|Some at-large decisions are a layup. This one, however, was most decidedly not.
Photo by Steve Frommell, d3photography.com
By Pat Coleman
How unusual was the selection of UW-Oshkosh as an at-large team to this year’s men’s basketball tournament? It’s practically off the charts.
For several years, the NCAA men’s basketball selection committee has set a .667 winning percentage as a floor, where teams below that were not going to get selected as at-large teams to the tournament. That’s why this year’s selection of UW-Oshkosh, at 17-10 (.630), was a surprise.
The Titans’ strength of schedule soared over .600 as a result of the Wisconsin Intercollegiate Athletic Conference tournament, where Oshkosh lost at River Falls in the title game.
“I do think the 10 (losses) is a round number and it jumps out at people,” selection committee chair Kevin Vande Streek, the men’s basketball coach at Calvin, told Dave McHugh on Hoopsville on Monday afternoon. “I think if they were 18-9, nobody would even say anything, probably. We have to get past a number that says, ‘you have to have an SOS of this number, or a win (percentage) number of this’ and try to figure out, especially when we're getting towards the end, and say, ‘who do we feel would be the best team here?’ ”
However, in recent years, the committee has done indeed exactly that, and it's noticeable how this affects Division III NCAA Tournament at-large selections. Let's take a look at the data.
This is compiled by Drew Pasteur, an associate professor in the mathematics department at the College of Wooster, and he lists his methodology in a post on D3boards.com.
This chart shows all of the Pool C decisions made by the Division III men's basketball selection committee from the 2013 through 2016 tournaments. Teams that reached the tournament, of course, had a higher winning percentage or strength of schedule, as you would expect, since those are two of the primary criteria used to select teams. It's important to note that it doesn't include results vs. regionally ranked opponents, and a graph can't easily quantify head-to-head results, or results against common opponents.
Nonetheless, you can see where this is a good proxy for determining who gets those at-large bids, if not a perfect replication of the process.
The middle of the three lines, the solid one, represents a 50 percent chance at getting in the field, with the line above it representing a 90 percent chance and the bottom line, a 10 percent chance. There are more details about the outliers in the post linked above.
Now let's take a look at this year's data:
Babson was such an easy at-large decision that it doesn't even appear on this graph. Someone such as Denison, with a quite low strength of schedule, doesn't appear on the graph in the opposite direction. The only team in the safe zone between 50 and 90 percent was Concordia (Texas), which fared poorly in terms of results vs. regionally ranked opponents.
And then you see UW-Oshkosh, which is here at the extreme bottom right of the chart. And you see the cluster of teams that the committee deals with in these situations, from LeTourneau and St. Norbert -- one of whom was on the table when the selection process finished and one who was not even regionally ranked -- to the cluster of Loras, Carthage, Illinois Wesleyan and John Carroll.
Will this selection serve as precedent in the future? Anything would be speculation at this point.
"I guess we'll look a couple years from now and say 'that committee really messed up,' " Vande Streek said. "Or maybe we'll say that was a good thing. I think all along we have been encouraged to go out and find good people to beat, and we've had it the past couple of years with teams like Penn State-Behrend or Lancaster Bible and they're hurt a little bit by their conference in terms of how many (conference) games they play and not being able to get it, but whenever they have contacted national committee members, we have said, you have to play better teams, you have to beat better teams."
Whether this gets carried forward by future committees remains to be seen. The only thing we can say for certain is that when the makeup of the committee changes, so do the points of emphasis, sometimes faster than coaches can adjust their scheduling practices.