We all know that pre-season performance isn't any indicator of regular season performance. With the way the pre-season has gone for the Cowboys so far, we've heard all about the 2008 Lions, the Colts of the last five years, the 49ers of the nineties and a host of other teams with pre-season results that did not match their regular season results.
However, at the very least, the performance in the games so far provides an indication of what the teams need to work on over the coming weeks until the regular season kicks off. In part, the pre-season is what it is because teams are knocking off the rust on their starters, playing mostly vanilla schemes, and using the games to test out the 2nd and 3rd stringers. So looking at W-L columns and general stats is largely meaningless.
BUT: Teams do play their 1st teams for some parts of the pre-season. Over the course of the pre-season games so far, enough data has accumulated to make looking at how the first teams performed worth it, and may provide some interesting food for thought. After the break, I'll look at the performance of only the starters/first teams over the pre-season so far for the four NFC East teams.
Disclaimer: In the NFL, past performance is not necessarily indicative of future results. This is particularly true as we look at pre-season results, but is equally true for most other stats, historical trends and any other observed data points. The stats we use to look at the NFL are suited to explain and perhaps better understand past performance to varying degrees, but for the most part, they are completely unsuited to predict future results. Ok, with that out of the way, let's focus on some pre-season stats.
Quarterbacks: Let's start with the QBs. Statistically, the starting QBs are the easiest to assess. All starting QBs in the NFC East are undisputed this year, and their performance is easily tracked:
STARTING QB pre-season passer rating, NFC East, 2010
| 2010 pre-season
||RATING||Pre- season||Regular Season|
In terms of passer rating, none of the NFC East QBs have delivered a particularly noteworthy performance this pre-season. In fact, it would be fair to say that all four have struggled so far. Shockingly, all four NFC East QBs have a lower passer rating than Tim Tebow (77.6). Maybe not quite as shocking, but at least a little disconcerting is to see who's at the top of the QB leaderboard this pre-season: Aaron Rodgers has racked up a 141.2 rating on 53 pass attempts, Tom Brady is a close second with a 140.6 rating on 42 pass attempts.
But if you look at's numbers from last season, perhaps there is hope yet: Eli had arguably his best season as an NFL quarterback last year despite coming in with a 66.3 passer rating in the pre-season. The other three had numbers with a lot smaller swing between the pre-season and the regular season (*Kolb's 2009 data is from only one pre-season game and a little over two regular season games).
First team performance
Just like we can look at the starting QB numbers, we can also isolate the team performance on drives when most or all of the first-team units were on the field. To do this, I went through the play-by-plays for all the games and tried to figure out which drives still had the first units on the field and which didn't. I believe I got the numbers right, but it's not always perfectly clear from the play-by-plays which unit, particularly on defense, was on the field.
I looked at number of drives and points scored or points allowed during those drives. With that data I calculated what I'll call Equivalent Points per Game: The average number of drives per game is about 12 in the NFL. Multiplying the points per drive by 12 gives an indication of how many points the teams would have theoretically scored or allowed with their pre-season performance over four full quarters in one game.
First team DEFENSE pre-season performances by drive, NFC East
|Points allowed per drive
||Equivalent Points allowed per Game
I'll be the first to point out that projecting individual drives to a full game (like you would with baseball's ERA for example) is tricky at best. There are many factors that can and do change the game situation from play to play, particularly in the pre-season. Yet the numbers do provide a glimpse of how the first team defenses have performed as a unit. And if nothing else, the numbers are an indication that the Cowboys and Giants still have work to do, and that the Redskins and Eagles would do well not to grow complacent with their pre-season performance on defense.
Obviously, you could always argue that the pre-season opponents also played a big role in these numbers. Were they playing vanilla schemes, how much did they want a pre-season win, etc. - we've been through all these arguments already - but at the end of the day, as the Cowboys defensive coordinator, you can't really be happy about these numbers.
Now the same exercise for the Offense.
First team OFFENSE pre-season performances by drive, NFC East
|Points scored per drive
||Equivalent Points scored per Game
This table does not look like it contains any of the teams formerly known as the offensive powerhouses of the NFC East. Right now it looks more like a Confederacy of Dunces division. The offensive showing of the four East teams is just as awful as that of their quarterbacks this year.
All four teams clearly have a lot of homework left to do, and I'm sure that every single team is taking a long, hard look at what they each need to do to fix their offenses. But rest assured, not a single one of the NFC opponents for the first couple of weeks is leaning back back right now and thinking "The Cowboys/Giants/Eagles/Redskins? Don't worry about them. Piece of cake."
Our opponents know that August numbers don't mean anything come week one. The August stats are completely useless in predicting how the season will unfold, particularly since they do not reflect one cold, hard truth: The most successful teams at the end of the season will be those that have found ways to continuously improve - and then improve some more - as the season progresses. Much to my dismay, I don't have any stats that can predict that.