I used to create a mathematical formula using money line and gambling lines to decide "expectations". I think ESPN and some of the others now use something similar as they create their "expected to win" chart during the games.
I mean, we were 6.5 point underdogs in week 1 against Buffalo and won the game. That, by definition, is exceeding expectations (and I would argue exceeding expectations by a lot). The same logic happened in week 2. We were 6 point favorites against Oakland and lost. Arguably, those two games kind of "cancel each other out", almost identically on the "expectation" factor. We were expected to be 1-1 after 2 games and we were 1-1 after 2 games, even if the wins happened kind of against the odds.
I mean, if you apply that same principle to every game using some funny math, you can get expected wins vs. actual wins and decide if Tomlin/team overachieved or underachieved.
Now obviously, this only takes into account the actual perceived level of your team. Bad teams can overachieve by winning 5 games. Good teams can underachieve by winning 10. Sometimes, as Bill Parcells says, you are what your record says you are. Expectations, by definition, are kind of arbitrary, perceived and imaginary. And fans sometimes have overly high expectations (which is why using gambling information seems more neutral).
There is also the thought of measuring expectations from year-to-year. If the expectations year-to-year are going down (in other words, you are favored in less and less games year to year), that might be the most damning thing about your team. If you are expected to win 10 games, then 9 games, then 7 games, you are really moving in the wrong direction and even "exceeding" those win totals doesn't cover up the general degrading of your team.
