You think preseason basketball polls are worthless? Think again
There's a narrative that exists and suggests that "preseason rankings are worthless" in college basketball. But it's not actually true if the goal is to know in advance which teams are most likely to be really good.
Some people, for reasons I'll never completely understand, love to insist on Twitter, message boards and talk radio that "preseason rankings are worthless" even though they're actually pretty good indicators of what's to come in college basketball. Ken Pomeroy, who is smarter than all of us, explained why in great detail a few years ago. And, contrary to popular belief, this season's preseason Associated Press poll is generally holding up nicely even if the preseason No. 1 already has three losses.
That's the issue, of course.
Kentucky and its three losses.
Everybody focuses on that and highlights it as proof that "preseason rankings are worthless" without recognizing that Kentucky isn't nearly as bad as most seem to think, and that Kentucky's fall from No. 1 in the preseason to No. 19 this week is a complete outlier when it comes to trying to gauge the quality of the preseason poll. As proof, consider that 17 of the schools ranked in the top 20 of the preseason AP poll are also ranked in the latest AP poll. That's 85 percent of the top 20, for those who aren't good at math.
Now I know what you're thinking.
That's fine, Parrish. But that has a lot to do with voters' reluctance to significantly punish teams that are ranked in the preseason for losing, and it also speaks to the schedules those teams have played. Preseason Top 25 teams aren't often placed in positions to lose early. So of course the overwhelming majority of them are still ranked six Mondays into the season.
To that, I say, you make some fine points.
But let me counter those points with this fact: Eighteen of the schools that were ranked in the top 20 of the preseason AP poll are, at this moment, ranked in the top 25 at KenPom.com, which provides an unbiased rating of teams that's rooted in data and takes schedules into account. In other words, if the AP voters thought you were a top 20 team in the preseason, there's a 90 percent chance you are currently performing like one of the nation's top 25 teams, according to KenPom.com. That's remarkable given that there are 351 schools that could theoretically be placed in a preseason poll, and it's undeniable proof, I think, that preseason rankings are worth way more than some simpletons try to insist.
So which two preseason top 20 teams are significantly underperforming?
Again, statistically, one of them is not Kentucky.
Statistically it's VCU and Marquette -- a pair of schools that were ranked 14th (VCU) and 17th (Marquette) in the preseason AP poll but have combined to lose seven games already and are now 39th (Marquette) and 40th (VCU) at KenPom.com. So what's going on with the Rams and Golden Eagles? A few things, honestly. But one obvious common denominator is that both programs are working with first-year starting point guards in the absence of Darius Theus (VCU) and Junior Cadougan (Marquette). Combine that fact with the fact that the other schools that (are still solid but) have undeniably underachieved relative to their preseason expectations -- Kentucky, Kansas and Michigan come to mind -- are also operating with first-year starting point guards, and there might be a lesson here worth learning.
What lesson, you ask?
That it's hard to be great early with unreliable play from a first-year starting point guard.
Which brings me to the top of the current AP poll.
Have you looked at it lately?
The top four schools are Arizona, Syracuse, <span data-shortcode= State" data-canon="Ohio Bobcats" data-type="SPORTS_OBJECT_TEAM" id="shortcode0"> and Wisconsin, and what do those schools have in common besides great tradition and top-tier coaches? Answer: Consistent and reliable point guards -- namely T.J. McConnell (Arizona), Tyler Ennis (Syracuse), Aaron Craft (Ohio State) and Traevon Jackson (Wisconsin). And, yes, I realize Ennis is a freshman just like the starting point guards at Kentucky, Kansas and Michigan. But the Canadian has adjusted more smoothly to the Division I level than Andrew Harrison (Kentucky), Frank Mason (Kansas) and Derek Walton (Michigan) have adjusted, so much so that Ennis' coach said this about him after Sunday's win in which Ennis got 21 points and six assists:
"He’s a very, very smart player," Syracuse's Hall of Fame coach Jim Boeheim told SNY.tv's Adam Zagoria. "As a freshman point guard, he’s playing better than anybody I’ve ever had, and I’ve had just a few pretty good freshman point guards. ... He’s really been unbelievable. If he played like a normal freshman, we’d be about 7-3, probably, right now. I thought we were rated too high [in the preseason]. But maybe the rating people are smarter than me."
And, maybe, just maybe, the "rating people" are also smarter than their detractors.
Sure, Twitter is a fine place to insist "preseason rankings are worthless."
But, the data suggests, they're actually worth quite a bit.
Zack Cimini is on a college basketball heater and just locked in a strong pick for Saturda...
Our advanced computer model simulated Saturday's Arizona-Stanford game 10,000 times
'If you want it, you're world champs,' A-Rod told the Spartans, which isn't exactly correc...
The players have the right to appeal the decision within three business days
Young is coming off his worst game as a Sooner and will try and bounce back vs. the Cowboy...
The Wildcats face a surging Stanford team that has risen as a potential Pac-12 contender