FanPost

Stats are for losers. Or are they? Why are the Patriots #1 in the league?

"You know how I feel about stats," Belichick said. "Stats are for losers. The final score is for winners."

This now famous Bill Belichick quote, right after Randy Moss was accused of "mailing it in" after a 20-10 win against the Panthers on December 13th, 2009, has always resonated with me, in particular because I don't believe it to be entirely true.  Well, the stats are for losers part, anyway.  Of COURSE final score is for winners.  Anyway...

I like stats or "metrics" as we call them in the office.  Properly utilized, they can take on the function of a patient monitor in an ICU: tell you a good story with a very quick glance.  The problem with most stats, however, is they are misused incessantly.  It's not the stats' fault; they are innocents, abused by us for personal gain, or to simply support a point we are trying to make.  Selective beasts, we humans are.  Case in point: the overused "yardage" stat when comparing skill players.  We need look no further than the Peyton Manning vs. Tom Brady debate.  As of right now, Manning sits atop the passing yardage race with 4,436 yards to Brady's "measly" 3,701, garnering a pedestrian 6th place on the list.  Clearly Manning's the better QB, right?  Now, I don't want to start another QB debate; I'm simply trying to point out the fallacy of using a stat (a data point, really) like this in any kind of thoughtful comparison.  Without characterization of the data point, we have no idea whether a passing yardage stat of 4,436 is a great accomplishment or not.

Exactly what do I mean by characterization?  Let's say you own a house that costs half a million dollars.  In some parts of the US, that's a comfortably sized colonial, 2,000 sq ft or so.  In other parts, it's a mansion replete with a 3 car garage, tennis courts, and a pool in the shape of Shaq's head.  That's what I mean by characterization.  What are the conditions surrounding the data point.  Take our yardage example above.  How was the yardage obtained?  Was it dome or cold weather?  How many TD's per yard?  How many interceptions?  How many different receivers were a part of that yardage?  Characterize and the data point begins to turn into a true statistic.

Several weeks back, I became curious as to why the New England Patriots were doing so well.  With a cursory glance of data points (ie: team stats at nfl.com), their success didn't make sense.  In fact, the team with the best record in the NFL, a division championship, and the #1 seed in the AFC isn't even in the top 5 of ANY data point in the above link.  Not one.  Clearly, Hoodie was right.  Or was he?

Football Outsiders may not be perfect, but (gratuitous ads aside) they do a pretty good job of characterizing their data, of adding some thought and analysis.  To use a common catch phrase in the business world, they turn data into information.  With their DVOA and DYAR statistics, the data points become a little more meaningful.  For example, is a rushing yard against Pittsburgh the same as a rushing yard against Buffalo?  Not even close.  How about a passing yard in a Chicago snow storm or one in Lucas Oil Stadium?  I think you get the picture.  Football Outsiders makes an attempt at characterization, at rationalizing the data.  I won't bore you with a look at how their statistics are created, but you may find it interesting to dig through the explanations on their website.

So, where am I going with this?  Remember my example from above?  The Patriots, according to nfl.com's data, are not in the top 5 of any category?  Characterize the data and the story changes.  For example, nfl.com has NE 12th in yards per game with 235.1.  Even looking at points per game, which NE leads with 32, doesn't necessarily give you enough to go on.  However, using FO's DVOA analysis, New England ranks #1 with a Passing Offense of 69.9%, 18.2% ahead of the closest team, San Diego.  That's no small margin.  In fact, it's 1.6km ahead of the nearest franchise (get it? 1.6km = 1 mile.  I kill myself).  Data points without analysis tell you nothing.

But even DVOA may mean very little in terms of success.  For example, San Diego had the highest offensive DVOA in 2009 with 29.4%.  DVOA does, however, give you an indication of how strong a team might be, or the possibility of success.  Football Outsiders, at the very least, turns data into information; it can be argued that information isn't necessarily the end all, be all.

Remember when I ranted about characterizing data, that data points without analysis are useless?  Well, I'm going back, sort of, on what I said.  Yardage, by itself, is a weak data point.  Even yards per carry or yards per reception don't necessarily tell you a useful story.  But there are a few data points that, by themselves, are interesting to think about.  Another great site, teamrankings.com, ranks teams using a number of data points.  Over the entire season, NE ranks 2nd in Takeaway Fumble Recovery Percentage with 63.16% (contrasted with 37.93% in 2009) and 1st in the last 3 games with 77.78%.  That means if the opposing team fumbles the ball, the Patriots will recover it 3 out of 4 times based on the last 3 games.  They have the highest Turnover Margin per Game with +1.8, a drastic improvement over 2009's +0.2.  Their Giveaways per Game are the lowest at 0.6 (1.5 in 2009) while their Takeaways per Game are the highest at 2.4 (1.8 in 2009).  If an opposing team fails to take care of the ball, they will pay for it.  Conversely, the Patriots do a very good job of hanging on to the rock.

Stats are nothing if not compared to a "control group" or, in this case, another team.  Let's take the 2009 Super Bowl Champion New Orleans Saints, shall we?  Check out the table below:

Statistic

2010 Patriots

2009 Saints

Takeaway Fumble Recovery %

63.16%, 2nd

62.96%, 1st

Turnover Margin per Game

+1.8, 1st

+0.9, 2nd

Giveaways per Game

0.6, 1st

1.5, 9th

Takeaways per Game

2.4, 1st

2.5, 1st

In every category but one, Giveweays per Game, the 2009 Super Bowl champs were ranked #1 or #2.  This bodes well for the Patriots who are in a similar, if not better, position.  Compare the numbers.  New England is better in every category but Takeaways per Game and that's only by one tenth.

Last, but certainly not least, New England has the highest Average Scoring Margin, +11.6, besting the 2009 Saints with +11.4.

The above data points are certainly not the sole indicator of the success New England has had during the 2010 season.  However, they DO make something very clear to me: Belichick has drilled home the value of fundamentals.  And it's paid off.  Do the basics well and the rest will follow.

What stats (or data points) are you fond of and why?  I'd like to hear.

The views expressed in these FanPosts are not necessarily those of the writers or SBNation.

X
Log In Sign Up

forgot?
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 Pats Pulpit

You must be a member of Pats Pulpit to participate.

We have our own Community Guidelines at Pats Pulpit. You should read them.

Join Pats Pulpit

You must be a member of Pats Pulpit to participate.

We have our own Community Guidelines at Pats Pulpit. You should read them.

Spinner

Authenticating

Great!

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.

tracking_pixel_9341_tracker