June 5, 2013
Minnesota’s 2012-13 RPI: Skill vs. Luck
The Minnesota Golden Gophers had the nation’s 34th best RPI on Selection Sunday. Despite the RPI’s inherent flaws, Minnesota’s RPI ranking likely was a significant aid in making a case for an invitation to the 2012-13 NCAA tournament.
Many have noted the Gophers’ top-5 strength of schedule (“SOS”) ranking and called it “difficult” and “impressive”. How much of the SOS should be attributed to intelligent scheduling as opposed to factors at least partially out of the team’s control?
We’ll share our thoughts. The analysis we’ve done is a bit technical, but we’ve tried to limit our content below to a more understandable, high-level commentary.
We believe it’s fair to give Minnesota a good amount of credit for their 2012-13 scheduling, but it’s also apparent that a number of other factors resulted in an SOS that pushed them into the NCAA tournament.
Did SOS really help Minnesota’s RPI a lot?
Yes. Heading into the tournament the Gophers had an RPI of 34 despite having an adjusted winning percentage (“AWP”) of 104. The AWP is the only piece of the RPI that looks at an individual team’s win-loss performance. The remaining components of the RPI consider what other teams did against opponents other than your team. The AWP is adjusted for where a team won or lost a game (for example, the credit received for a road win is greater than a home win).
Of the top 68 ranked teams per the RPI on Selection Sunday, there were only two with a worse AWP than Minnesota. Villanova (#52) was had the 117th best AWP in America and was given a 9-seed in the NCAA tournament. Tennessee (#59) came in at 105 and played in the NIT.
Villanova and Tennessee both had similar AWP’s as compared to Minnesota. Their RPI rankings were considerably lower as a result of their strength of schedule.
This does not mean Minnesota played more challenging nonconference games than Nova or the Vols. In fact, the opposite is probably true.
Once you get past neutral site games against Duke and Memphis, there weren’t any other games in which Minnesota had a tough challenge on paper. 2012-13 Gopher opponents Stanford, Florida State and Southern Cal all have recognizable names, but those schools combined for a 51 wins and 49 losses. None of them made the NCAA tournament.
Prior to SEC competition, Tennessee played Oklahoma State on a neutral court; Georgetown and Virginia on the road; and hosted Memphis, Wichita State and Xavier.
Factor #1 – Avoiding the awful teams
There are a few major factors in Minnesota’s top-5 SOS (overall, including conference and nonconference foes) in 2012-13. Factor #1 can be attributed mostly to the team’s scheduling efforts.
A key to a strong SOS is playing a minimal number of awful teams. There are many teams that should do well in their respective conference and post a good win-loss record for the season, yet aren’t a school that should give Minnesota a very tough test. These are the teams you want to play.
Only two Gopher nonconference opponents had a pre-tourney RPI of 200 or higher: American (#278) and North Florida (#230).
Tennessee played more games that could be described as “difficult” than Minnesota did, but they also played some really bad competition including Kennesaw St. (#341), Presbyterian (#340), Western Carolina (#234) and UNC-Ashville (#217).
Whether Lafayette (who the Gophers spanked 75-50) or Kennesaw St. visits Williams Arena, the expectation is going to be the same: Minnesota should win easily. However, the RPI sees those two opponents far differently. This is mostly due to Lafayette’s record being 18-15 and Kennesaw St. finishing just 3-27.
Factor #2 – Neighbors to the West
Minnesota was helped by their neighbors to the west who frequently visit Minneapolis in the nonconference. Both South Dakota State (who was without Nate Wolters, but the RPI doesn’t know that) and North Dakota State had excellent 22-9 regular season records (against D-I teams, including conference tourney).
These two teams are regulars on Minnesota’s schedule. It just so happens that both had very good win-loss records in 2012-13.
Factor #3 – Battle 4 Atlantis bracketing
The Battle 4 Atlantis was in its second year as an exempt tournament. In year one the championship game featured Central Florida and Harvard. In year two the final game was upgraded to Louisville vs. Duke.
Minnesota saw a nice SOS bump by being placed into a first round match up against the Blue Devils. By losing, the Gophers were tremendously rewarded by falling to a game against Memphis.
The Tigers were a decent team, but the they are especially rewarding as an opponent because of their great record. It’s true that Memphis skated through to a 19-0 record against Conference USA teams, but Minnesota’s SOS gets rewarded for the Tigers’ conference dominance.
From a SOS perspective, getting to playing Memphis is about the best thing you could hope for after a loss in a nonconference tournament.
Factor #4 – Big Ten schedulers
The Big Ten schedule worked out well for Minnesota. Although they only played Ohio State and Michigan once, their other single-plays were Penn State and Purdue.
In addition, the Gophers were fortunate to play Illinois (strong overall record) in the Big Ten tournament.
Minnesota’s average (pre-NCAA tournament) Big Ten opponents’ win-loss percentage was approximately 61.5%. The Gophers’ weighted-average (weighted for how many times Minnesota played each team) Big Ten opponents’ win-loss percentage was approximately 62.0%.
Quantifying the Impact
Let’s just consider Factors #2 and #3 above. We’ll make the following changes to the 2012-13 schedule and determine the impact of the lower SOS on Minnesota’s RPI:
a) [Factor #2 – SDSU & NDSU]: Assume Minnesota replaced SDSU and NDSU with an decent opponent that carried an RPI of around 100. We’ll use Tennessee State (RPI of 105 and an NBA prospect in Robert Covington) as the replacement.
b) [Factor #3 – Battle 4 Atlantis]: Assume Minnesota played Missouri and Northern Iowa instead of Duke and Memphis.
Assuming only these two changes, we estimate that Minnesota’s Selection Sunday RPI would have been 51 instead of 34. The day would have been far more uncomfortable for the Gophers.
Of course, had the Big Ten schedule also worked out differently – for example, if Minnesota played Purdue twice and Michigan State only once – or if Minnesota replaced SDSU and NDSU with awful opponents, the impact could have been enough to push the Gophers off the bubble.