Commentary

A look at how the S-curve works

Originally Published: February 11, 2013
By Charlie Creme | espnW

With the football season in the rearview mirror, the sports world begins to focus more on college basketball. There's no better time to hit the reset button on Bracketology, bring everyone up to speed and brace for the sprint that is the next five weeks to Selection Monday.

Since most good ideas are stolen, this is a good time to point out that friend Joe Lunardi did this for ESPN.com's men's college basketball coverage. What follows is my version of bracket math, detailing the S-curve that the committee uses to rank all the teams in the field.

My brackets are an attempt at duplicating the committee's process as closely as possible for picking, ranking and placing the teams in the bracket. Bracket updates, which are a depiction of where teams would be if today was Selection Monday, will be weekly from this point forward. As you follow the updates, these are some things to keep in mind.

• The automatic bids are the current conference leaders. If there is a tie atop a league, head-to-head is the tiebreaker, if applicable, followed by highest RPI ranking.

• Host schools, by rule, have to play on their home floor in the first and second rounds. Host schools cannot play at home in the regionals.

• The S-curve is not used exactly the same way it is used on the men's side. Teams are given geographical preference (shorter travel) based on their position on the S-curve. Balancing the bracket is still a priority, but keeping the higher seeds closer to home in the opening rounds and regionals is also included in the decisions on placing teams.

• As part of its evaluation, the committee uses something called the "Nitty Gritty" report, which breaks down statistics like road record; W-L against top-50 RPI, 51-100 and 101-200, conference record, nonconference record, etc.

• Teams can be moved up or down one seed line to accommodate bracketing principles (this was done five times in this bracket). Such a scenario might be to avoid placing teams from the same conference close enough to meet before the regional finals. Also, sometimes the concept of host schools makes moving teams necessary since all the hosts must be separate for the first two rounds.

Here is how it all breaks down:

So, with 33 teams "safely" in the field, plus the other 21 automatic bids, 54 of the 64 spots are accounted for. That leaves 10 spots, which essentially accounts for the bubble.

The teams in are (in S-curve order): Duquesne, DePaul, Virginia, Villanova, West Virginia, Illinois, St. Joseph's, LSU, Kansas and Miami.

The teams out are (in S-curve order): Rutgers, Florida, Charlotte, Washington, Minnesota, Creighton, Arkansas, Georgetown, St. Mary's, South Florida, BYU, Wyoming, UTEP and Drexel.

If you are a fan of those teams on the out list, don't get too excited. Yes, your team in on the board. No, their chances of breaking into the field are not good. Most of the teams on the "OUT" list have lost plenty or just don't have any quality wins. Neither one of the those factors is likely to change in the final month. But they do warrant some consideration and analysis, which is why they are on the list.

Finally, here are some notes/observations on this week's bracket:

• Fourteen of the 16 sites included a host school. If Ohio State and St. John's were having the kind of years most thought they would, it's possible we'd have a clean sweep in that department, which has never happened under the format of pre-determined sites. Nonetheless, 14 of 16 is still a high number.

• Notre Dame is the only No. 1 seed not hosting, so the Irish remain susceptible playing a second-round game on their opponent's home floor. That last happened to a No. 1 seed in 2009 to Duke (the Blue Devils lost at Michigan State). It was avoidable this week, but one of my toughest decisions was at which of the remaining sites -- Queens or Columbus -- did I think the committee would place Notre Dame. Columbus is closer to South Bend, but Queens is a Big East location. I went with Queens with the logic being that the schools involved in the other part of that bracket (Florida State, South Carolina, Middle Tennessee and SMU) would have a better chance of drawing fans in Columbus, while the big New York following of Notre Dame, plus those same fans' familiarity with the Irish from playing St. John's, would maximize that attendance.

• As is the case most years, the bubble is not strong. The last few teams in, despite plenty of losses, have shown at least an ability to play at a high level (LSU beating Georgia, Miami over Penn State, Saint Joseph's over Maryland, Kansas over Iowa State). Those outside of the field haven't done that yet. That's the difference right now.

• Creighton checks in with a solid RPI but is not in the field. At this time of year, sometimes there are still RPI anomalies, and this appears to be one. The Bluejays lost two games this week and saw only a small drop. Also, remember Western Kentucky was 17th in the RPI in 2006 and did not receive an at-large bid. Creighton's omission is hardly unprecedented.

Charlie Creme | email

Women's College Basketball
Charlie Creme projects the women's NCAA Tournament bracket for ESPN.com.

Comments

Use a Facebook account to add a comment, subject to Facebook's Terms of Service and Privacy Policy. Your Facebook name, photo & other personal information you make public on Facebook will appear with your comment, and may be used on ESPN's media platforms. Learn more.