Analyzing the Analyzers

A couple of stories were linked to way down in the mega-comments on the Reid post.  The first is by the guy whose name is two sandwiches and has the sort of statistics that you just know are going to become very popular talking points in this town:

They're now 0-4 this year in games decided by seven points or less, with all four losses coming to winning teams...

After going 2-5 last year in games decided by a touchdown or less and 4-5 in 2006, they're 6-14 in close games over the past three seasons — with three of the wins coming during Jeff Garcia's late-season run in 2006.

Take out the games the backups started while Donovan McNabb was hurt, and the Eagles are 2-10 in the last 12 close games McNabb has started.

In 10 years under head coach Andy Reid, the Eagles are 101-65 overall, which is .605 winning percentage — one of the best in NFL history. But in games decided by a touchdown or less, the Eagles under Reid are only 36-36.

The Eagles' performance against winning teams is equally poor. Since 2005, the Eagles are 13-26 against teams that are .500 or better. That's a .333 winning percentage...

Starting with Super Bowl XIX in Jacksonville after the 2004 season, the Eagles are an astonishing 6-17 against .500 or better teams in games decided by a touchdown or less.

I want to make a few points here, one macro and the rest micro. 

The big picture issue here is that it a) shouldn't be surprising that the Eagles do better against bad teams than good ones and b) the fact that the Eagles have lost a lot of close games doesn't tell you very much about why the Eagles lost a lot of close games.

Let's take the Giants as an example.  Because the Eagles failed to convert that fourth-down run, many people are taking that game as evidence that this team somehow just can't get it done in the clutch.  Hey, I don't know if that's true or not -- it certainly seems like it has been the last couple years -- but this seems like a complete misreading of that game.

The Giants dominated the Eagles on Sunday night.  If the game had really been as close as the score indicated, I wouldn't even be that upset right now.  It wasn't.  In that game, the Giants were a far better team.

So what does it say that the game came down to the last possession before being decided?  Is that an indictment of the Eagles' "clutchness" and "get 'er doneness"?  Or is it really more an indication that this team actually has a fair amount of fight and don'tquit in it, such that they were in a game they had no business winning?  I think it's the latter.  And I don't even think it's really arguable. 

This is one case where the stats -- without context -- are pretty meaningless.

Ok, now the micro points:

They're now 0-4 this year in games decided by seven points or less, with all four losses coming to winning teams. Their five wins have been by an average of 18 points a game, all but one by 13 or more points.

This is part of the problem with these stats.  The "one" exception to the blowout wins was the Steelers game, in which the only reason the Eagles won by nine is because of a late safety and field goal that helped nail the door shut.  For 26 minutes in the second half, this was a four-point game.  By any realistic assessment of what happened that night, this was a close game.  But the cherry-on-top points at the end take it out of that category.

Check out last year's second game against the Redskins for a similar example.  That was a 15-13 game through three quarters (with the Eagles down), but then the good guys piled up a 20-10 advantage in the fourth to make this an eight-point win.  When you adopt an arbitrary cutoff point, you strip away a lot of detail.

After going 2-5 last year in games decided by a touchdown or less and 4-5 in 2006, they're 6-14 in close games over the past three seasons — with three of the wins coming during Jeff Garcia's late-season run in 2006.

Take out the games the backups started while Donovan McNabb was hurt, and the Eagles are 2-10 in the last 12 close games McNabb has started.

And:

Here are the six — at the Chiefs early in 2005 in that big comeback when T.O. had a huge game (11-for-171); by three points over the Chargers a few weeks later; over the Panthers and Giants (in the playoffs) with Garcia at QB in 2006; and last year in Minneapolis over the Vikings and late in the season in Dallas after the Eagles had been virtually eliminated from postseason contention.

This is skillful propagandizing.  Early in the piece he's doing everything he can to blame McNabb without saying it out loud, just to slip it past your defenses.  Then, in this later paragraph, look how he tries to devalue any of the good wins the Eagles actually did have. 

Oh, and by the way, TO's "huge game"?  McNabb threw for 369 yards and three TDs on a day when the team managed all of 28 rushing yards.  And yes, he was playing with the sports hernia, chest contusion and ankle problem that day. 

Furthermore, that Chiefs game is another good example of how late scores can change things.  The Eagles actually had a 13-point lead until the Chiefs scored with 1:33 left in the game.  Failed onsides kick and the game was over. 

The Eagles' performance against winning teams is equally poor. Since 2005, the Eagles are 13-26 against teams that are .500 or better. That's a .333 winning percentage.

From 2005 to 2007, the Eagles were a .500 ballclub.  It's therefore not tremendously surprising that they couldn't beat the good teams ... since ... they ... weren't ... one.

In close games against winning teams? The Eagles have very little chance.

Starting with Super Bowl XIX in Jacksonville after the 2004 season, the Eagles are an astonishing 6-17 against .500 or better teams in games decided by a touchdown or less.

Propaganda again.  Look, at least I try to be honest with my stats.  This guy just can't help himself, so he tacks on one last game from 2004 to pad his numbers, because 6-16 wouldn't have looked bad enough.

- - - - - - -

Now for the second piece, which comes from Les Bowen's blog post today:

If I'm Lurie or Banner right now, I'm looking at things like the miscalculation involved in the Lorenzo Booker trade, the fullback follies, and the inability of the defensive line to stop the Redskins or the Giants, despite a bunch of money and draft picks spent in that are the past several years. I would have serious questions about whether Reid is the guy to rebuild this team in 2009, the way he was in 1999.

I'm not saying anything is going to happen. Lurie owes Reid at least $10 million on a contract that runs through 2010. The Eagles could still make the playoffs, could still catch a Washington or New York napping in the rematch and succeed just enough to allow management to cling to the mistaken belief that it has the personnel in the trenches to line up with the best.

This is pretty much exactly where I am right now.  Great summary of the "Oh $#@! maybe it is time to get rid of Reid" position, which I know a number of us share. 

I'm less in agreement with this part:

Overall, I still favor the blame-it-on-Donovan scenario: reporters historically sympathetic to management favored by postseason not-for-attribution whispers about how, you know, film review showed the team would actually have won this game or that game if McNabb weren't so erratic, it wasn't really that the team or the playcalling weren't good enough,  you see.

I don't see any way Donovan stays, at age 32, and Andy leaves. That wouldn't really make sense. And though I don't blame McNabb for what has happened this year, I also don't think he's been good enough when it counts. He is not the QB he was four years ago, and I guess now he never will be that guy again.

First of all, I'm all in favor of reporters > turned bloggers making catty comments about other reporters and how they handle their Eagles business.  You can hear the snark dripping off that first sentence. 

The thing is, though, I don't see how this could work out this way.  McNabb has been very, very good this year.  I think the fact that it's Reid taking all this crap for the 5-4 record tells you just how hard it has been to find fault with McNabb's game so far, since McNabb gets blamed for everything around here. 

I think Bowen's also way wrong about McNabb not being the same guy any more.  I guess if he just means he's not the singular offensive weapon he was back when he could run, ok, I can buy that.  But if he instead means there's been a production dropoff, that's crazy.

McNabb's putting up the third-highest passer rating of his career, he's completing over 60 percent of his passes, he's got his Y/A figure up to 7.3, he's finally adjusted his game to avoid taking drive-killing sacks, and he's doing all of this in a year when he's had to play significant chunks of time without:

• Brian Westbrook
• Kevin Curtis
• Reggie Brown
• LJ Smith

Yeah, I bulleted the list because I think people have a tendency to handwave away this reality. 

Given all the guys he's had to work through this year, combined with the complete lack of an effective running game for most of the season, McNabb is quietly compiling one of the finest seasons of his career.

Now I realize it could all change in the blink of a serious [I-word].  But right now, McNabb is the team MVP.  And it's not a close race.