aroused by guys named “Matthew.”
I have this theory about Caldwell. The theory is that he basically sucks as a coach. He’s the kind of coach for whom extra time is bad, because it doesn’t benefit him — he doesn’t know what to do with the extra time — and it benefits opposing coaches, who can use extra time to get an edge on him.
Under this theory, one would assume that Caldwell’s teams would do relatively well in games played on a short week: Thursday games.
And that’s exactly what we see. In his six Thursday games as a head coach — three with the Colts (2009-2011) and three with the Lions — Caldwell’s teams have destroyed their implied totals with a differential of +8.25 points, reaching their implied total 100 percent of the time with an average of 31 points scored. Five of those six games were against divisional rivals.
I can’t believe that I’m saying this, but (relatively) it might actually be a good thing for the Lions that they’re playing on a short week. The disadvantage in preparation their head coach has isn’t as large as it usually is.
My saying that probably makes me sound like a d*ck, but whatever. I’m open to other theories — one of which could be that Caldwell’s teams have been fortunate to play against uninspiring defenses for all six of his Thursday games. That’s possible. And it is worth remembering that the Lions are facing the Vikings, who have a good defense anyway and are also a divisional rival.
But . . . Caldwell might have a little bit of Thursday magic to him.
—–
By the way, I wrote that entire section, and then Mears pointed out to me that Caldwell actually told reporters this week that he thought that having to prepare for the game on a short week could be good for the Lions.
Circle gets the square.
—–
Also, the six teams that have gone against Caldwell on Thursdays have reached their implied totals 66.67 percent of the time with a differential of +0.42 points. That’s not great, but the Vikings at least might not be total trash this week.
Jason Garrett
Garrett is (I believe) the anti-Caldwell. He’s the type of coach who gets better with more time to prepare. He needs more time to prepare. I would expect Garrett to be comparatively disadvantaged by a Thursday game.
And that’s what we see.
Since Garrett became the official Cowboys head coach in 2011 (after replacing Wade Phillips on an interim basis to finish the 2010 season), his team hasn’t done well in its six Thursday games. The Cowboys have been favored in each game — as they are now — and they’ve produced a differential of -2.46 points per game on Thursdays. Only once have they covered the spread.
The over has been hit in four of the Cowboys’ six Thursday games with Garrett, and the Cowboys have reached their implied total in three of those games, so it’s not as if they’ve been total trash, but opponents have covered and reached their implied totals 83.33 percent of the time with a differential of +6.79 points.
Within the context of ‘Garrett the Thursday Tactician,’ the Cowboys historically at least don’t look all that great this week, although this year could be very different. As for the visiting Redskins: Even though they’re playing against a division foe, the Redskins are in a good spot to take advantage of Garrett’s Thursday ineptitude. Under Garrett, Cowboys defenses have allowed 29.17 points per game on Thursday contests.
Garrett might be a dark horse for coach of the year — but he’s historically been something of a Thursday WOAT.
Chuck Pagano
In Pagano’s time as the Colts head coach, the team has played only one game on a Thursday. I don’t want to put too much (if any) stock in a single-game sample — but it is at least intriguing that the game was played last year when someone other than Luck was playing at QB. Again, Luck (concussion) isn’t expected to be active this week.
In their single Thursday game with Pagano, the Colts were road underdogs. Even without Luck, they managed to cover the spread and reach their implied total against a divisional opponent in the Houston Texans.
Mike Tomlin and Todd Haley
Tomlin has been the Steelers head coach since 2007, and Haley has been the offensive coordinator since 2012. Since 2007, the Steelers have played four times on Thursday, all of them with Haley as the coordinator.
One of those games was a Thanksgiving game on the road against the Ravens. Three of those games have actually been against the Ravens — because apparently the NFL thinks that low-scoring football games on Thursday nights are awesome. Three of the four games have seen the Steelers play on the road. And in all four of these games the Steelers have been underdogs.
Only once in these games did the Steelers manage to cover the spread or reach their implied team total. They have an actual/implied differential of -3.75 points. Only one of these games managed to hit the over. And in three of these games the Steelers allowed their opponent to reach their implied total.
So the Steelers have been #notgood on Thursdays. At the same time, it’s hard to read too much into this sample because three of the games were played within the division, three were played against the same team, and three were on the road.
Then again . . .
Ben Roethlisberger
The Steelers are on the road, where Roethlisberger has been an amazingly suboptimal QB, even with wide receiver Antonio Brown being perhaps the best WR in the NFL:
Do you know what’s sad? Ben’s been even worse as a road favorite:
The Steelers have a high implied total of 28.25 points, and the Colts have one of the worst pass defenses in Football Outsiders’ Defense-Adjusted Value Over Average (DVOA) . . . but that’s some awfully muddy water for drinking.
Matthew Stafford
Since entering the league, Stafford has done well as a home favorite. Per RotoViz:
In his six home games against the Vikings . . .
. . . Stafford hasn’t done quite as well.
The Lions are implied to score 22.75 points, and the Vikings defense is top-10 in pass DVOA.
As if by the end of this piece you needed more reasons to distrust a guy named “Matt.”
The Last Piece of Pecan Pie
Thanks for reading, and, whatever you do this year for Thanksgiving, may you experience the delicious thrill of eating the last piece of pecan pie.
———
The Labyrinthian: 2016, 92
This is the 92nd 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. Previous installments of The Labyrinthian can be accessed via my author page.
