Its always debatable when the RPI begins to "matter". Some say its always completely useless, others think its a valuable tool at the end of the season, and other's think it matters at the end of non-conference play. Non-conference play isn't officially over, but most leagues are entering conference play now. Leagues like the MVC only have 1 non-conference game left this week, and then the BracketBuster event later in the year. Most Pac-10 teams only have 1 non-conference game left, like UCLA who takes on West Virginia later in the year. Gonzaga's one of the few teams left with 3 non-conference games, with Nevada, Virginia, and Memphis still on the schedule. For my money, non-conference RPI is the real statistic that matters.. so now's basically the time when the RPI matters.
My big issue with the RPI is that it doesn't take into account whether games are played at home or on the road. Duke currently has the #1 strength of schedule (SOS), this despite the fact they haven't played a true road game yet. It's a complete farce to think a team can play the hardest schedule without playing road games. Sure, Duke has played a lot of above-average teams in the 50-100 RPI range at home.. but going on the road to play teams in the 100-150 range is more of a challenge in my opinion.
Because the RPI doesn't take into account whether games are at home or on the road.. the SOS for teams that stay at home like Duke can be misleading. However, it can also show just how pathetic teams's schedules are.. when they have a low SOS and they've played all their games at home. UCONN has an SOS of 311.. which is the worst of any power-conference team. Yet UCONN is the ONLY team in the entire country that has yet to play a road game or even a neutral court game. For some reason though, the Huskies are ranked 9th in the AP Poll.. this can only be due to bias amongst the voters.
UCONN though isn't the only team that has yet to win a game away from home. Louisville is 0-2 in road/netural (R/N) games, Virginia's 0-3, Washington is 0-1 (hasn't even left the state), and Missouri is 0-2. All of these teams are given respect, and either are ranked or have receive votes at some point in the season.
Not all power-conference teams have taken the easy way out by playing a bunch of home games. Villanova's an impressive 6-1 in R/N games, while Oklahoma State is 5-1 along with Alabama. It's a mystery how Villanova can be 9-2 overall and 6-1 away from home and not receive a single vote in the AP.. while Syracuse is 10-3 with just 1 win away from home, yet is ranked 31st by AP votes. Just one of a hundred examples while the voting is so biased. People often complain about the RPI, or the BCS in football.. but these rankings put the human voters to shame. The only fault with the BCS is that it simply weighs the human vote so heavily..
If R/N records really matter.. here are some teams that should rise to the top in conference play. Butler (7-1 in R/N) games is obviously the favorite in the Horizon now. UMASS (7-2) and SLU (4-2) should make a run in A-10 play. Drexel and VCU, both 5-2, were the best R/N teams in the CAA, though Drexel did it against much better competition. Oklahoma State (5-1) has been the cream of the Big 12's mediocre crop. Villanova (5-1) should surprise in the Big East. Davidson (6-3) and Appalachian State (5-3) will battle it out in the Southern Conference. Jackson State (5-7) will own the SWAC, why SELU is the Southland favorite according to R/N records. Despite a slow start, Bucknell (5-4) still did enough in R/N play to be worthy Patriot League favorites. Southern Illinois (5-2) could surprise higher ranked MVC teams like WSU and Missouri State according to R/N records..