With free agency and the draft in the rear view, it's time to look ahead to the Buccaneers season and play everyone's favorite game: How Many Games Will They Win?
The South Point casino in Las Vegas recently released its projections. It set the over/under for Tampa Bay at eight wins.
Eight? For a team that won nine last season and narrowly missed the playoffs? After it signed receiver DeSean Jackson and drafted tight end O.J. Howard, you'd think it'd be projected to win more.
And get this: The casino's sports book set a higher over/under for every other team in the NFC South. The Falcons are projected to win 10, the Panthers nine and the Saints 8.5.
Or maybe the over/under for the Bucs is right where it should be.
How do oddsmakers determine over/unders?
For South Point sports book director Chris Andrews, his process begins at the end of a season, which is when he develops a "power rating" for each team. It's part conjecture, part science. Points scored and points allowed, for example, are among the variables he considers. (More on points scored and points allowed in a moment.)
He assesses which teams got better and which got worse during the offseason, giving more weight to free-agent transactions than draft picks, and adjusts his rating accordingly. Then he pores over the NFL schedule for about two weeks, calculating the win probability for each team in each game. To get a team's season win total, he adds up the probabilities for each of its games.
So, no, he doesn't hate your team. He's not trying to disrespect your team. He's just trying to elicit bets.
"You strive to be objective and mathematical," he said.
If you're using last season's win totals as a reference point for predicting this coming season, you might feel South Point is underestimating the Bucs.
Is there a better measure of team quality than wins?
We like to think of football as a game of will, but it's full of luck. A 16-game sample is too small to mitigate the influence of random results. With apologies to Hall of Fame head coach Bill Parcells — "You are what your record says you are" — a team's record does not always accurately reflect its performance.
This brings us back to points scored and points allowed, which, on the average, are better representations. Plus, they're more correlated to wins the next season than even wins are.
It's a concept rooted in Bill James' baseball research. Decades ago, he developed a formula that estimated the number of games a team "should" win based on the runs it scored and the runs it allowed. Football statisticians have since adopted and adapted that formula to gauge team performance.
Here's the formula for a team's "expected wins" (there's no test afterward, I promise): points scored^2.37/ (points scored^2.37+ points allowed^2.37). Then multiply the outcome by the number of games played in a season (16).
How fair is an eight-win forecast for Tampa Bay?
South Point's projections bear a remarkable resemblance to 2016 expected wins. For 25 of the NFL's 32 teams, there is no more than a win difference.
The Bucs are one of the 25. If a team scores 354 points and allows 369, as Tampa Bay did last season, you'd expect it to win 7.6 games. By this measure, the Bucs overachieved.
What about the Saints? An 8.5 over/under is reasonable when you consider that they underachieved last season — at least based on their point totals. They "should" have won 8.3 games.
It's natural, fresh off the free agency and draft buzz, to rationalize a bump from 9-7 to 10-6, or even 11-5. If we use expected wins, however, instead of actual wins to inform our projections, a scenario in which the Bucs improve and yet finish with the same record seems plausible.
Expecting Tampa Bay to accomplish much more is a risky gamble.
Contact Thomas Bassinger at firstname.lastname@example.org. Follow @tometrics.