Arrow-right Camera

Color Scheme

Subscribe now
Gonzaga Basketball

John Blanchette: Gonzaga proving that RPI is no longer the golden child of college hoops metrics

So the Big Giddy that figured to linger at least through the week after the Moraga massacre? Not so much.

Oh, Gonzaga kept rolling Thursday night, but the 76-46 steamrolling of Loyola Marymount had the feel of a mercy killing – the Lions already well under the pile in the West Coast Conference and missing two of their three top scorers to boot. The Kennel Club, meanwhile, was so pumped by last Saturday’s blitz of Saint Mary’s that it left its two end sections of seats for this one unclaimed.

Hey, no one needs a college degree to smell a blowout in the works.

But maybe the sizzle got doused way back on Sunday, when the NCAA selection committee did its in-season bracket preview for the second year. And the most revealing part of the big reveal was that the committee turned its latest statistical toy into the same old worn rag doll.

Also, Gonzaga wasn’t one of the top 16 seeds.

This wasn’t what had Bulldogs coach Mark Few shaking his head this week. He knows the Zags will be seeded somewhere come Selection Sunday and will show up and play their opposite seed, which will be a good team whether it’s a 13 or a 12 or whatever. You hope for the best matchup, but a seed number never guarantees it.

But Few’s been part of a National Association of Basketball Coaches committee that’s made recommendations to the NCAA about its assessment process – trying to build the better mousetrap knowing all the while that occasionally the cheese will disappear with no critter in sight.

“It’s never going to be perfect,” Few said. “But there’s a better way that will advance us down the road.”

Does it involve blowing the RPI to smithereens? Does it?


Sorry, no.

The ever-maligned Ratings Percentage Index has been the committee’s Linus blanket since ones started being done, always with the doth-protest-too-much disclaimer that it’s strictly a reference tool.

Right. The NCAA would rather put Duke on probation than wean itself off that metric.

But there are other metrics available now – different, deeper, better – that the NCAA has been slow to warm to, and Few’s committee has tried to turn up the heat.

Ken Pomeroy’s ratings are the designer drug of college basketball’s literati. Those and Jeff Sagarin’s USA Today numbers are the “predictive” measurements; the RPI is “results based,” which would seem to be what you’d want, except the RPI has a haywire notion that your winning percentage has the same value as that of your opponents’ opponents. Among other dubious math.

Which has led to the nonsense of Gonzaga being 36th in Thursday morning’s RPI, but eighth in KenPom and 10th in Sagarin. Conversely, Oklahoma is 30th in both of the latter, 21st in RPI.

Guess which team snagged the last seed in the NCAA’s early reveal?

Few led the charge in his group to use a composite of four or five rating systems. The NCAA – insisting it will eyeball those numbers – still can’t bring itself to make the leap, well, official.

“I don’t know if it’s so much that they can’t let go,” Few said, “but it’s, ‘We don’t know what these other metrics are all about.’ It’s like there’s some deep, dark secret involved, or they think it can be manipulated. Well, probably – but the RPI’s clearly been manipulated over the years.

“Or maybe manipulated is the wrong term. With any statistical measure, you have outliers. If you run a composite, we think you have a better chance of minimizing those.”

One suggestion the NCAA did glom on to was the quadrant concept – moving away from all victories over top 50 or top 100 teams being equal, and instead lumping home wins over top 30 teams, neutral-court wins over the top 50 and road wins over top 75 in the prized Quadrant 1 category.

Except, oops – it’s based solely on that pesky RPI again.

And the Zags still have a gripe. They had five Quadrant 1 wins when the committee made its announcement. Of the teams on the No. 4 line, Oklahoma had six, Tennessee four, Arizona three and Ohio State – which lost to Gonzaga by 27 points – two. But their resumes are laced with stronger mid-level wins, so once again the Zags get punished for Loyola Marymount being lousy.

Eye test, anyone?

Actually, Few wouldn’t necessarily object – if there were more, in his words, “basketball people” on the committee. He’d even recommend – horrors – media members “who are involved in the process and really understand, and aren’t just watching it because they have to.

“I can watch these teams and know, ‘Hey, they’re really good’ or “They’ve got some holes.’ If you don’t have that confidence, you’re going to lean really heavily on a number.”

And the wrong one, at that.