Discussion in 'PatsFans.com - Patriots Fan Forum' started by SVN, Feb 2, 2012.
FOOTBALL OUTSIDERS: Innovative Statistics, Intelligent Analysis | Super Bowl XLVI Preview
Wow, a bunch of stat heads who are confident in their system basically admit they have no idea how to size up this game.
All that statistical masturbation, which they say favors the Patriots, and then they conclude the Giants win in a squeaker?
I prefer the discipline index Parcells champions.
Still had some useful statistical information that isn't found publicly anywhere else.
At least at FO they are writing about football. Not sure what the NY tabloids are gossiping about.
It's refreshing to see someone actually analyze available information to try and draw a conclusion. All I've seen or heard for ten days is superficial sound bites based on a meaningless stat or an irrelevant ranking. Essentially people have been doing just the opposite of what FO set out to do: they made an opinion and then looked for a stat, ranking or piece of information to justify it.
The original consensus, before the New York papers and world wide leader started spouting off was that these were two very evenly matched teams, each with strengths and weaknesses that meant that the game could easily go either way. That opinion has been drowned out by all the noise emanating from New Jersey. Schatz gets a bit flip-floppy and wishy-washy with his conclusion, but basically what he is saying is that it's his opinion that it is too close to have a strong conviction in the outcome. At least he came up with a good reason to take that position.
The basic theme is that stats over the full season favor the Patriots, but the Giants have strongly been opposing those stats the last few weeks. In addition, strong X-factors such as the Gronk injury and healthy situation along with the sudden strong defensive play from both teams play into the decision. It's not outside of logic to stray from season stats when you have one of the best players on a team hurt and two teams playing much differently (and better) than they have all year. As a Giants fan that isn't buying into the media hype, I think any person looking at this game from a neutral perspective should basically look at it as a coin toss and not an easy game to read.
This makes no sense. They keep talking about the Giants front four... but they dont mention the Pats O-Line (the best in the league). Also, the amount of games they took for analysis is casually during the Giants winning streak. Why instead of 5 games they don't take 10? Or 8 or 6... very convenient.
Still as Tune says numbers gives the edge to the Pats, FO picks the Giants.
Sunday cant get here soon enough...
Interesting read, thanks for posting.
CHFF has a good read on the Pats D:
The Curious Case of the New England defense | Cold Hard Football Facts
A Couple of other good articles there also.
Ignore the mediots.
Where all this statistical analysis falls short is in failing to quantify the Myra Factor, the Giants Gumflapping Factor, the Eli Goober Factor, and the Moronic New York Tabloids Factor.
All of these stats an analysis and nobody mentions the simple fact special teams defense allows just as many yards as your actual defense.
Your defense amounts to about only 50% of your total defensive field position given up by your defense. Every time you look at your defense in terms of yards per game or yards per drive, you're looking at only half your defense.
If people are interested in yards can someone please explain to me who really gives a crap that Washington is
-ranked 12th in yards per game
-if they are 15th in yards per drive with 29.45 yards per drive
-when their special teams gives up 30.66 yards per drive(28th)
30.66 > 29.45
Who cares that Jacksonville is ranked 6th in yards per drive with 26.12 when their special teams give up 29.74 yards per drive.
29.74 > 26.12
Over 50% of Washington's or Jacksonville's total "yards allowed" come from their special teams defense, or even your quarterback's picks, which is never tracked as part of the team's total defense. And nobody talks about it. This is the case for most NFL defenses.
The more yards your special teams gives up, the less yards your actual defense tends to "give up". Only a few defenses does both, and those are the only ones yards stat might actually match up for. So if you're going to look at yards at least count the full freaking thing in terms of field position given up. And even then, it's a terrible measurement. But looking at half of it, you're bound to have every defense completely wrong.
The NFL really screwed everybody up when they called half of a team's defensive yards allowed "Total defense". The name doesn't actually represent the meaning of the word folks. NE's #2 special teams(24.06 yards per drive) have a lot to do with NE's true total defense success.
Vacuum stats are useless. They're just there to confuse you. Yards is one of them.
I don't want to knock Schatz because I really like what FO is all about. Still, it seems clear that he is hedging his bets here. If the Patriots win, his models were correct. If not, he called the outcome.
I think its more of a case of him being burned the last several years, as he mentions. Statistically there isn't much to the notion of teams "getting hot", but several hot teams have made it to the superbowl recently and have played very well which flies in the face of that notion.
I think after reading all of these predictions, especially those with quality analysis, the pattern seems to be that the game is really too close to call but they have to pick someone. So what I take out of each of these isn't the prediction itself but rather the analysis behind it. What I like about FOs analysis is that they are one of the few that target the Patriots running game as a key element in this game. I agree with them there.
I think it's more that it's just difficult to know what sample to use. If you use the whole season the Patriots will win comfortably. If you use the last 5 games the Giants will probably win a nail biter. And how healthy is the Gronkle, a huge part of our offense. I actually feel the same way he does, I could see the small sample of the Giants playing well falling apart and leading to an easy Patriots win but I also don't trust our secondary at all - even though they've been solid for a while - and could see the Giants taking advantage of that.
Yeah but the guy uses the most irrelevant, meaningless specific statistics, writes 3 pages on it, and can't even come to a solid conclusion.
At least pick a definitive side if you're going to do all that.
"Well the Patriots look better overall based on their entire body of work, but the Giants look about even if we look over the past couple of months, with the exception of the game when they faced each other, but it's an aberration anyway, and in these selective stats here's how each team looks, so the Patriots should win, unless the Giants pull out a tight win, though another aberration is not expected, but since it did the first time, who knows if it will happen again, and I can't explain it, therefore I predict, it might go either way....maybe. DVOA rules!" -Aaron Schatz.
Separate names with a comma.