Regression to the mean

This is truly one of my favorite terms and devices of wishful thinking to predict how things might play out later in the season. Does it have any real scientific merit in a 16-game season? Probably not. But it might be cause for optimism, so let's check it out.

It's a mixed bag with the Texans so far, as they've looked flat out dominant in some aspects of the game and league-worst in others. However, I can pretty much guarantee that there will be some major statistical shifts by the end of the season.

The term "regression to the mean" has been used around here recently to describe why Shaun Cody won't be as effective this year after posting a 100% stuff rate last year. It lead to questions about our run defense as a whole going forward. Would they be able to sustain the success they had during the last twelve games of last season? On a more positive note, it might suggest that the Texans will recover more of their own fumbles on offense this year (they only recovered 3/13 last year). It's a double-edged sword, really.


There are many factors that play into these statistics, but one thing is for sure: four games is not a good sample size. I'd argue that 16 isn't either, but there's little we can do about that. So over the next twelve weeks, what can we expect to improve, and what probably won't hold up? Here are a few statistics after four games, starting with the offense:


  • The Texans are the fourth-ranked scoring offense, putting up 27 points per game.
  • They're second in yards, with 415 per game.
  • A large portion of that comes from rushing yards, where they rank tops in the league with 172 per game, which is what the Jets finished with last year to lead the league in rushing. The difference? They did it with a higher volume, averaging 4.5 yards per carry. The Texans are averaging a full yard more per carry at 5.5.
  • The passing offense is on a much more modest pace, averaging 244 yards per game (following up last year's 291), but on a solid 8 yards per attempt, which is nearly identical to last year.
  • As a team, they've turned the ball over five times and taken eleven sacks (on pace for 44. Last year, they took just 25).
  • They're also fourth in 3rd-down conversions, converting them nearly 47% of the time.


Aside from the sacks, it all looks pretty nice. Defensively, not so much. Here are a few:


  • They're dead last in yards allowed, giving up 408 per game.
  • They're 26th in points allowed, at a clip of 25.5 per game.
  • They're last in passing yards allowed by a wide margin, surrendering 338 per game. Over a full season, that projects to 5,408, which is 324 more than Dan Marino's single-season passing record. The volume of passes against them hasn't helped matters, but they're next-to-last in YPA allowed, at 8.7.
  • They're 2nd-best in rushing yards allowed, giving up just 70 per game on the ground at a clip of 3.6 YPA (good for 7th in the league).
  • They've forced and recovered two fumbles and notched two interceptions, alongside 9 sacks (tied for ninth in the league).
  • They're 20th in 3rd-down defense, allowing a 1st-down 39% of the time.

What does it all mean? Well, nothing substantial, technically. The schedule is different, there has been some turnover on the roster and coaching staff, and teams go worst-to-first and vice versa fairly often in this league. But that's no good for our purposes, so let's make some fairly baseless predictions:


Things that will improve:

1) The passing defense, primarily in terms of yards. Last year's worst passing defense (Detroit) surrendered 266 yards per game, a full 70 less than we're giving up now. In 2008, Seattle was the worst, at 259 yards per game. It tends to get lower the further back you go. Could the continuing trend towards passing offenses increase that number? Sure, but not by 70 yards. Consider also the competition we've faced: Peyton Manning, Donovan McNabb, and Tony Romo. There haven't been many more productive quarterbacks over the past several years. Say what you will about Gradkowski, but even a relatively good game by him doesn't make me afraid of the prospect of Sanchez, Vick/Kolb, Flacco, Garrard, or Young.

2) Turnovers forced. Three teams grabbed less than ten interceptions last year. All were bottom feeders who spent most of the season playing from behind and getting the ball rammed down their throats by running backs. The Texans' offense is good enough to keep teams throwing, and the defensive line is good enough to force bad throws. If the DBs can catch, they'll finish the season with 14+.

3) Points per game allowed. Nobody is happy with the state of this defense right now, but I don't think anybody buys that they're one of the worst in the league. The run defense has been good, and they're gotten a lot of pressure on opposing QBs. They'll still give up yards through the air, but when Cushing comes back, expect more stuffs, three-and-outs, and turnovers. They'll ride that to mediocrity, and all will be well in the world.


Things that probably won't hold up:

1) The rushing offense. Sorry to say it, but Arian Foster won't rush for 2,148 yards as he's on pace to do, nor will he average 6.4 YPC. On the schedule still: NYJ, Baltimore, Tennessee twice. If you believe the KC hype, lump them in there. It's not going to get any easier for our boys up front. That's not to say they won't be productive over the rest of the season. However, we're not going to win many more imbalanced games. If this team wins the division, it'll be because Schaub and the passing offense stepped up and won them some big games down the stretch.

2. Points scored per game. Not in a major way, however, the Texans are second in yards and fourth in points. Nothing I've seen so far this season leads me to believe that they're solved their red-zone issues. I'd expect another 5-6 place discrepancy between yards and points.


Things I'm just not sure about:

1) Sacks on Schaub. As noted earlier, Schaub is on pace to take 44 sacks this season. To put that into perspective, it's three more than David Carr took in 2006. It's a lot, but not surprising given how the pass protection has looked. It can't be pinned on any one player, either. It's a combination of every player on the offensive line making a mistake here and there. However, some of those could have been prevented by Schaub and/or better play-calling. Has the change in offensive coordinator put the passing game on the back-burner? Who knows. But pass protection needs to improve for the well-being of this team however, and it brings me to my next point:

2) Schaub's TD:INT ratio. It's early. 7:4 isn't awful at this point in the season (unless you consider that Manning is sitting at 11:1, and Sanchez at 8:0). However, at the end of the season, it needs to be better than 2:1. With the red-zone pass protection (or lack thereof) we have, will it be? I always made the claim that Schaub would become a more efficient QB when he actually had a running game to lean on. I'd like very much for him to prove me right. Oh, and speaking of Sanchez, that aught to be a fun regression to the mean. He's due for some stinkers.

3) RB ball security. One fumble in four games for the RBs. I like that very much, and it's certainly sustainable. However, last year's debacle has me worried that we're going to have a couple of rough weeks in that area that cost us two very winnable games.

4) Sacks on defense. With Mario, Smith, Okoye, and Mitchell, there's no reason this team shouldn't be harassing opposing QBs every week. The only thing that gives me second thoughts is that they never have in the past. Even with one of the league's premier pass-rushers, they've been in the bottom half of the league in sacks every single year since 2006 (and every year going further back, I assume, but that's ancient history). Even if Mario and Antonio Smith each notch 12+, it won't be enough. We need contributions from the DTs and LBs. Pick it up, Frank Bush. DeMeco and Cushing should have three or four each.


And ultimately, the most important stat that all of this leads to: 3-1. On pace for 12-4. If you ask me, that's division-title worthy. Based on everything else that could improve/go wrong over the next twelve weeks, how do you think it will translate to wins and losses?

Log In Sign Up

Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

Join Battle Red Blog

You must be a member of Battle Red Blog to participate.

We have our own Community Guidelines at Battle Red Blog. You should read them.

Join Battle Red Blog

You must be a member of Battle Red Blog to participate.

We have our own Community Guidelines at Battle Red Blog. You should read them.




Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.