We're far enough into the season now where disagreements between pollsters on the one hand and computer rating systems on the other no longer have to be settled with just: "Well, you wait and see." Now we can start to look at actual performance, and if you've been reading along with me for a while you know I especially like to draw on the lessons offered by conference play.
Take Kansas State. Classic pollsters vs. computers dispute, right? The Wildcats are No. 18 in the polls, but both the BPI and KenPom have K-State ranked in the low 40s nationally. Who's right?
I asked Bruce Weber that very question this week, and, yes, he came down firmly on the side of the pollsters. "I think we're a top-25 team," Weber told me by phone. "I'm not sure we're as high as 11," he added quickly, referring to his team's lofty standing before it lost at home to Kansas and suffered a defeat on the road at the hands of Iowa State. "But we've shown we can beat Florida, and we played Kansas down to the wire. It's not like Michigan blew us out. We just have to play consistently."
Weber forthrightly asked me why his team appears in a different light when viewed through the lens of computer ranking systems. ("Why do computers have us that low?"). I suggested to him it may be because computers think this is not a particularly good offense. "We've made some progress offensively led by Rodney McGrduer," Weber responded. "He's starting to get a feel for how to get open and move without the ball."
You can write off Weber's opinion as biased, of course. (Though it occurs to me certain coaches would likely answer this question in the affirmative and say their teams really are being overrated. Doing so could constitute a motivational ploy in some quarters.) But in this instance I think the coach may actually be on to something.
This might be one of those cases where the narrative about players needing some time to adapt to the new coach's system really does have value. Weber's in the middle of his first season in Manhattan, and with six Big 12 games now in the books the Wildcats have actually been very good on offense. K-State has recorded 1.05 points per possession against their league opponents, a level of offensive efficiency second only to that displayed by Iowa State.
This upward trend on offense is captured well by Kansas State's featured scorer, McGruder. On the season as a whole the 6-4 senior looks pretty good, but in the past six games the Wildcats' star has taken his game to a new level, making 57 percent of his twos and 44 percent of his threes. Even more impressively, McGruder's doing this while taking 28 percent of K-State's shots during his minutes. (And you can largely dispense with the "during his minutes" part. By my count McGruder's been on the floor for all but 28 of Kansas State's 376 Big 12 possessions.)
With a featured scorer of McGruder's caliber, and a team-wide 51 percent accuracy rate inside the arc in conference play, Kansas State's offense has definitely answered the skeptics to this point in the season. Don't sell your Wildcats' stock just yet.
Computers or pollsters? Pollsters. After all, I did have Kansas State at No. 24 on my ballot in this week's Power Rankings. My vote says that as of Monday I thought No. 18 was closer to the truth than the low 40s. I still think that.
Butler is baffling
AP ranking: 9 BPI: 33 KenPom: 28
First I want to offer a note of sympathy to pollsters, computers, or anyone or anything else trying to make some evaluative sense of Brad Stevens' team. Just to touch a few bases here: Butler beat Indiana, Marquette, and North Carolina on neutral floors, and prevailed at home over Gonzaga in rather memorable fashion. That's the impressive portion of the Bulldogs' resume, and it's very impressive indeed.
This is the same team, however, that also lost by 15 at Xavier (the Musketeers have dropped home games this season to Vanderbilt and Wofford), and came up on the short end of a 54-53 decision just last week at La Salle (the Crusaders lost at home earlier this season to Central Connecticut). True, Rotnei Clarke, Stevens' leading scorer, missed the La Salle game, so maybe we can give the Bulldogs a partial pardon there. Still, isn't No. 9 a little high for this team?
Right now, yes, it probably is. Butler has outscored five Atlantic 10 opponents by 0.12 points per possession. That's very good, and at the end of the season we may find that the Bulldogs are the best team in their 16-team league. But at the moment I'm a little concerned that BU is actually operating at a turnover deficit relative to their conference opponents. The Bulldogs take very good care of the ball (sporting a 17.4 percent turnover rate in-conference), but when the other team commits a turnover on just 13.8 percent of their possessions, you're going to get fewer chances to score relative to your competition.
Computers or pollsters? Computers. Butler is good, and Stevens' team has improved dramatically since last season. But a top-10 ranking is just too high. For my part I have the Bulldogs at No. 17.
Pittsburgh is perplexing
AP ranking: Unranked BPI: 14 KenPom: 6
At the risk of oversimplifying matters, computers have loved the Panthers all season long, but Jamie Dixon's team is still seeking that proverbial signature win. Right now their best resume bullet would be their 73-45 win at Georgetown. (The Hoyas are currently perched just outside the top 25.) Then again Pitt will have an excellent opportunity to change this state of affairs when they host Syracuse this Saturday.
Pollsters see a 17-5 team that (all together now) "hasn't beaten anybody," but computers see a team that's excellent on both offense and defense. And if nothing else, as always in Pittsburgh, the Panthers can crash the offensive glass. Talib Zanna, Steven Adams, J.J. Moore, and Dante Taylor have combined to haul down 39 percent of Pitt's missed shots in Big East play.
I have the Panthers listed as "unranked" in the AP poll, but if you look at the poll's small print ("Also receiving votes"), you can find Pitt down there around No. 32 or so. I guess that means Dixon's team is at least on voters' radar, but being buried in the small type is no way to treat a group that's outscoring the still-strong Big East by 0.12 points per possession.
Computers or pollsters? Computers. I have Pitt as a top-20 team right now, and I wouldn't be surprised to see them climb higher.
A version of this article originally appeared at ESPN Insider .
John Gasaway is an author of Basketball Prospectus.
You can contact John by clicking here or click here to see John's other articles.