The strength of a college basketball team's schedule should manifest itself in, basically, two rules:
1. Playing good teams gives you an opportunity for a good win.
2. Playing bad teams both gives you an opportunity for a bad loss and removes an opportunity for a good win.
It is a give-and-take. Either give yourself a host of wins that the NCAA Tournament committee can ignore (while risking a catastrophic loss), or give your resume a pleasant spark. But the tournament committee seems to look at strength of schedule in three ways: a reward for big wins, a punishment for bad losses, and a reward or punishment for the aggressiveness of your schedule. In theory, this is fine -- we want as many lovely marquee games as possible in non-conference play (even though we half-ignore basketball until February). In practice, however, it means that certain teams get rewarded or punished for reasons beyond their play on the court. That is unacceptable, especially when we take a look at exactly how strength of schedule tends to be measured.
Low Fat Cupcakes
The commonly accepted strength of schedule measure appears to be that provided by Jerry Palm and his RPI. The RPI is, despite its detractors and limitations, a solid, at-a-glance tool, and it serves a purpose. It is an objective, numbers-based tool that, while not nearly as useful and descriptive as other tools, gives us a much-needed non-human component. For instance, you can very quickly look at a team's high and low points using the per-team breakdowns on Palm's site. Such a breakdown very quickly shows us an enormous hole in Colorado State's resume. Yet, because of the way strength of schedule is used, somehow RPI became a tool that helped the Rams get into the tournament.
On the road, Colorado State played seven opponents ranked 51st or worst in RPI. They went 2-5 in those games. They beat No. 147 UTEP by three points and No. 168 Air Force by 10. Meanwhile, they lost to No. 73 Northern Iowa (83-77), No. 83 Wyoming (70-51), No. 97 Stanford (64-52), No. 113 TCU (75-71) and No. 173 Boise State (70-69). It probably goes without saying that they also went 0-4 versus Top 50 opponents (Duke, UNLV, San Diego State, New Mexico) as well. (Average loss in those four games: Opponent 82, CSU 61.) They did beat UNLV, San Diego State and New Mexico at home, and they did beat two other NCAA Tournament teams -- Colorado and Montana -- at home as well. (They also lost to Southern Miss by a whopping 21 points at home.)
But their play away from Fort Collins very clearly painted a picture of a team that probably wasn't deserving of inclusion in the Field of 68. And yet, the 19-11 Rams, who went just 9-7 in the Mountain West, not only got into the field, but got in rather easily. Why? Because of strength of schedule.
The Rams' non-conference strength of schedule was a cool 26th in the country. Good for them. It isn't the worst thing in the world to want to reward a team for taking on challenges, and using this single number, it appears that Colorado State indeed took on some heady challenges before Mountain West play began. Only … they didn't.
They played Duke, yes, and Southern Miss as well. But those were their only two RPI Top 50 non-conference opponents. How exactly did their schedule rank so high? Because they ate lighter cupcakes than everybody else. The Rams played just four teams ranked 200th or worse in non-conference play, and only one ranked worse than 250th (N.C. State played five worse than 200th, while BYU played four against teams ranked 270th or worse), and they managed to play five teams in the no-man's-land area of 50th to 100th. They played two opponents good enough for at-large inclusion in the tournament and lost to those temas by a combined 44 points, but because they were willing to play No. 74 Montana and No. 73 Northern Iowa, instead of, say, No. 275 Montana State and No. 262 Northern Colorado, they got into the tournament.
Their schedule got them in, even if their performance against the schedule seriously left something to be desired. Again, theory and practice significantly disagree when it comes to rewarding strength of schedule.
No. 43 Vs. No. 222
The committee's most surprising at-large decision had to be the inclusion of Iona in the Field of 68. Tim Cluess' Gaels produced a solid resume, finishing 15-3 in the MAAC and taking out favored Nevada on BracketBuster Saturday. They knocked off one-time at-large candidate St. Joseph's in November, and their only two non-conference losses (Purdue and Marshall) were at least semi-respectable.
When Jim Nantz and company talked to committee chairman Jeff Hathaway after the field was revealed, they asked him about the inclusion of Iona over a team like Drexel, whom many (including myself) thought deserved inclusion. Hathaway quickly dismissed the question because Iona's non-conference strength of schedule was 43rd, while Drexel's was 222nd.
As we saw later, Drexel wasn't even the first team cut out of the field; they were fourth, behind Oral Roberts (non-conference strength of schedule: 17th), Miami (99th) and Nevada (193rd). If Drexel had been included, they would have had the worst non-conference strength of schedule of any at-large team included.
That's great and all, but it ignores two simple facts: Drexel's best wins were better than Iona's, and Drexel's worst losses were not as bad as Iona's.
- Drexel's Five Best Wins (according to Ken Pomeroy's rankings, which we will use here because there is more of a quality aspect involved): No. 46 VCU, No. 80 Georgia State, No. 82 Cleveland State, No. 89 Princeton, No. 104 Fairfield. Average Pomeroy rank: 80.2. (Average RPI of top five wins: 79.2.)
- Iona's Five Best Wins: No. 65 St. Joseph's, No. 78 Denver, No. 104 Fairfield, No. 105 Richmond, No. 112 Nevada. Average Pomeroy rank: 92.8. Average RPI of top five wins: 77.4.)
- Drexel's Three Worst Losses: No. 213 Norfolk State, No. 163 Delaware, No. 80 Georgia State. Average Pomeroy Rank: 152.0. (Average RPI of bottom three losses: 140.3.)
- Iona's Three Worst Losses: No. 226 Siena, No. 221 Hofstra, No. 128 Loyola (Maryland). Average Pomeroy Rank: 191.7. (Average RPI of bottom three losses: 209.7.)
When you look at the actual resumes instead of simply the non-conference strength of schedule, Drexel trumps Iona in just about every way possible. But they "only" played Cleveland State, Virginia, St. Joseph's, Princeton and six cupcakes ranked 200th or worse in RPI in non-conference, while Iona played Purdue, Marshall, Nevada, St. Joseph's, Long Island and four teams ranked 200th or worse.
(Meanwhile … the fact that those two non-conference schedules were separated by 179 spots further enunciates why this may not be a very good tool to use for these matters. We need to make sure we know what numbers are telling us before we actually use them.)
Heading into the 2011-12 campaign, Missouri was looking at a non-conference schedule that included the following:
* vs. Notre Dame, a 2-seed in last year's NCAA Tournament, in the CBE Classic in Kansas City.
* vs. California, the preseason No. 23 team in the country, in the CBE Classic.
* vs. Villanova, the preseason No. 33 team in the country, in a "semi-road" game at Madison Square Garden.
* vs. Illinois, an annually solid team (if one seeing diminishing returns from the Bruce Weber era), in St. Louis.
* at Old Dominion, a team that lost via last-second bucket to eventual national runner-up Butler last season.
* Five scheduled cupcake games, plus two more as part of the CBE Classic preliminary rounds.
In all, nobody thought this would be a world-beater schedule by any means, but it did not look as if it would barely crack the Top 300 in terms of non-conference strength of schedule. But then Illinois and Villanova tanked, and Old Dominion, with whom the trip to Virginia had been scheduled for two years, faded after losing quite a few seniors. Notre Dame lost Tim Abromaitis and just looked lost, period, until February. California made it into the NCAA Tournament but failed to live up to any sort of Top 25 hype. And, of course, the cupcakes turned out to be really, really bad. The Tigers ended up playing seven teams ranked 201st or worse, and they ended up with both a non-conference strength of schedule ranking of 294th and an overall RPI ranking of just 10th despite a 30-4 record.
When asked about Missouri's status as a potential 1-seed following the revealing of the bracket, Hathaway said, simply, that they were not only out of consideration for a No. 1, but they were the lowest-ranked No. 2 because of their strength of schedule. It didn't matter that they had as many wins (11) versus the RPI Top 50 as anybody else in the Field of 68; it only mattered that they didn't try hard enough to schedule better teams in non-conference … even if, technically, they had.
I never really considered Missouri seriously, even though the team certainly plays like a top seed, because its profile just didn't match up well with the Selection Committee's typical opinions on scheduling. Perhaps in the future, teams in the Tigers' position -- and really any team that wants to qualify for the Tournament -- will finally take the Committee's opinions to heart. If they do, college basketball fans will have plenty of November and December games to look forward to in the future.
Now, to be sure, Missouri's chances at a No. 1 should have, and did, take a fatal blow when Michigan State won its conference tournament. There was very little reason to argue with any of the No. 1s chosen, especially considering, among other things, that Missouri likely had the single worst loss (to Oklahoma State) of any of the top eight teams. But viewers were left with a bit of an impression that, should Missouri have lost to Baylor (the top-ranked 3-seed) in the finals of the Big 12 Tournament on Saturday, the 29-5 Tigers might very well have fallen to a 3-seed despite what would have been a 10-4 record versus NCAA Tournament teams (with five wins coming by double digits).
Quite a bit of Missouri fans' potential outrage was quelled by the fact that their draw ended up perfectly reasonable. But the remarks by Hathaway once again left us with the impression that who you play is infinitely more important than who you actually beat; not only that, but most analysts seemed to agree that Missouri should be punished for not trying hard enough to schedule up, potentially giving them a lower ranking/seed than their play deserved because of contracts signed and agreements made by someone other than the players themselves. Missouri was the anti-Colorado State in this regard. Even though all major conference teams (and most mid-majors) schedule sure wins, Missouri's cupcakes were fattier than others', and they nearly paid a significant price.
Aside from the cupcakes, however (and the idea of judging a team for reasons beyond "How good are they?"), the problem with this line of "Team A should be punished, while Team B should be rewarded," of course, is the idea that, in advance, you know with certainty who is going to be good or bad from year to year. Last year, a schedule featuring Notre Dame, California, Villanova, Illinois and Old Dominion would have been considered reasonable. This year, four of those teams regressed, and the fifth (California) did not live up to expectations. Obviously if you schedule more big-name teams -- annual powers like Kentucky, Duke, Kansas, Syracuse, etc. -- you can be more assured of ending up with a better strength of schedule. But among other things, those teams are infinitely less likely to agree to home-and-home series; a good portion of the time, you must draw them in an early-season tournament or a made-for-TV "classic" (like Coaches Vs. Cancer). And when you agree to participate in a given tournament, you obviously then have little control over who you might play.
Really, I have two recommendations for the committee when it comes to strength of schedule.
- Ignore the number. In the end, what matters to your resume should be what I mentioned at the top: who you beat, and who beat you. If you arrange a weak non-conference schedule, it punishes you by quite simply giving you fewer opportunities to prove yourself. (And if you do prove something against really weak teams, it probably isn't good.) A tool like RPI can be used in positive ways when it comes to the quality of your best wins and worst losses. But leaning heavily on a given number is erroneous thinking, especially if, as in the example of Drexel vs. Iona above, the difference between the No. 222 and No. 43 schedules is, for all intents and purposes, a trip to Purdue and two extra cupcakes.
- If you aren't going to ignore the number, then adjust strength of schedule to ignore the (lack of) quality of a team's cupcakes. If you are an NCAA Tournament team, you are almost certainly going to beat either the No. 200 team in the country or the No. 350 team in the country, potentially by equal margins. Since so many teams schedule weak games for sure wins, the "quality" of the sure win should not even remotely matter. It would make sense, then, to assign any team with a ranking worse than a certain benchmark with a certain ranking. For instance, if your cupcake opponent ranks 225th or 325th, it goes in the books as 225th. Most teams' schedules can be broken into sure wins, possible wins, and major challenges. If you are going to be punished or rewarded for your mentality in scheduling in a certain way, then one sure win should be viewed as any other. In terms of perceptions, your "schedule strength" is determined by the best teams you played, not the quality of the worst.