FanPost

Luck Be A Lady

Greetings all. I’m back from a long week of camping in stupid hot heat, swimming in a lake, singing medieval-themed songs, and, yes, drinking beer. When I’m camping my preference is Ska (Durango, CA) Modus Hoperandi IPA because it’s delicious, relatively inexpensive, and comes in cans making the logistics easier. Right now I’m drinking a Martin City (Martin City, MO) IPA and contemplating luck.

We often want to put so much into the skill and talent of our favorite players and teams, but, in truth, luck is often more important. A 16-game season is simply not enough of a sample size to smooth the vagaries of luck, and of course this is especially true in the playoffs where the important sample is each single game. To put it simply, sometimes the lesser team gets lucky.

It’s been proven in a number of places that games in the NFL that are decided by less than one score that the quality of the teams is irrelevant. Here is one good look at this phenomenon: http://community.advancednflstats.com/2010/12/do-good-teams-win-close-games-part-2.html. Here’s another using a different metric: http://community.advancednflstats.com/2010/12/is-close-game-clutch-play-story.html. Basically, the winner of close games is determined by small things, like whether ball bounced or not in the Immaculate Reception or whether Dez’s finger touches the back line of the endzone, and these are as much or more luck rather than skill.

Teams that win close games more often usually have winning records, but it’s not necessarily a sign that they are a good team. Good teams dominate games, winning by more than a score and so avoid close losses. Bad teams get dominated, losing by more than a score more often. A team’s record in non-close games is a better example of its true quality than its overall record.

With three 8-8 years in a row we can probably guess that Dallas is a middle of the pack team, but I thought I would take a look and see how lucky we have been and how good we actually are, based on our record in close games. I am going to tabulate the results of the last 11 years, corresponding to Bill Parcells’ first year to the present. Again, I’m defining a close game is one decided by 8 points or fewer either way and a game as a blowout when one team wins by more than one score, hence 9 or more. I’m going to include playoff games, by the way.

Here are the raw numbers (CG is Close Games, CG Pct. Is the percentage of all games that were close, CW is Close Wins, CL is Close Losses):

W

L

Pct.

CG

Pct.

CW

CL

Pct.

2003

10

7

58.8%

6

35.3%

5

1

83.3%

2004

6

10

37.5%

6

37.5%

4

2

66.7%

2005

9

7

56.3%

11

68.8%

6

5

54.5%

2006

9

8

52.9%

6

35.3%

2

4

33.3%

2007

13

4

76.5%

6

35.3%

4

2

66.7%

2008

9

7

56.3%

6

37.5%

3

3

50.0%

2009

12

6

66.7%

8

44.4%

4

4

50.0%

2010

6

10

37.5%

11

68.8%

3

8

27.3%

2011

8

8

50.0%

9

56.3%

4

5

44.4%

2012

8

8

50.0%

12

75.0%

7

5

58.3%

2013

8

8

50.0%

10

62.5%

5

5

50.0%

Total

98

83

54.1%

91

50.3%

47

44

51.6%

Parcells

34

32

51.5%

29

43.9%

17

12

58.6%

Phillips

35

24

59.3%

25

42.4%

11

14

44.0%

Garrett

29

27

51.8%

37

66.1%

19

18

51.4%

This shows a few salient features about various Cowboys teams.

In 2003, the Cowboys were not a good team. They only went 5-6 in blowouts. Basically, they were an average team that caught some breaks and got to the playoffs.

In 2004, they were simply awful, going only 2-8 in blowouts.

In 2005, they were an average team that needed one more break.

In 2006, they were actually a solid team that went 7-3 in blowouts. Not the greatest of teams ever, but better than we realize. Parcells said that he was really disappointed about the Seattle loss because he thought they were in position to beat the Saints and Bears and go to the Super Bowl. He was probably right, too, given that the Bears went 5-1 in close games and 10-3 in blowouts and the Saints went 5-3 in close games and 6-4 in blowouts. The Cowboys were definitely good enough to compete. And the Colts that won that year? 7-1 in blowouts and 9-3 in close ones, including their loss to Dallas.

In 2007, the Boys were, however, darn good. They went 9-2 in blowouts. That’s excellent, but of course one of their close losses was to the Giants in the playoffs. The 16-0 Patriots, by the way, were 5-0 in the regular season and 0-1 in the playoffs in close games, meaning they went 13-0 overall in blowouts. The Giants went 8-1 in close games and 6-5 in blowouts. I said it then, and I’ll say it now, the 2007 Giants lucked into a Super Bowl.

In 2008, it was generally a pretty average team.

In 2009, the Cowboys were a very good team, going 8-2 in non-close games. Yeah, we got destroyed by the Vikings, but the combined 58-14 destruction of the Eagles in Game 16 and the Wildcard round were not mirages. That Vikings team was very good, going 3-2 in close games and 10-3 in blowouts, including the NFC title games to the Saints. The Saints that year were great, going 5-2 in close games and 11-1 in blowouts.

In 2010, the Cowboys were an average team, going 2-3 in blowouts but getting extremely unlucky in close games.

In 2011, that’s about as average as a team can be.

Finally, in 2012 they were a mediocre team at best, going 1-3 in blowouts but getting a break or two to have a chance at the playoffs.

Some other notes. First, we underrated Wade Phillips. Even I did, and I was generally a Phillips partisan. He had a 24-10 record in blowouts. In other words, in the games where one team dominated the other it was the Cowboys about 70% of the time doing the dominating. By comparison Parcells was 17-20 and Garrett has been 10-9 in blowouts. I also will point out that he had the fewest close games by percentage, though it’s not really enough of a sample to prove anything compared to Parcells.

Take a quick look at the 2010 season, by the way. How unlucky was Phillips that year anyway? Phillips lost all 5 close games his team had that year. Garrett went 3-3 in his. Couldn’t catch a break in close games and lost his starting QB 6 games in. And then his job. How many black cats did it take to do that?

Second, Garrett really likes close games. Essentially 2 out of every 3 games Garrett has coached has been one. He’s gotten about as average a result in all of those games as he could, so the Cowboys have neither been lucky nor unlucky in those games, but that’s irrelevant to the simple fact that his teams have been vastly more likely to play in close games in the first place.

This means either one of two things. Either Garrett is unaware that luck is the prime determinant of close games or that he knows this team was bad and playing to create opportunities where the other teams’ skill and talent was nullified. Neither of these are good things, though it’s much preferable to me that the teams were bad and he was maximizing their opportunities because that would be a better process.

We will know quite a bit more about Garrett after this year, too. As we look at the talent base that he had when he got here to now, it seems clear he’s improved the depth and resiliency of this roster. Hopefully, this will mean fewer close games and more big wins. It will tell us quite a bit, though, if Garrett continues to push for close games, and none of it good.

Overall, we can say that Parcells had a bit of luck on his side, Phillips had a bit of unluck on his, and Garrett has been about average. That earns Parcells a Breckenridge (Breckenridge, CO) Lucky U IPA, Phillips a Lucky Bucket (La Vista, NE) Certified Evil Belgian Strong Dark Ale, and Garrett a Jackie O’s (Athens, GA) Middle of Nowhere Stout.

Another user-created commentary provided by a BTB reader.