The Denver Broncos are on their way to breaking records this season, having scored 230 points in five games. That's 51 percent more points than the second-highest scoring team and over twice as many points as the average N.F.L. team.
You would think that a team like Denver wouldn't give up many points because its offense moves the chains and rarely turns the ball over to put the defense in bad situations. But Denver's defense has been seemingly poor, allowing the fifth-most points so far in 2013.
It has long been suspected that defenses on teams with excellent offenses are underrated by their statistics. Defenses may be strategically trading away yards and points in exchange for time off the clock, increasing their teams' chances of winning at the expense of their statistical rankings. It seems plausible, but how can we tell if it's true?
To untangle the relationship between offensive and defensive production, we need to use the concept of Expected Points. Expected Points measures the net scoring potential of each combination of down, distance and yard line. For example, consider a situation in which a team turns the ball over at its 20-yard line, and its opponent goes on to score a touchdown. Instead of counting as 7 points against the defense alone, it would count as about 4 Expected Points "Added" against the offense and only about 3 more against the defense.
Perhaps we should expect defenses on teams with great offenses to be relatively poor, regardless of game-strategic considerations. The zero-sum nature of the league's salary cap means that every cap dollar that goes to a top quarterback can't go to a pass rusher. But in general, offensive and defensive Expected Points Added is correlated only very weakly, at 0.07 on a scale from zero to one (no systemic association to lockstep connection).
We get a very different result when we look only at teams with the very best offenses in recent years. Since the 2000 season, the 32 teams with the best offenses allowed 2.1 more Expect Points per game than the average team over the same period — a significant difference. This translates to about a 4 percent chance of winning a game when matched against a roughly equal opponent, and slightly less when matched against a lesser opponent.
So it's likely that statistical models, like the one I use to estimate weekly game probabilities, do undervalue the defensive strength of teams with top offenses. But the effect isn't large and won't often make the difference between favorite and underdog. And it's not likely to make any difference to the Broncos this Sunday, who are favored by four touchdowns over the struggling Jaguars.
At right are the game probabilities for Week 5.
Brian Burke operates Advanced NFL Stats.
Anda sedang membaca artikel tentang
Week 6 N.F.L. Game Probabilities
Dengan url
https://suporterfanatikos.blogspot.com/2013/10/week-6-nfl-game-probabilities.html
Anda boleh menyebar luaskannya atau mengcopy paste-nya
Week 6 N.F.L. Game Probabilities
namun jangan lupa untuk meletakkan link
Week 6 N.F.L. Game Probabilities
sebagai sumbernya
0 komentar:
Posting Komentar