indiegames
February 8th, 2010, 00:05
http://cache.gawker.com/assets/images/9/2010/02/500x_10.8.184.18-image185.jpgEA Sports (http://kotaku.com/tag/easports/) is mighty proud that its Madden NFL simulation has correctly predicted the Super Bowl (http://kotaku.com/tag/superbowl/) winner all but one time since it's run the promotion. But how did its predictions for the season, made back in August (http://kotaku.com/5333184/madden-simulates-nfl-season-we-offer-predictions-too), turn out?
To recap: On Aug. 4, a full month before the first regular-season kickoff, EA announced the division winners, playoff teams, MVP and all-Pro selections returned by a simulation using Madden NFL 10 (http://kotaku.com/tag/maddennfl10/). Granted, this was before the preseason even kicked off - before Brett Favre announced his return to Minnesota, which absolutely changed the NFC dynamic. Still, let's grade how they did.
The Madden sim accurately named just three of eight divisional winners and one of four wild card participants. Overall, it did get seven of the 12 playoff participants. And none of its playoff team predictions finished worse than .500. Given how the year started for the Titans and Panthers and how it ended for the Steelers and Giants, that's actually pretty good.
EA Sports said the AFC division winners would be New England, Pittsburgh, Tennessee and San Diego. Only New England and San Diego panned out. EA Sports said the NFC division winners would be Philadelphia, Arizona, Atlanta and Chicago. Just Arizona won its division, although Philadelphia made the postseason as a wildcard.
For the wildcard teams, EA said Indianapolis and Baltimore would get bids in the AFC. Both did visit the playoffs, but Indianapolis won its division with a 13-3 record. In the NFC, the Madden simulation said New York and Carolina would make the playoffs. Neither did; both finished 8-8.
That means the Madden simulation's division winners' combined, real-life records were six games worse compared with the records of real life division winners in the AFC (42-24 to 47-17). They were 7 1/2 games worse in the NFC (39-27 to 45-18).
Now, for the individual honors. The variables at work here make picking even a majority of actual NFL All-Pro first team selections a real longshot. But Madden only got three out of 25 right - Texans wide receiver Andre Johnson, 49ers linebacker Patrick Willis, and Browns kick returner Josh Cribbs. None of the awards Madden picked - MVP, Offensive Players and Rookies of the Year, Coach of the Year, etc. - came true.
What does this tell us? Dunno. The game's on. But it's one thing to simulate one game between two teams of 50 players and make five correct predictions in six years. Quite another to simulate a few hundred football games and thousands of events and come up with anything close to reality. It's probably an unreasonable expectation, especially considering how the rosters vary over the season.
So with that, enjoy Super Bowl XLIV (http://kotaku.com/tag/superbowlxliv/), and feel free to talk about it in the comments here.
And, apropos of nothing, on Friday Luke and I were talking about Rules We'd Like to See Added to the NFL. Being that he's an Australian who enjoys American football, he's got a pretty smart perspective on this sort of thing. I said I wanted to see a rule where a punting team that catches its own punt on the fly is awarded a free kick from placement on the spot it is caught - similar to the mark in Aussie rules (http://en.wikipedia.org/wiki/Mark_(Australian_football)). Luke pointed out the total destruction awaiting anyone attempting such a catch, and a fair-catch rule in such an instance would almost defeat the point. It would increase scoring and make the kicking game more relevant, however.
Luke instead said he'd like to see a penalty rule similar to rugby, in that for certain infractions (say, 10 yards or more) the ball is kicked across the sideline in the direction of the offending team's territory and where it goes out of bounds is where play resumes. While pass interference (fresh set of downs, and a spot-of-the-foul placement) is the closest U.S. football comes to this, I like Luke's idea better for adjudicating the game's harsher penalties.
http://ads.pheedo.com/img.phdo?p=1 (http://ads.pheedo.com/click.phdo?p=1)http://a.rfihub.com/eus.gif?eui=2226http://feeds.feedburner.com/~ff/kotaku/full?d=yIl2AUoC8zA</img> (http://feeds.gawker.com/~ff/kotaku/full?a=3AHJ4zbLkws:KOIqsZrN9x4:yIl2AUoC8zA) http://feeds.feedburner.com/~ff/kotaku/full?d=H0mrP-F8Qgo</img> (http://feeds.gawker.com/~ff/kotaku/full?a=3AHJ4zbLkws:KOIqsZrN9x4:H0mrP-F8Qgo) http://feeds.feedburner.com/~ff/kotaku/full?i=3AHJ4zbLkws:KOIqsZrN9x4:D7DqB2pKExk</img> (http://feeds.gawker.com/~ff/kotaku/full?a=3AHJ4zbLkws:KOIqsZrN9x4:D7DqB2pKExk) http://feeds.feedburner.com/~ff/kotaku/full?i=3AHJ4zbLkws:KOIqsZrN9x4:V_sGLiPBpWU</img> (http://feeds.gawker.com/~ff/kotaku/full?a=3AHJ4zbLkws:KOIqsZrN9x4:V_sGLiPBpWU)
http://feeds.feedburner.com/~r/kotaku/full/~4/3AHJ4zbLkws
More... (http://feeds.gawker.com/~r/kotaku/full/~3/3AHJ4zbLkws/grading-maddens-preseason-predictions)
To recap: On Aug. 4, a full month before the first regular-season kickoff, EA announced the division winners, playoff teams, MVP and all-Pro selections returned by a simulation using Madden NFL 10 (http://kotaku.com/tag/maddennfl10/). Granted, this was before the preseason even kicked off - before Brett Favre announced his return to Minnesota, which absolutely changed the NFC dynamic. Still, let's grade how they did.
The Madden sim accurately named just three of eight divisional winners and one of four wild card participants. Overall, it did get seven of the 12 playoff participants. And none of its playoff team predictions finished worse than .500. Given how the year started for the Titans and Panthers and how it ended for the Steelers and Giants, that's actually pretty good.
EA Sports said the AFC division winners would be New England, Pittsburgh, Tennessee and San Diego. Only New England and San Diego panned out. EA Sports said the NFC division winners would be Philadelphia, Arizona, Atlanta and Chicago. Just Arizona won its division, although Philadelphia made the postseason as a wildcard.
For the wildcard teams, EA said Indianapolis and Baltimore would get bids in the AFC. Both did visit the playoffs, but Indianapolis won its division with a 13-3 record. In the NFC, the Madden simulation said New York and Carolina would make the playoffs. Neither did; both finished 8-8.
That means the Madden simulation's division winners' combined, real-life records were six games worse compared with the records of real life division winners in the AFC (42-24 to 47-17). They were 7 1/2 games worse in the NFC (39-27 to 45-18).
Now, for the individual honors. The variables at work here make picking even a majority of actual NFL All-Pro first team selections a real longshot. But Madden only got three out of 25 right - Texans wide receiver Andre Johnson, 49ers linebacker Patrick Willis, and Browns kick returner Josh Cribbs. None of the awards Madden picked - MVP, Offensive Players and Rookies of the Year, Coach of the Year, etc. - came true.
What does this tell us? Dunno. The game's on. But it's one thing to simulate one game between two teams of 50 players and make five correct predictions in six years. Quite another to simulate a few hundred football games and thousands of events and come up with anything close to reality. It's probably an unreasonable expectation, especially considering how the rosters vary over the season.
So with that, enjoy Super Bowl XLIV (http://kotaku.com/tag/superbowlxliv/), and feel free to talk about it in the comments here.
And, apropos of nothing, on Friday Luke and I were talking about Rules We'd Like to See Added to the NFL. Being that he's an Australian who enjoys American football, he's got a pretty smart perspective on this sort of thing. I said I wanted to see a rule where a punting team that catches its own punt on the fly is awarded a free kick from placement on the spot it is caught - similar to the mark in Aussie rules (http://en.wikipedia.org/wiki/Mark_(Australian_football)). Luke pointed out the total destruction awaiting anyone attempting such a catch, and a fair-catch rule in such an instance would almost defeat the point. It would increase scoring and make the kicking game more relevant, however.
Luke instead said he'd like to see a penalty rule similar to rugby, in that for certain infractions (say, 10 yards or more) the ball is kicked across the sideline in the direction of the offending team's territory and where it goes out of bounds is where play resumes. While pass interference (fresh set of downs, and a spot-of-the-foul placement) is the closest U.S. football comes to this, I like Luke's idea better for adjudicating the game's harsher penalties.
http://ads.pheedo.com/img.phdo?p=1 (http://ads.pheedo.com/click.phdo?p=1)http://a.rfihub.com/eus.gif?eui=2226http://feeds.feedburner.com/~ff/kotaku/full?d=yIl2AUoC8zA</img> (http://feeds.gawker.com/~ff/kotaku/full?a=3AHJ4zbLkws:KOIqsZrN9x4:yIl2AUoC8zA) http://feeds.feedburner.com/~ff/kotaku/full?d=H0mrP-F8Qgo</img> (http://feeds.gawker.com/~ff/kotaku/full?a=3AHJ4zbLkws:KOIqsZrN9x4:H0mrP-F8Qgo) http://feeds.feedburner.com/~ff/kotaku/full?i=3AHJ4zbLkws:KOIqsZrN9x4:D7DqB2pKExk</img> (http://feeds.gawker.com/~ff/kotaku/full?a=3AHJ4zbLkws:KOIqsZrN9x4:D7DqB2pKExk) http://feeds.feedburner.com/~ff/kotaku/full?i=3AHJ4zbLkws:KOIqsZrN9x4:V_sGLiPBpWU</img> (http://feeds.gawker.com/~ff/kotaku/full?a=3AHJ4zbLkws:KOIqsZrN9x4:V_sGLiPBpWU)
http://feeds.feedburner.com/~r/kotaku/full/~4/3AHJ4zbLkws
More... (http://feeds.gawker.com/~r/kotaku/full/~3/3AHJ4zbLkws/grading-maddens-preseason-predictions)