Tuesday, September 8, 2009

Division Sandwich

Theory:  A team will not play as hard for a non-division game, when sandwiched between division games.
Why? Teams know that division games are extremely important.  They just played one and have another one next week.  The current opponent might be overlooked.

I went through data from the last 7 years to find a trend with the division sandwich game. There were WAY too many scenarios to work with, so here were the criteria I chose for this study:

  • Game is week 10 or later
  • Game is played by a team at least 2 games over .500
  • No data from Week 17 is used (so no games played week 16 or 17).
Here we go:
Per the criteria, these are clearly good teams overall- a .67 winning percentage going into the game in question.  We see the effect a division game has even on good teams, with the Division Game 1 (ie. the game before the sandwich game) having only a .59% winning percentage.  The sandwich game itself shows some of the issues I was afraid of, with a .54% winning percentage.
Lets look a bit deeper.

Ok.  This is good.  The .54% is broken down nicely depending on whether the sandwich game is at home or on the road.  [Side note:  We have previously discussed that the win % after a division win is 52% which matches these numbers very nicely].  
The Division Game 1 Loser does the best in the sandwich game, as expected, because they are a good team that's pissed off they lost now paying at home.
Let's look even further:
The break down of home vs road still looks nice, but nothing earth shattering.
What do we learn from this?
A non-division game played in between two division games are usually called a "trap game".  This is definitely true to an extent.  The good teams are still winning at over 50%, but that is a far cry from their record to date on the season.   But, if you separate the games based on home vs road, this game can definitely be considered a trap game.
[I was not able to get spreads on games back to 2002, so the data is a bit flawed.  We don't know if the teams were supposed to win the games or not.  Nonetheless, with the teams excellent record going in to the sandwich game I would expect most of the teams to be favored, road or home.
Eventually, I will see if these numbers differ from games after a division game home/road.]
Games to Note:
TBD half way in to the season.


Super Side Note-  The game I lost on last year was a sandwich game with ATL winning the first at home then playing DEN at home.  Those games are only 4-4 total.  Not the odds you want for survivor.

No comments:

Post a Comment