Dr. Jeckyll and Mr Hyde: The Michigan St. Story 2003-2006

There are a lot of adjectives-none of them complimentary-that can be used to describe John L. Smith's four year tenure as head coach of the Michigan State Spartans, but no adjective is more applicable than to describe the teams under Smith's leadership as inconsistent. Usually, Michigan State had the habit of playing either up or down to their competition, beating a top ranked team one week only to lose to a below-.500 squad the next. Not only that, but they would play one half like a real football team and then the second one like a Pop Warner team full of unathletic, uncoordinated little kids.

Taking the first half-second half observation a step further, I looked at those Michigan State teams and see if the first half of their seasons were significantly better than the second half of their seasons. For those seasons in which Michigan State played an odd number of games, I always included half of the games plus one in the first half and classified the other games as being part of the second half.

It turns out that there were significant differences between Michigan St. in the first halves of seasons and Michigan St. in the second halves of seasons. For the offense, all of the significant differences had to do with passing: attempts, yards per pass attempt, and yards per catch. All of these differences are easily explained. Whereas Michigan State averaged 9.5 more points per game than their competition in the first halves of seasons, they averaged 5.4 points per game less than their opponents so they were forced to pass more in an attempt to come back to win those games. The difference in yards per pass attempt (8.0 to 6.5) and yards per catch (12.4 to 10.9) showed that those extra pass attempts did nothing to help the teams.

The significant differences in defense go hand in hand with the ones on offense. Against Michigan State in the second half, opponents passed less and ran more, indicating they were doing more to protect their lead and run out the clock than they were to score more points. Also, Michigan State did a horrible job of stopping the run in the second halves of seasons, both in terms of total yards (123.0 yards per game to 179.0 yards per game) and yards per rush (3.6 to 4.4). Needless to say, they gave up a whole lot more points in the second halves of the four seasons (22.3 points per game to 31.9 points per game).

After I completed this study and looked back at the results, I wondered if the data were not being skewed by the fact that most teams schedule their easiest opponents in the first halves of seasons. To avoid this problem, I eliminated the teams I deemed "cupcakes" and ran the test again. The results did not change so Michigan St. wasn't just inconsistent from play to play and from half to half and from game to game, but they were also inconsistent from the first half of their season to the second.