clock menu more-arrow no yes

Filed under:

Could a Syracuse win over Georgetown actually hurt the Orange’s RPI?

New, 11 comments

These guys ruin everything...

NCAA Basketball: Connecticut at Syracuse Nicole Sweet-USA TODAY Sports

You know the Ratings Percentage Index (RPI) is an imperfect metric. Yet, it’s the primary one that the NCAA utilizes when determining who makes the postseason. So as much as we admonish it, it’s important to have quality numbers over there, however they come about. KenPom is better to evaluate a team-versus-team argument, but as Pomeroy has agreed with himself, it’s not for seeding purposes.

The RPI lays out as follows:

Your team’s win percentage (WP) x 0.25 + opponent win percentage (OWP) x 0.5 + your opponents’ opponent (OOWP) win percentage x 0.25

There’s an additional caveat to your WP as well, which weights wins and losses (but not opponent or opponents’ opponent wins and losses) based on where they occur. A home win is worth 0.6 wins and a home loss is worth 1.4 losses. True road wins are worth 1.4 wins, and true road losses are worth 0.6 losses. Neutral site remains at one apiece.

This is all a long way of bringing us to the Syracuse Orange’s Saturday opponent, the Georgetown Hoyas. What’s their current stat profile look like?

WP = 0.7

Okay, that’s fine, since they’re undefeated.

OWP = 0.35

Uhhhh...

Strength of Schedule = 351

(no, I’m not sitting around and calculating OOWP right now -- it’s far too extensive of a process; just trust that SOS is a worthwhile stand-in)

Georgetown’s currently tied for 195th in the RPI.

And now let’s look at Syracuse’s numbers (WP, OOWP, SOS)

WP: 0.58

OWP: 0.54

SOS: 80

RPI: 25

Assuming Syracuse wins, the WP number goes up to 0.66. the OWP jumps up as well to around 0.6 or so. But SOS is sure to take a dip by beating a team ranked just above the RPI’s top 200. And obviously we now inherit all of the Hoyas’ trash opponents in our OOWP, which further brings that number down. Georgetown could actually jump up in RPI themselves, however, just by facing an Orange team with better computer numbers.

And therein lies the flaw with RPI. We get dinged because Georgetown played bad teams. The Hoyas get a boost because SU’s the one good team they’ve scheduled.

Now, much of this one-game effect disappears as the season goes on. Sample size right now is pretty small, so RPI is going to skew in one direction or another much more easily. Win or lose, this could just be a footnote to Syracuse’s season by the time we hit March.

NCAA Basketball: Colgate at Syracuse Rich Barnes-USA TODAY Sports

Still, as we know, RPI’s not the only factor for NCAA Tournament bids, though. The committee looks at who you beat and where. A road win would be a plus for the Orange, but would it matter as much if it was against a team that finished around 200 in the RPI?

This game makes sense when Georgetown’s RPI is top 50-100. It’s no major penalty to the loser (other than pride) to fail. For the winner, you get the positive of beating a potential tournament team, plus the win over a hated rival to boot.

This leaves us in the precarious situation of hoping the Hoyas improve. It may be what’s best for the rivalry. But is it what’s best for us? As I started getting into this morning, it’s hard to tell.

(forgive the rough math above, as I’m no Hoya Suxa, and thus can’t number crunch on the level that this topic would otherwise require -- still the sketch above should provide an idea of how dumb the RPI always is and how pointless it deems this game this year)