South Carolina didn't clinch their regional while in Hoover, and there has been talk that they could lose it to Arkansas. Early on Saturday the twitterverse consensus was that an Arkansas win over LSU would move the Columbia Regional to Fayetteville. The rationale was that Arkansas's RPI would move up into the 20s which coupled with a third place finish in the SEC would be enough to earn a host site.
Of course, Arkansas did not beat LSU on Saturday, and it looks like their RPI will remain in the 30s. That alone should be enough to deny them a host site, but they are still in the conversation. For the most part the narrative has been South Carolina or Arkansas. If that is the option the committee is looking at, Carolina should still host.
In fairness, the Hogs have two very good arguments:
- They finished 3rd in the SEC, in front of South Carolina.
- They swept South Carolina in their regular season series.
Those arguments are relevant, but they are not conclusive. First, conference finish. Conference finish is relevant but it is incomplete. Arkansas played 29 regular season conference games and 28 games outside of the conference. You have to consider all those games.
The same is true of the head to head results. Three games are too few to compare teams for the season. Head to head results are relevant when the teams are very close in other metrics, but in the key metric, RPI, the teams aren't close.
Arkansas is going to finish about 30th in the RPI; South Carolina is going to finish 13th. The two aren't close, and RPI has historically been important to the selection committee.
RPI aside, the Gamecocks still finished with more wins and fewer losses against a harder schedule. USC finished 39-18 against the nation's 14th toughest schedule. Arkansas finished 37-20 against the 48th toughest schedule.
Bottom line: Over the course of the season South Carolina won more games and lost fewer against a harder schedule. How can you discount that?
- RPI and SOS numbers come from Warren Nolan's website, which you can find here.
