Finding the Winning Factors - Turnovers

[UPDATE] We've had a lot of post for today. So, I placed most of the cool stuff in this post after the jump so we can conserve space on the frontpage. Be sure to rec this. It's on of mgrex03's best. --bbs

This week in the Finding the Winning Factors series, let's take a look at a very common stat:  Turnovers. This stat is talked about over and over when talking about winning, but are the "experts" correct when they say turnovers (or lack thereof) are crucial to winning? Here's the obvious calculation:

Mind blowing mathematics there.  I actually looked at a few rate stats with respect to Turnovers, such as Turnovers / Yard Gained, or Turnovers / Drive (which I included in the tables), but none of them were better predictors than just raw Turnovers, so we'll go with that.  Occam's Razor wins again.

I went back to 2001 to see how teams do.  That means I'll be looking at 4,080 games over the past 8 years.  I did this calculation for every game, then found the overall average for a game.  I then went back and compared every game to this average, and found wins/losses based on being above or below average.

Let's get to the numbers...

• The Overall Average Turnovers / Game over the past 8 years is 1.76.  That means causing just 2 Turnovers in a game makes your defense Above Average.  I was expecting it to be just slightly over 2, so I guess offensives have been doing a better job protecting the football than expected. 
  
• The best team since 2001?  The Jaguars, with an average of 1.406 Turnovers/Game, just barely edging out the 2nd best team, the Colts, who have an average of 1.414 Turnovers/Game.  It's actually just 1 less turnover over the past 8 years (180 to 181).  After Week 1, I hope the Colts have passed them up.
  
• Defensively, the Bears and Ravens are tied at #1, with an average of 2.01 Turnovers/Game.  They've each caused 257 Turnovers over the past 8 years.  That's pretty good.  How about the Colts?  8th best at 1.86 Turnovers/Game.  That's better than the Steelers, Titans, and Eagles.  This is one of the best categories for the Colts' defense, and one I expected them to be high up on.
  
• So how often do teams win when they are Above Average on Offense (Having 0 or 1 TO in a game)?  1370-651-1, which is 67.8%.  That's the 2nd best we've seen so far.  Pretty damn good.  Having an Above Average defense (causing 2+ TOs in a game) produces a record of 1387-668-3, which is 67.5%, which is a slightly worse than the offense, but still excellent. Turnovers really do lead to Wins and Losses.
  
• When both the Offense and Defense are Above Average, their record is 912-193-1, or 82.5%.  This ranks 3rd out of 12 stats so far, only behind ANY/A and Drive Success Rate.  Not turning the ball over, and causing Turnovers, is, like many "experts" say, key to winning games.
• How about just winning the Turnover battle?  Teams that have less TOs than their opponent are 1307-335-2, or 79.6%.  So 4 times out of 5 if you win the Turnover battle, you win the game.  Not bad.  The Colts are 54-6 when they win the Turnover battle.
• There is no sense going over the best game since 2001, as many teams have had 0 TOs in a game.  But I can certainly look at the worst.  Week 7 in 2001, the Rams turned the ball over 8 times against the Saints, and lost 34-31.  Eight Turnovers, and they still scored 31 points?  That Rams offense was pretty good.  The worst Colts games was the 6 INT game vs. the Chargers in 2007, but I bet everyone here knew that.
• The Colts have fumbled the ball the 2nd fewest times (73), and caused the 3rd most fumbles (109).  Coach Dungy preached it, and the players lived up to it.  Hopefully those things don't change.
  
• In terms of Net Turnovers over the past 8 years, the two teams tied at the top, with 57 more caused turnovers, are the Colts and Patriots.  In other news, water is wet.  As we've seen with the other stats that lead to lots of wins, these two teams are always at the top.

So what did we learn today?  The "experts" are right when they say that teams who limit turnovers, and cause turnovers, win games.  Just remember, a broken clock is still right twice a day.

Unfortunately, this will have to be the last stat looked at for this offseason.  Life has gotten in the way (an unexpected move), so I won't have time to write up the last 3 stats in the table below.  I have run the numbers though, and none of them are that great.  I've included the summary numbers so that you can take a look at how all of them stack up.  Here goes:

I'm going to write up a summary article sometime in the next 2 weeks or so, where I'll have where the Colts rank in each one of these categories (Real short summary: Offense Good, Defense Not so Good), and I'll talk about how I'm going to use this to predict games.

After the jump you'll see the offensive and defensive numbers for each team, as well as Win/Loss records when going above/below the league average.  Click on the headers to sort.  Here's a link to the last 5 years worth of data.  There's not much of a difference, especially since it only took out 3 years worth of data.

Here are the Offensive/Defensive Numbers...

Here are the Win/Loss records of teams when they go above/below average:

And finally, Win/Loss records when the offense is above average, and the defense holds the other team below average: