Statistics geeks favor Miami | FinHeaven - Miami Dolphins Forums

Statistics geeks favor Miami

NYCphan

NYCphan

Stuck in Jetsville
Joined
Dec 3, 2002
Messages
1,205
Reaction score
1
Location
New York
I’ve been reading an interesting site using a statistics-driven, mathematical approach to rating teams and probable outcomes. If you like this kind of thing, like I do, you’ll really enjoy it. And if you don’t, you should still like what they have to say about Miami right now.

Advanced NFL Stats uses a formula to rate teams by their offensive, defensive and combined efficiency. Right now, it ranks Miami #2 in offense (lack of turnovers clearly boosting the score), #19 in defense, and #9 overall. Statistically, that translates into beating an average NFL opponent 66% of the time at a neutral site.

The Pats and Jets are ranked 20 and 21.

Based on that, they use the same ratings to calculate the probabilities of this week’s games, giving Miami a 77% chance of beating the Chiefs, New York a 57% chance of beating the Seahawks, and New England a 52% chance of beating the Cardinals.

Finally, they rank the league's luckiest teams, meaning ones that have won more games than their efficiency predicts they should. Tops on the list is the Jets, who have won three more games than their statistical performance would predict. The Pats are second.

Another site, NFL Forecast, takes those efficiency ratings, adjusts slightly for a home-field advantage and simulates the remaining game results 5,000 times, in order to calculate the probability of each team making the playoffs.

It estimates Miami has a 56% chance of winning the division and a 2% chance of winning a wildcard. Jets are next (35% division, 5% wildcard), followed by the Pats (9% division, 18% wildcard).

There are quite a few other sites like this with statistics-based rankings. Football Outsiders is another interesting one. It’s DVOA system (used at Fox, I think) ranks Miami #15, ahead of the Jets (17) and Pats (20).
 
The basic fundamental problem I have with statistics is that they are only accurate indications of a teams past performances but really have no basis for their future performances. Using statistics to predict the outcome of a game is a very flawed approach, IMO.
 
Miami and Atlanta having the two most efficient offenses is interesting and fairly true. both teams are more than capable of moving the ball with ease and controlling the clock and minimizing mistakes. and i'd say our defense being 19th sounds very accurate, too.
 
Statistical methods hate us because turnovers are really bad, more so to statistical evaluators. A turnover not only takes away a chance at points for us but gives the other team a chance at points. I know it sounds obvious but that's why they are so crucial to formulas like that.
 
Statistics are great to get approximation of what is supposed to happen, but there are variables in the real world that you can not account for solely using statistical formulas.
 
shula_guy said:
The basic fundamental problem I have with statistics is that they are only accurate indications of a teams past performances but really have no basis for their future performances. Using statistics to predict the outcome of a game is a very flawed approach, IMO.
Click to expand...

I don't disagree. It's a question I would pose to Vegas though.
 
LOL. Hey man, the stats say what they say. I'll gladly cite these as a means of needling my Patriots and Jets fan friends. ;)
 
ckparrothead said:
LOL. Hey man, the stats say what they say. I'll gladly cite these as a means of needling my Patriots and Jets fan friends. ;)
Click to expand...


Will you at least conceede the fact that yes when the numbers favor your argument they are fun to use against rival but there really is little substance to them when you look at them in the microscopic perspective. Which is my biggest issue. Alot of stats are over or under valued without propperly factoring in varialables.


Marino613

To you I would say if we are dicussing point spreads by vegas, I think as inaccurate as they are they are the best at a very fallable scientific approach. If you do a history of predictions on outcomes, I would readily agree that Vegas is probably alot more accurate then most but if you look at their sucess rate of getting it right it is probably a dismally low percentage. If they got it right consistently they would not hedge their bets with one another and generally professional gamblers would bet with the house.
 
You must log in or register to reply here.
Back
Top Bottom