Bracketology: Archaic RPI Formula Must Be Replaced to Truly Judge Bubble Teams

RPI, three letters you are bound to hear as often as “bubble” and “Cinderella” in the coming days.

With conference tournaments in full-swing, experts and “Bracketologists” alike are crunching the numbers trying to figure out the 68 team’s worthy of a ticket to the exclusive dance.

It’s been the gold standard for evaluating each team’s resume, and stats like “record against top 100 RPI” become incredibly important.

It shouldn’t.

The RPI is a dated metric that needs to either be severely tweaked, or completely overthrown.

The NCAA Tournament committee uses the RPI formula (team’s winning percentage (25 percent), its opponents’ winning percentage (50 percent) and its opponents’ opponents’ winning percentage (25 percent)) to determine who is worthy of a bid and who is NIT-bound.

Aside from a few tweaks to the recipe, this has been the golden standard since 1981.

What’s the Problem?

With the RPI, your winning percentage carries the exact same weight of the team’s your opponents play winning percentage. So your RPI is affected by schools that weren’t even on the schedule. It’s completely out of the hands of that team.  

For example, Kansas State crushed Alabama back in December when the Crimson Tide were at full strength and a top-25 team. At the time it was considered a fairly large upset and a win the Wildcats could hang their hat on come March.

Is it their fault ‘Bama ended up dealing with multiple issues in SEC play and had a nasty four-game losing streak a month later against teams that don’t even know where Manhattan, Kansas is?

Absolutely not, but because ‘Bama’s RPI suffered, so did the Wildcats. In the span of three months a shocking upset is now nothing special. If anything, win or lose, playing a team that goes on to struggle negatively effects your ranking more than beating an average team in conference play.

Something is wrong with that.

With the tools we have in today’s world, the RPI looks ancient. In a recent rant on his ESPN radio show, Scott Van Pelt compared the system to an old man that is just a bit out of touch of today’s world:

"

(W)alking around with a big Walkman on his hip the size of his toaster, who is flipping over his cassette tape, who wants to run home to program his VCR on his standard-definition television.

"

He went on to call it the “worst metric in sports”.

The real problem is the way most people look at the RPI like it’s the end-all for college basketball metrics.

In baseball we have evolved with statistics to the point where a .500 pitcher can win the Cy Young. It’s not just about wins, loses and ERA. In college football, margin of victory is heavily considered into the massive computer system they use.

We can do this because of computers, because of excel spreadsheets and the ability to look up and plug in an entire night of scores in a matter of minutes. The formula needs to be more complicated, with much more factors incorporated into it.

What’s the Solution?

There is no definitive way to compare a six-loss team from the MAC to an 11-loss team from the Big East aside from actually playing a game, but Ken Pomroy comes close.

KenPom.com takes into account each and every possession. What opponents do aside from that particular game carries very little meaning. Pomeroy focuses on each individual team’s success with stats like point margin, the amount of close losses against strong opponents and a team’s offensive and defensive efficiency rating.

According to the Washington Post, almost 100 schools subscribe to his site and Pomeroy has even been a guest on Duke Coach Mike Krzyzewski’s radio show.

If you don’t like that there is Jeff Sagarin, who has one formula that only measures wins and loses without margin of victory and another where point spread counts, but only to a certain point. Blowouts can only count for so much, and beating other opponents handily that are considered strong means more.

There is no easy solution to truly gauge the top 68 teams every year. Whatever system is implemented is going to face harsh scrutiny no-matter what. That’s the nature of sports.

But it’s a tough pill to swallow knowing how valued the RPI is and how skewed the numbers can be. This should not be the end-all on how to evaluate collegiate basketball and potential solutions exist.

The biggest problem is convincing the 11 suits that make up the committee and have been using the RPI for 30 years to change. Like the BCS and a potential playoff system, the men in power are the last to jump on board with the ever-changing landscape of the world around them.

Who knows when things will finally change. Just be aware that when you are putting on you 'Bracketolgist hat and predicting the field this weekend, use the RPI as your metric of choice because that’s what the committee does, even if it’s more dated than a Walkman. But remember that they may not actually be the best 68 teams, even though that's the whole point. 

Something is seriously wrong and changes need to be made from the top. 

RIP, RPI.

Follow @BigLeagueEball