1. You know how the band was called 10,000 Maniacs but actually it was just Natalie Merchant and four, sometimes five, other people? So, at most, six maniacs? Well, this headline is kind of like that, in that I don’t have literally 10,000 reasons the Cowboys are better than you think. That’s the bad news. But the good news is that this will be actual reasons the Cowboys are better than you think, whereas I’m not sure the Natalie Merchant and her bandmates were clinically diagnosed with anything.
When I sat down to write about the Dallas Cowboys this week—something no one on the internet has ever done as far as I can tell—I expected to be laying out an argument for why, due to their unsustainable takeaway rate, they’d be coming back down and nestling in with the second-tier contenders in the NFC. But a closer look at the numbers revealed a different story. But before we get to the Cowboys, some words about numbers. And if you’re not a numbers person, you might want to order and wait for the arrival of this series of books before you read the rest of this column:
I’m a strong believer in the volatility of what I call the “small-sample-size stats.” In the NFL, we often lose track of the fact that the talent is quite evenly dispersed—there’s not a lot of separation between, say, the fifth- and 25th-best teams in the league. Most of the time, the difference is a play or two each week, and those plays are usually turnovers, come in the red zone, or come on fourth down.
But there are so few of those plays over the course of a season—turnovers, total series in the red zone, and fourth-down attempts—and we often fail to recognize that, when dealing with such a small sample size, it’s difficult to develop any kind of projection or draw broad conclusions about how teams will perform in these categories. There are some exceptions: If your quarterback is Jameis Winston you’re at risk of turning it more than the average team, if you have an elite four-man pass rush you will typically get more takeaways than an average defense, and if you’re a good overall team you typically perform better in fourth-down efficiency because your opponent is often forced to go for it on fourth-and-long to stay in the game while you can pick and choose your spots. Meanwhile, red-zone stats are particularly fickle not only because of the sample size but because the geometry of the game changes so drastically when the field is shortened (though teams with great running games typically fare better).
There’s also the ultimate “luck” stat: fumble recovery rate. Recovering a fumble requires all the skill of winning a coin toss; it’s pure luck and every team’s expected recovery rate is 50%, no matter how many Cardinals fans DM me to say Arizona’s 80% recovery rate through seven games (on 25 fumbles—they were picking up an extra possession per game!) was a result of them “wanting it more.” Though that apparently wasn’t the case on Thursday night when they recovered only one of the game’s four fumbles against the Packers.
Ergo, I keep a close eye on the following stats, which often show which teams are outperforming or underperforming against their record. Yes, it's great if your team is having success in these categories, but that success is difficult to maintain. You can expect them to regress toward the league average as the season goes on:
• Turnover differential: How many possessions a team is gaining/losing over the course of the season
• Fumble recoveries: Not just the percentage but, based on how many fumbles, the number recovered over expected (example: A team that has a 100% recovery rate on four fumbles hasn’t picked up as much of an advantage as a team with an 80% recovery rate over 25 fumbles)
• Fourth-down conversions made/allowed: A failed fourth down is a turnover, even if it doesn’t factor into turnover differential. Like with fumble recoveries, this stat is not just percentage, but possessions gained/lost when factoring how many fourth-down conversions tried or defended.
• Red Zone efficiency: Based on the current league average—5.0 points per red-zone possession—how many points above expected is a team scoring over all red-zone trips? How many points above expected is their defense allowing in the red zone?
Now, back to the Cowboys. Their turnover differential will almost surely regress (they’re fourth in the league at the moment), in part because their fumble luck will as well (fifth). But the number that really stands out is their red-zone offensive efficiency, which is far lower than expected. With one of the NFL’s best rushing attacks and an elite quarterback, you could identify the Cowboys as an offense that should perform above league-average in the red zone. But, through six games, they’re sitting at 27th in the NFL in red-zone points above expected. If they were league-average in the red zone, they’d have scored 13 more points this season. (And if they matched the red-zone efficiency of, say, Arizona, they’d have 28 more points.) All that is to say: While the Cowboys will probably take the ball away less frequently as the season goes on, they’ll likely start scoring more touchdowns in the red zone, canceling that out.
But the reason to be bullish on the Cowboys is more than that. Right now, they’re winning primarily through run game and defense, which is really how everyone in the league would like to win (get a lead, ram it down your opponent’s throat, go home and enjoy some celebratory kombucha). But while some teams are less equipped than others to come back if they fall off-schedule, Dak Prescott gives them an opportunity to hang around in any game and steal it back late. He’s had to do it twice this season, and he succeeded both times: re-taking a lead in Tampa on opening night (even if Tom Brady took it right back), and getting an overtime touchdown in Foxboro in their last game, two weeks ago.
In short: Dallas can win in a slog, and they can win in a shootout, and the more paths you have to winning on a weekly basis, the better your chances of winning a Super Bowl.
1b. Did you really go to a 10,000 Maniacs show expecting to see 10,000 band members on stage? Let's be reasonable, shall we?
