The debate over which conference is the nation’s best is fairly low on my list of priorities this time of year, mainly because there’s no accepted way of measuring conference strength. At least when comparing teams, one might agree on a method of comparison. Or the teams in question might actually play each other, providing some evidence to guide the discussion. Conferences don’t actually play each other and figuring out how to compare one group of teams to another is a challenge.
Typically, when someone asserts that this conference is the best or that conference is overrated, there is no interest in discussing the details of how that conclusion is reached. But the details are important because there are many ways to determine conference strength. Let’s look at a few of these methods and determine which conference can plausibly be considered the best in the land to this point.
Average strength of the teams. This is what I use on my site. It’s good enough for most cases, but if you’re more interested in the strength at the top of the conference, then it could leave you unsatisfied. The biggest weakness of this method is that an outlier can skew the rating. Whether San Jose State is merely one of the ten worst teams in Division-I or worse than about 200 D-II teams (as the Massey Ratings suggest) shouldn’t affect our perception of the Mountain West much, but it makes a difference when computing average strength. If you use this method and my ratings, the Big 12 is the best conference in the nation by a healthy margin over the Big East and the ACC. Though almost any ratings system you use will give you the Big 12 as the best conference using average strength.
Median strength of the teams. We can completely remove the San Jose State effect by looking at the median strength of the teams. As far the effect on the top conferences, this only increases the Big 12’s dominance since its middle teams are actually in my system’s top 20, while the ACC falls off rather quickly after you get past its top five. The Big East becomes the clear second-best conference in this scheme.
The “Tri-Mean”. Jeff Sagarin uses something called the tri-mean to rank conferences. This is essentially a compromise between the average and median, giving more weight to the average teams in a conference, but still giving some weight to the outliers. I haven’t run this on my numbers, but he has the Big 12 over the Big East and ACC and I suspect my results would be the same.
Projected tournament teams. Counting methods like this are generally unfair to smaller conferences. Consider that the 2011 Big East sent 11 teams to the tournament positioning itself as the best conference of all time in the minds of some. But the current configuration of the Big 12 can’t match this simply because it only has ten members.
This gets to the ultimate challenge of ranking conferences: handling leagues of different sizes. On the one hand, using a counting measure puts a smaller conference at a disadvantage, but a percentage measure is unfair to larger conferences. The latest consensus bracket at bracketmatrix.com has seven teams each from the ACC, Big 12, and Big East. The Big 12 appears likely to get 70 percent of its teams into the tournament. (I’m not as confident about the Big East accomplishing this.) In order for the ACC to do this, it would need to put a record-tying 11 teams in the tourney. At any rate, since current brackets have the ACC with two one-seeds and five teams seeded four or better, this method favors it as the best league.
Projected Final four teams. Maybe you only really care about the teams capable of winning a national title in each conference. Based on probabilities at TeamRankings.com, here are the expected number of Final Four teams for each of the top six conferences: ACC 1.08, SEC 0.67, Big Ten 0.55, Pac 12 0.44, Big 12 0.43, Big East 0.26. I can buy the ACC as the best conference if we are focused on the best teams, but this method seems to weight the top teams too much. I don’t hear anyone making the case that the SEC is the second-best conference or that the Pac-12 is better than the Big 12.
Projected Sweet 16 teams. We can dial back the love for the top teams by looking at the projected number of Sweet 16 teams in each conference. Again, based on the probabilities at TeamRankings, here’s how the top six conferences fare: ACC 3.32, Big 12 2.37, Big Ten 1.88, Pac-12 1.51, Big East 1.39, SEC 1.24. That matches national perception a little bit better. We still have the issue of this being a counting metric which leaves the Big 12 and Big East at a disadvantage. If we normalize for the number of teams in each league, which seems like the fairest thing to do, the Big 12 would come out on top here followed closely by the ACC.
Schedule difficulty. There is something to be said for a measure that in some way accounts for the strength of the entire conference. The top of the ACC is tremendous, but it also has seven teams currently ranked 100th or worse in my ratings. That’s almost half of the conference and it leaves me uncomfortable declaring the ACC the best in the land.
A more fair way to do this, and one that I think lines up better with the conventional thinking, is to look at the expected winning percentage of a good team against a round robin of conference opponents. The definition of a good team is up for debate but fortunately it doesn’t matter for this season. If we assume greatness (pythagorean rating of .98), the expected winning percentage against each conference would be: Big 12 .881, ACC .885, SEC .898, Big East .902, Pac-12 .911, Big Ten .919.
If we assume a fringe top-ten team (pythagorean rating of .92) we get: Big 12 .649, Big East .702, ACC .704, Pac-12 .726, SEC .730, Big Ten .746.
And with a bubble-type team (pythagorean rating of .81) we get: Big 12 .428, Big East .486, Pac-12 .520, ACC .522, SEC .540, Big Ten .546. Bubble teams pick up wins over the soft underbelly of the ACC while those opportunities are fewer in the Big 12 and Big East. Thus, as we weaken the test team in these cases, the ACC gets hurt more relative to the Big 12 and Big East.
Regardless of what quality of tournament team we are talking about, the Big 12 is the toughest conference to navigate. Though there’s a reasonable debate to be had between the Big 12 and ACC when considering the difficulty for an elite-level team.
There are many different ways to measure the quality of a conference, but just about any reasonable method is going to identify the Big 12 as the best in the land. Only if one ignores the size of each league and focuses exclusively on the very best teams in each conference can the ACC emerge as the nation’s top league.