The 2004 Season: A Crunching the Numbers Retrospective

2004. It was the highlight of Andy Reid's tenure here as the head coach of the Philadelphia Eagles and was probably a memorable time in the football lives of many readers here (except for the Super Bowl, of course). I unfortunately started following football two years too late and have no idea what it feels like for my favorite football team to win a conference championship game. That does not, however, prevent me from conducting research for fun since I my final exams are over. (NOTE: This is a lengthy post. Just giving a heads up to the "too long; didn't read" crowd.)

Below you will find the results of my algorithm on the NFL as it was in 2004. I'm sure most of the readers have fond memories of that year, but if you're like me I'll give a brief overview of what the playoffs looked like back then. I'm sure it will be refreshing to see the double digits on the other side of the dash on the Eagles' record.

Playoff Seeding, 2004

Seed	AFC	NFC
1	Pittsburgh (15-1)	Philadelphia (13-3)
2	New England (14-2)	Atlanta (11-5)
3	Indianapolis (12-4)	Green Bay (10-6)
4	San Diego (12-4)	Seattle (9-7)
5	New York Jets (10-6)	St. Louis (8-8)
6	Denver (10-6)	Minnesota (8-8)

So there you have it. Seattle, San Diego, Denver, and Green Bay took an early exit. St. Louis, Minnesota, Atlanta, New York, Pittsburgh, and Indianapolis each won one game. And Philadelphia and New England each one two games to reach the Super Bowl, where the Patriots won 24-21.

In fairness, I'm already planning on improving my algorithm for next season. I debated tossing in the improvements for this post but decided against it. Keeping the same formula that I've been using all season will help with comparisons and consistency. The results were... interesting, to say the least. I'll break it down in a little bit, but first here's how all 32 teams fared in the rankings:

Rank	Team	Score
1	Indianapolis	26.198
2	New York Jets	24.500
3	New England	24.347
4	San Diego	23.134
5	Philadelphia	18.665
6	Pittsburgh	17.473
7	Buffalo	16.359
8	Baltimore	15.097
9	Carolina	7.917
10	Jacksonville	5.134
11	Seattle	-0.526
12	Atlanta	-0.796
13	Washington	-1.950
14	Cincinnati	-3.551
15	Houston	-3.818
16	Detroit	-4.856
17	New York Giants	-5.010
18	Arizona	-5.676
19	Minnesota	-7.070
20	New Orleans	-7.274
21	Denver	-8.368
22	Tampa Bay	-11.821
23	Kansas City	-13.994
24	Green Bay	-20.279
25	Tennessee	-20.433
26	Chicago	-21.362
27	Cleveland	-31.014
28	Miami	-32.330
29	Dallas	-34.368
30	Oakland	-38.546
31	St. Louis	-43.140
32	San Francisco	-47.622

First, a comparison of the playoff picture. Unlike my other posts, where I seed each team based on their ranking, I'll do the opposite. I'll look at the playoff seeding as it happened and compare each team to their rank in my algorithm.

Now onto the original question. How good were the 2004 Eagles? This discussion could have taken two paths depending on how Philadelphia performed in my system. If they had done poorly, like St. Louis, I would be looking into exactly what differences led to the discrepancy. But there was a nice harmony between my rankings and the actual outcome; according to the formula the Eagles were the best team in the NFC that year, and they ended up representing the conference in the Super Bowl. Unfortunately that does not close this issue, because we still have St. Louis, an 8-8 team that was second-worst in my rankings but still won a playoff game on the road. Therefore, it must be decided how flawed the system actually is. If it is deemed to be inaccurate, then the Eagles' ranking as the best team in the conference is inaccurate, at least from a correlation standpoint. Let's dive in and see how the rankings compared with reality in the end.

St. Louis is an interesting case because they are very similar to the 2012 Colts. Both teams are near the basement in my system but both are wild card teams. The 2004 Rams are much more extreme in this inconsistency, because they could not do anything right statistically. The were outscored by an average of five points per game. They allowed over three sacks per game. They gave up almost five yards per carry on defense. They averaged one fumble lost per game and committed twenty-four more turnovers than they forced. Twenty-four! How they managed to win eight games is beyond me, especially when you consider how similar those statistics are to the 2012 Eagles. And yet, the 2004 Rams not only made the playoffs, but they beat the Seahawks on the road. Is that a fluke? Or is there something fundamentally wrong with my system? I might be a little biased since this is my algorithm, but I'm inclined to think that this is a fluke.

Let's look at the other ten playoff games. Based solely on the rankings above, my algorithm would have predicted the winner in eight of them. In the case where the rankings were wrong (Pittsburgh over New York, New England over Indianapolis) the higher-ranked team was playing on the road in a hostile environment. Only St. Louis managed to win a road game as the worse team. And then they were beaten badly in the next game against the Falcons where they were outscored by thirty points. Likewise, the low-ranked Broncos (21) were trounced by the number one team, Indianapolis, by twenty-five points. The middle-of-the-pack Vikings (19) also defeated a lower-ranked team in the Packers (24), despite playing in Lambeau.

All in all, if you took someone who knew nothing about the 2004 and asked them to blindly pick the games each week in the playoffs, they would be expected to guess half the games correctly. I'll give them the benefit of the doubt and say that they guessed six winners correctly out of eleven games which is a success rate of 54.5%. If they made picks based upon the rankings above, they would have correctly picked two additional games, with a success rate of 72.3% (a 17.8% improvement). Had they picked all of the home teams, they still would have only correctly picked six games, discounting the Super Bowl. So, even though St. Louis was able to pull one out (I can only imagine how that game went), the rankings for the most part were accurate.

Accepting that these rankings are overall accurate for the 2004 season makes the Eagles the best team in the NFC for that year. But does that still make them a great team? After all, the NFC was terrible that year - only the four division winners finished with a record above .500. In the rankings, the Panthers (who went 7-9 that year) were the only other team in the conference to finish in the top ten in my rankings. Finding it hard to believe that Buffalo and Jacksonville would finish in the top ten, I looked at the records in the AFC for 2004 (both the Jaguars and Bills went 9-7). In addition to the six playoff teams in the AFC, four other teams managed at least eight wins. The Eagles most definitely had weaker competition, and they showed it by going 11-1 in the conference and 13-3 overall. Then again, the Broncos - who were less-than-stellar in my rankings - managed ten wins in the AFC. The Eagles probably would have been successful even against better opponents. Whether or not they still would have won thirteen games is debatable.

What do the numbers say? To put it bluntly, the Eagles were average or slightly above average in almost every category in 2004. Whether it was sacks allowed, yards per rush, points forced, turnover ratio... it was all middle-of-the-pack. However, they did stand out in several areas. That was McNabb's best year as a starting quarterback, and it showed in his statistics. His yards-per-attempt were decent (7.2) and his interception percentage (2.12) was excellent, although we already knew that part about Donovan. They really shined on defense, where they were second in the league in sacks (2.9 per game) and first in scoring (16.2 points per game).

So, where does that leave us? How good were the 2004 Eagles? What does all of this say about Andy Reid? In my opinion, the Eagles were an all-around good team, but their defense made them a great team. It was not a team that will be forever remembered like the 1998 Vikings or the 2007 Patriots as one of those groups who "just didn't win the big game." But they were still great in the same way that the 2006 Bears were great (without the hilariousness of Rex Grossman). Sure, they may have had weak competition, and in all fairness they would have likely lost the Super Bowl no matter who they played. The AFC powerhouses that year were just too good. And as far as the idea that Andy was only really successful with Jim Johnson running the defense, that is true. But it shouldn't take away from anything that Reid did. After all, he made the decision to hire Johnson and successfully crafted an offense that complemented Johnson's blitz-heavy defense. And there have been good defenses out there without an offense to keep them successful - just look at the Browns in recent years.

That, to me, is the legacy of the 2004 Eagles. They carried all the hallmarks of a great team, such as determination and perseverance (especially when McNabb got injured). They had great, emotional leaders on the field like Brian Dawkins. Statistically, they had a stifling defense with an efficient offense to match. The 2004 Eagles had earned the right to play in the Super Bowl that year, but they were outmatched by their competition.

But that's just me, some college student looking at numbers on a computer. You guys were there. You guys lived it. What do you think? Am I on the right track, or am I just full of it?