If you look at some of the game stats, the Giants game appears to have been a lot closer than the 33-20 score would suggest. In fact, if you look at detailed team stats for the game, you could get the impression that the Giants actually won the game. Here are a couple of statistical categories in which the Giants dominated the Cowboys:
| Stat | Giants | Cowboys |
|---|---|---|
| Time of Possession | 37:51 | 22:09 |
| Total 1st downs | 25 | 15 |
| Third down conversion | 6-15 | 3-11 |
| Total Yards | 480 | 427 |
| Sacks | 1 | 0 |
In his press conference on Monday, Coach Garret said: "I’m not a big statistical guy. I don’t really care too much about statistics."
So what does this mean? Does our coach think stats are overrated? The numbers from the Giants game sure do suggest that they are. Or is this just a clever ploy by Garret, meant to confuse the press and opponents alike?
Just because you like what Garrett is saying and you like his demeanor while saying it, do not automatically believe everything this head coach tells you. He’d have you believe he went to the Doofus School of Head Coaching, and he’d tell it to you with a straight face and a sincere smile.
At the very least, Garrett is being purposefully vague and non-committal on many of his answers, if not outright misleading. Ever notice that sly smile right after he finishes answering a question? That ought to tell you something right there. When he said "I'm not a big statistical guy", I wouldn't be the least surprised if what he was thinking was "Well, I'm only 6-2, 200 pounds, so I'm not outright lying when I say I'm not a big statistical guy".
Another give-away is that he talks at length about turnovers right after saying he doesn't really care about statistics. Here' the full quote from the press conference:
I’m not a big statistical guy. I don’t really care too much about statistics. But the overwhelming statistic in the NFL last year, last ten, last twenty, last thirty, last fifty is turnovers.
Ultimately, at the end of the day, it’s about who has more points. But the one statistic that factors into who has more points – more than anything else – are turnovers. You have to take care of the football. So we make that emphasis in meetings, in walk-throughs, on the practice field. We keep putting that in front of them.
But that still leaves the question of why the Cowboys won the game when the Giants won many of the statistical battles. Part of the answer are obviously the big plays the Cowboys had throughout the game, none bigger than McCann's goal-line interception for what was arguably a 14 point swing in the game.
The other part is that of the answer is that we are looking at raw volume stats which don't have a high correlation with winning in the NFL. Just last year, Patriots coach Bill Belichick joined a long line of coaches before him (and after him) who have all said that stats are overrated:
"Stats are for losers," he said. "The final score is for winners."
Yesterday, Belichick spoke with the Boston Herald and clarified what exactly he meant with that quote:
"Well," he said, pondering the issue. "I think they all have meaning. It’s just the priority of the stats."
"Wins is number one," Belichick said. "Points is number two, because that correlates to winning. And then you get to the things that correlate to scoring, which (are) red area, big plays, and third down becomes a part of that because of being able to sustain drives and that type of things. But if you make big plays, then third down becomes less important. You can offset any good numbers with bad numbers. You can offset bad numbers with good numbers, but in the end, it’s about getting points on the board and keeping them off. You always want to improve on the things that you’re doing in all areas of the game. You want to run for more yardage, run for more consistent yardage, pass for more consistent yardage, defend it, all those things - get more negative plays, turnovers, on and on. You’re always striving to improve in every one of those areas."
"I'm not saying (stats are) not significant, they are," Belichick clarified, "but the ones that correlate the highest to winning, you still have to consider them as the most important."
What Belichick is basically trying to say is that only the wrong stats are for losers. So let's look at some of the 'right stats' and see how the Cowboys fared versus the Giants:
| Stat | Giants | Cowboys |
|---|---|---|
| TO differential | -2 | +2 |
| Field Position | 318 | 420 |
| ANPY/A | 6.7 | 14.7 |
| DVOA | -56.1% | 61.2% |
| Passer Rating | 88.3 | 124.1 |
Note on the stats: ANPY/A is explained in detail here, DVOA is Football Outsiders proprietary stat and you can read up on it here.
The Cowboys walked away with a victory on the winning stats, or the stats that count, and while many would argue that passer rating should not be included here, at the end of the day, it is correlated to winning.
Traditional, volume-based football stats are pretty useless outside of fantasy football and are usually not a great help in explaining how a game ended. But a good, efficiency-based stat can go a long way towards explaining what happened on the field, and perhaps even why. Only a loser would think otherwise.