You're The Last In Line - The Myth of the "Tough" Schedule

"We'll know for the first time, if we're evil or divine." 

--Ronnie James Dio (Heavy Metal Poet & NFL Schedule Soothsayer)

This week I was going to preview the Broncos 2010 schedule and give you a game-by-game breakdown of the brutality of the first six games:

9/12 -- at Jacksonville 1:00 pm
9/19  -- Seattle 4:05 pm
9/26 -- Indianapolis 4:15 pm
10/3 -- at Tennessee 1:00 pm
10/10 -- at Baltimore 1:00 pm
10/17 --  NY Jets 4:05 pm

At first glance, none of these games (outside of Seattle) are games that you would circle and say, "yeah, Denver's got that game."    In fact, I was ready to lament facing Peyton Manning and the Colts.  I was ready to cry at having to go on the road to face Tennessee and Baltimore in back-to-back-early road games.  I was ready to hang my head at the prospect of facing the mighty Gang Green.  

But then I came to my senses and checked my biases at the door--except that ones against Raiders fans and drunk street rioters, which I should combine into one bias, because they're always the same sets of people.

For several weeks now, I've been parroting the false expectation that the early part of Denver's schedule is somehow going to be tougher than the latter half.  But then I realized, it's almost impossible to know in the preseason if a schedule is either hard or easy based on last year's results.  

Let's face it.  There are no easy games on an NFL schedule. After years of watching the Broncos-Raiders, Broncos-Chiefs, and the Broncos facing anyone with a losing record in December, my intuition should have told me better.    I guess watching Jabar Gaffney shred Kansas City's 3rd-string corner backs in last season's finale blinded me to the fact that superior teams often lose to teams with Matt Cassel.

You've heard the myth of the tough schedule a lot in the media as well.  The good folks over at NFL.com described the Broncos 2010 schedule as follows:

Difficult Early Schedule Will Test Broncos

The Broncos play host to the Colts in Week 3 only to follow that up with road games at Tennessee and Baltimore before returning home to face the Jets. Denver got off to a hot start in 2009, racing out to a 6-0 mark. The Broncos will be hard-pressed to match that accomplishment with such a tough slate against three playoff teams and one of the hottest teams to close out last season.

It's easy to fall victim to the myth of the tough schedule because the last images we remember from the previous year are of a handful of teams enjoying success.  Peyton Manning shuffles his feet really really fast, Vince Young is again somehow, suddenly-last-summer going to really revolutionize the NFL, Ray Lewis really does eat children for breakfast, and Rex Ryan is really really mean-looking.  

But that was last year.   And predicting which team's are going to be difficult based on what they did last year is about as smart as trying to predict the stock market or JaMarcus Russell's pre-camp weight.  You just can't do it.   Finishing first in line one year doesn't mean you get shortcuts to the front of the line a year later.  

So, please, for the love of Ronnie James Dio, let's start worrying about the important things in life.  Like Tim Tebow's jersey sales.

Winning Percentage & Correlation - Killing the Dragon

There simply isn't any strong correlation between what a team did last year in the NFL to what it's going to do this year.  When I looked at the correlation between winning percentage from a team's current year to it's previous year all the way back to 2000, the correlation coefficient (with p-values less than .005) was a rather small .293.  In the world of correlation, .293 is weak at best.  

Is this lack of correlation actually an indication that the strength of schedule concept is actually working?  Perhaps. Given that, in theory, teams that finish with poor records are playing a supposedly easier schedule, so one might expect to see these teams winning a lot more in subsequent years.  I did find that, since 2000, those teams that had .500 or worse records did have, on average, a 10% higher winning percentage than they did the year before. That's 1 extra win per year, on average.  But it's certainly nothing you want to write home to momma (or Al Davis) about.   If you win 4 games in one year and 5 the next, your coach is still getting the devil's horns. 

While it's never fun to play around with stats (you might put your eye out and serve as the next Raiders mascot), this data would suggest on the surface that even if Rex Ryan made the playoffs this year, he might be sitting on the couch the next.  And further, the year after he sat on the couch, he might win a game or two more (who wouldn't with Darrelle Revis?), but wouldn't necessarily be fist pumping his way to a division title.  

The Dude's Pythagorean Theorem, which is simply an adjusted version of the Football Outsider's Pythagorean Wins Formula, enjoys a much better correlation with respect to predicting winning percentage.  In the same time period 2000-2009, the Dude's Theorem from a team's previous year had a correlation of .608 to a team's current winning percentage.   And according to The Dude's Theorem, the Titans, Jaguars, Raiders, Colts, and Chargers all had their fair share of "luck" last year.  

But Why am I not Smarter than John Clayton?

Since preseason predictions are notoriously difficult, so I'm not going to pile on John Clayton or the NFL-Network crew for getting the Broncos record wrong last year.  A lot of people get the predictions wrong, whether they reside in Bristol, Connecticut or within the confines of the MHR universe.  But at least we have the decency to wear contacts, Mr. Clayton.

More interesting to me is why these preseason projections are so difficult.    Let's explore some of the more (and less) obvious reasons:

1. Free Agency.  Since 1992, free-agency has obviously driven the league to parity.  A team can, with the signing of 2 or 3 players in a given free-agency period, go from the outhouse to the penthouse very quickly. We've seen numerous instances of this, but perhaps the Green Bay Packers of 1992 and 1993 come to mind with the signings of Brett Favre and Reggie White.  

2. Coaching turnover.  Generally, coaches have a 3-4 year window in which to hire their guys, implement their systems, and benefit (if they have input into free agency and the draft) from their work.  Given this 48-month window, and the fact that only 12 teams each year make the playoffs, you're left with a league that simply grinds coaches to meat.  And predicting what will happen with new and retread coaches is not something that one ought to make a living doing.  You might as well take a job as a security guard at the Oakland Colosseum.  It's likely more stable work.  

3. Injuries.  Imagine a world in which Peyton Manning doesn't play for the Colts for a 6-game stretch.   Do you need more evidence than this?  In the preseason, the Colts may look like division winners.  In week 8, without their holy diver, they could be just another .500 football team.  

4. The Draft.  Like free-agency, but on a smaller scale, draft picks can transform a team.  Usually it takes a bit longer, but even so, just the throughput of new players makes predicting anything in the preseason hard.  

5. Luck.  Yes, yes, I've heard it too.  Luck is what happens when preparation meets opportunity.   Usually the people that say this are busy selling you the tee-shirt or poster with a 50% mark-up.  In the NFL, we all know, sometimes the wind kicks up and the ball hits the crossbar.  Or if you are the Raiders, there's always the tuck rule.  

Wanting to be the First in Line 

Now that I've probably convinced you of the complete and utterly silliness of trying to predict wins and losses in the preseason, let me be the first idiot to jump to the front of the line and do so.  According to my own proprietary formula, in which I give equal weight to last year's strength of schedule, Al Davis' monthly need for live-goat sacrifices, and the wonderlic scores of every nickel corner back on every roster, here is how I see it going for the Broncos in 2010:

Sept 12. at Jacksonville -- Tyson Alualu fooled everyone, except the Jags.   After all, he is Samoan.  And we all know Samoans kick ass.  Actually, Denver, humidity, and early games never mix.  LOSS.

Sept 19. Seattle --  Pete Carrol is back in the NFL.  It's just too bad the Seahawks are not still in the AFC West so the Broncos could get 2 wins per year.  WIN.

Sept 26. Indianapolis -- Age begins to catch up to Peyton Manning.  Unfortunately, he can still no-huddle the Broncos into the dust.   LOSS

Oct 3. at Tennessee --  Tim Tebow busts off some Wild Horses and shows Vince Young who is the better running quarterback.  WIN

Oct 10. at Baltimore --- Okay.  Okay.  Ray Lewis really does eat children.  That and the Broncos do poorly in early game trips to the east coast.  LOSS.

Oct 17. New York Jets --  Rex Ryan had stomach bypass surgery.  He doesn't look so mean anymore.  Also, Marc Sanchez isn't as good as you've read.  WIN.

Oct 24. Oakland - You're always the last in line, Oakland. WIN

Oct 31. at San Francisco - Mike Singletary can't suite up or play quarterback.  But his eyes will still pop out of his skull when he sees how good Robert Ayers has become.  Vernon Davis regresses.  WIN

WEEK 9 BYE - Ryan Clady injures his eyelashes playing basketball.  But it's thought to be nothing serious.

Nov 14. Kansas City -- Matt Cassel tries to switch teams by halftime because he likes McDaniels shotgun/spread more.  He's barred entry into the Denver locker room.  WIN

Nov 22. at San Diego -- Tim Tebow is now fully integrated into the Broncos offense with his own specialized sets and formations.   Philip Rivers looks into his own stadium to realize that his jersey sells 2nd to Tebow's...in San Diego.   WIN 

Nov 28. St. Louis - Alphonso Smith works himself back into the nickel corner spot and takes back an INT to the house.  WIN

Dec 5. at Kansas City -- December.  Kansas City.  Ouch.  LOSS.

Dec 12. at Arizona -- Denver sweeps the NFC West.  Why?  Because Arizona can't commit to a QB.  WIN

Dec 19. at Oakland -- You're always the last in line, Oakland.  WIN

Dec 26. Houston --  Brady Quinn injuries his hamstring in pre-game warm-ups, after a drunk female fan storms the field demanding some post-Christmas mistletoe .  The only home loss of the year.  At least it's to Kubiak.  LOSS

Jan 2. San Diego -- Jamal Williams really is one hour ahead of Philip Rivers.  The Broncos exorcise (and exercise) their demons.  WIN 

So there you have it.  The Broncos, on my watch, will finish 11-5 and win finish at the front of the line.  

Silly?  

No sillier than anyone else's preseason predictions.  At least I accounted for increased Tim Tebow jersy sales.

Go Broncos!