Aikman Efficiency Ratings

[thesmee1]

For what it's worth.. It's nice to see the Colts' offense rated way below ours.

http://troyaikman.wordpress.com/category/aikman-efficiency-ratings/

 

[emoney_33]

Whats his formula for this poop. Our defense below average?

 

[Satchboogie3]

His formula must be utter bull****. He has the Colts, Packers, Texans, Vikings..., BROWNS, BENGALS!!!! Listed ahead of the Patriots. And this is supposed to be through week 2. How on Earth are 2 teams they gave up more than 40 points each in one game manage to get a better defensive rating than the Patriots who are OFFICIALLY ranked #1 in Defense? This guy needs to lay off the drugs. Oh no, another scandal coming...

 

[unoriginal]

His formula must be utter bull****. He has the Colts, Packers, Texans, Vikings..., BROWNS, BENGALS!!!! Listed ahead of the Patriots. And this is supposed to be through week 2. How on Earth are 2 teams they gave up more than 40 points each in one game manage to get a better defensive rating than the Patriots who are OFFICIALLY ranked #1 in Defense? This guy needs to lay off the drugs. Oh no, another scandal coming...

Browns are second to last there chief. Bengals are also below the Pats.

 

[Satchboogie3]

Opps, must have accidentally scrolled back up. Anywho, doesn't change the fact that that is just plain wrong. Ranking 21 teams higher on Defense when we are ranking #1 officialy is so no right. This guy must have gone to Cape Cod Community College.

 

[JJDChE]

A defensive ranking of 22 for the pats seems kind of hard to fathom under any ranking system.

I think the Aikman system probably relies heavily on red zone performance. I believe we've only allowed a team in our red zone four times and they've scored TDs all four times. It could be something like that.

 

[Satchboogie3]

And if that is the case, as I said, it is flawd.

 

[JJDChE]

And if that is the case, as I said, it is flawd.

Well, it's a small sample size. The rankings are trying to predict winning tendencies, not a compilation of defensive stats. Our defensive stats are just skewed away from the metrics those ratings are tracking right now. It'll work itself out.

 

[Fencer]

Opps, must have accidentally scrolled back up. Anywho, doesn't change the fact that that is just plain wrong. Ranking 21 teams higher on Defense when we are ranking #1 officialy is so no right. This guy must have gone to Cape Cod Community College.

We look at five major categories - yards per play, turnovers, points, first downs and red zone performance.

I believe Patriot opponents have a 100% success rate in the red zone so far this season, which is also why the points-against results are good but not great.

3 takeaways in two games is also nothing special.

Now, I don't like a below-average outcome for the Pats. I think the defensive result was OK in the first game, and excellent in the second, so they should have an above-average rank overall, perhaps high single-digit. But he's also not crazy.

 

[FirstAndGoal]

It's probably based on factors that relate to things like opponet's previous production. That would explain it for our defense. The Jets and Chargers have faced two stiff defenses so they haven't been that productive. I imagine that once a team like the Chargers gets going, this will be more accurate. Remember, it's only got two games so far. Many statistical models require a minimum sample size before they start being acurate.

 

[unoriginal]

And if that is the case, as I said, it is flawd.

I think Belichick would agree with Aikman that Red Zone performance is one of the more important defensive statistics. This has always been a bend-but-don't-break D.

 

[Satchboogie3]

We'll yes, but I think the most flawed thing is not taking into account the opposing teams. We destroyed (defensively) one of the top 3 offenses in the NFL. I think a teams ratings should go beyond just stats...

I've never liked formulas for sports. There are soo many non-statistic factors that are important.

 

[emoney_33]

We'll yes, but I think the most flawed thing is not taking into account the opposing teams. We destroyed (defensively) one of the top 3 offenses in the NFL. I think a teams ratings should go beyond just stats...

I've never liked formulas for sports. There are soo many non-statistic factors that are important.

They were one of the top offenses in the NFL last year. I guess the formula only takes into account this year where the Chargers, in 2 games, have not been a top offense (29th in his ratings).


As for stats, with the right formula and a large enough sample size, they can tell the story an extremely high percentage of the time. Finding the right formula however, well that's another story.

 

[FirstAndGoal]

They aren't one of the top three this year and that's my point. I'm sure his model doesn't take last year's stats. Maybe it should but I'm guessing it doesn't. You can't measure "good" so even though you think their offense is good, it's not a quantifiable measurement. You can only measure descrete values the Chargers offense sucked in the only other game they played because it was the Bears and the Jets sucked because they played the Ravens. When you measure production to production with such a small sample, that's what you get. It doesn't mean that the model is flawed. It just means that more numbers need to be plugged in.

 

[FirstAndGoal]

They were one of the top offenses in the NFL last year. I guess the formula only takes into account this year where the Chargers, in 2 games, have not been a top offense (29th in his ratings).


As for stats, with the right formula and a large enough sample size, they can tell the story an extremely high percentage of the time. Finding the right formula however, well that's another story.

Looks like we were thinking the same thing.

 

[unoriginal]

We'll yes, but I think the most flawed thing is not taking into account the opposing teams. We destroyed (defensively) one of the top 3 offenses in the NFL. I think a teams ratings should go beyond just stats...

I've never liked formulas for sports. There are soo many non-statistic factors that are important.

I don't think San Diego has a Top 3 offense. Us, Colts, Bengals, Cowboys...

San Diego should be top 10 this year once LT gets going.

 

[Fencer]

Many statistical models require a minimum sample size before they start being acurate.

And this one will never be accurate. It's just a made-up formula, not tested in any serious way.

Even so, your comment applies. This model is also subject to wild early-season fluctuations due to small sample size. E.g., does anybody seriously think the Pats' red-zone performance is indicative of their likely full-year result?

 

[Fencer]

I'm sure his model doesn't take last year's stats. Maybe it should but I'm guessing it doesn't.

It doesn't account for opponents' stats or strength AT ALL.

 

[DarrylS]

He has us at 71.5 for Defense, which is below the league average of 74.3, there are statistics and damed lies.. in this case it is the latter and this negates anything else he says.

 

[Lloyd_Christmas]

Now, I don't like a below-average outcome for the Pats. I think the defensive result was OK in the first game, and excellent in the second, so they should have an above-average rank overall, perhaps high single-digit. But he's also not crazy.

Maybe not crazy, but his system has a MAJOR flaw....

100% failure rate in the redzone doesn't mean jack S_H_I_T if you keep your opponent out of the redzone for 95% of the game, resulting in very little point allowed.

Seems to me that his 'System' doesn't put enough weight on pts allowed/game and the ability of a defense to keep the opponent out of the redzone for more than twice a game.