Football Outsiders predicts us as 10th best team, missing playoffs | FinHeaven - Miami Dolphins Forums

Football Outsiders predicts us as 10th best team, missing playoffs

dolphone

In memoriam
Super Donator
Joined
Aug 8, 2011
Messages
961
Reaction score
15
Location
Mexico City
According to their latest season predictions, Football Outsiders predicts we're going to be about the 10th best team in the NFL in overall performance. Their model sees a top-10 offense (9th), top-half defense (12th), and lousy special teams (23rd).

Since they also predict the Pats as division champs and the Jets and Ravens as wildcards, we would end up out of the playoffs. The difference isn't that big (we're projected for 8.5 wins, BAL/NYJ for 9.5/9.6 respectively), but it would mean an 8-8 or 9-7 record, and out of playoff contention.

-- ADDENDUM --

I forgot to say this outcome is mostly related to our schedule. We're currently facing the 4th hardest schedule in the league (according to them, and their predictions, obviously). Makes it kind of hard to succeed even with a good team...

-- ADDENDUM --

For full disclosure, here's a track record of their notable hits and misses (in preseason projections):

2004: predicted a bounceback season for the NYJ (6-10 in 2003, 10-6 in 2004), as well as the meteoric rise of the Chargers (4-12 in 2003, 12-4 in 2004).

2005: predicted a NE decline in wins, but still holding on to the division. Missed badly on Miami (predicted around 5 wins, actually had 9). Had Indianapolis with the highest projected wins in the entire AFC. Missed awfully on the Eagles, forecasting around 12 wins and practically guaranteeing playoffs; conversely, had the Giants as a lousy team (the NYG won the division that year at 11-5, and Philly completed a crappy 6-10 season). Correctly saw TB improving dramatically, and Atlanta declining as sharply. Pretty much guaranteed a deep playoff run for the Seahawks, as the second best team in the NFC after the Eagles.

2006: Correctly saw a Miami downturn. Pegged Baltimore as a below average team (actual record: 13-3). Accurately predicted a great season for Indy and SD. Badly missed on the Steelers (projected high, actual 8-8) and the Seahawks (projected as best team in NFL, actual 9-7). Had CHI as easily the best defense in the league, by a wide margin. Projected New Orleans as the worst team in the league (this was the Katrina "recovery" year, when they went to the playoffs).

2007: Correctly predicted Miami's collapse, particularly on offense (pegged as the worst in the league); also correctly saw NE running away with the division and had them as the best team in the league by a wide margin. Badly missed on the NFC East and AFC North, predicting Philly and Baltimore as division champs while forecasting the Cowboys as bottom-dwellers (both predicted champs ended up as the worst in the division, while Dallas soared to a 13-3 record). Nailed the NFC North and South, predicting huge bouncebacks from Green Bay and Tampa. Also correctly saw a Jaguars improvement, but overestimated it (had them as division champs).

2008: About everything went wrong this year.

2009: Correctly saw Indy and SD as the two best teams in the AFC. Had the Jets and Bengals as below average teams (they both went to the playoffs) and saw division championships for the Giants (8-8), Seahawks (5-11) and Panthers (8-8), while missing on New Orleans' monster of a season. Incorrectly saw Arizona declining dramatically. Correctly saw an upswing for the Vikings.

2010: Correctly saw Pittsburgh, Baltimore, Indy and NE as a closely grouped bunch of the best teams in the AFC. Predicted the division championship of the Chiefs. Had Miami as a wildcard and the Jets declining. Correctly saw San Diego's regression. Correctly saw Philadelphia and Chicago bouncing back, and also had Green Bay and Atlanta as very strong teams and division champions. Missed badly on the Cardinals (8-8 division champs, actual 5-11) and predicted the Redskins as a wildcard.
 
Last edited by a moderator:
I'm sure if I spouted out a crap load of predictions, one of them is bound to be correct.
 
I'm sure if I spouted out a crap load of predictions, one of them is bound to be correct.

Well, they don't just shoot from the hip with these things. They have a(n improved over the years) mathematical model, and they seem to be generally succesful with their predictions. They have accurately expected things no one else had, and it's hard to say it's just dumb luck.
 
@dolphone
Seems to me they are about 50/50, and almost all their mistakes were from assuming that the good teams stayed good and the bad teams kinda lolly gagged. The definition of CHANCE
 
@dolphone
Seems to me they are about 50/50, and almost all their mistakes were from assuming that the good teams stayed good and the bad teams kinda lolly gagged. The definition of CHANCE

The model (AFAIK) doesn't "assume" "that the good teams stay good and bad teams kinda lolly gagged". There are some presumptions built into the model from historic observation (e.g. that offenses tend to be more stable year-to-year than defenses) but they consider a lot of variables when running their simulations.

I do agree they're about even with great hits and lousy misses. That's still much better than most other prognosticators (who are once in a blue moon hitters and constant misses).
 
You have to throw in injuries, which I assume are not calculated. How does Peyton not starting affect the scores? If Brady rips an ACL, it could change everything. Let's play, c'mon Monday Night Football!!!
 
You have to throw in injuries, which I assume are not calculated. How does Peyton not starting affect the scores? If Brady rips an ACL, it could change everything. Let's play, c'mon Monday Night Football!!!

Peyton's injury has been taken into consideration, and they also have a statistical model of injuries that assumes (underline that) teams with bad injury luck one year will do better next year and teams with good injury luck will do worse. This is called regression towards the mean and is not a great predictor of future outcomes, but it's pretty clever and does have some predictive ability.


That being said, trying to explain FO in this forum is like Don Quijote fighting the windmills.
 
Peyton's injury has been taken into consideration, and they also have a statistical model of injuries that assumes (underline that) teams with bad injury luck one year will do better next year and teams with good injury luck will do worse. This is called regression towards the mean and is not a great predictor of future outcomes, but it's pretty clever and does have some predictive ability.


That being said, trying to explain FO in this forum is like Don Quijote fighting the windmills.

Trying to explain FO to *anyone* is like that. It's a complex model.
 
to say we will be a top 10 offense at this point is quite a stretch
 
Vegas is the best prognosticator, and will be dead on, as usual.
 
Vegas is the best prognosticator, and will be dead on, as usual.

You do know their prognostications are made with the goal of earning the most money, not necessarily being the most accurate, right? Whatever gets about the same money from both sides is the best projection for them. They're aiming for public perception, not accuracy.
 
Back
Top Bottom