Tim in black; John in red.
As we approach Selection Sunday, we’re inundated with various criteria to whittle down to the most deserving 65 teams.* There’s the record, RPI, records against the RPI top 50, strength of schedule, and of course, the “Eye Test.”
*It is NOT the best 65; it is the most deserving 65.
I hate the Eye Test.
The Eye Test works neither theoretically nor practically. In what other aspects of life is the Eye Test appropriate?
You may not have made the right diagnosis, but you looked like you knew what you were doing. Congratulations, Doc, you’ve passed my Eye Test.
The Eye Test is college basketball’s equivalent of Josh’s beloved Handwriting Effect. It’s a way to impose overly subjective measures into a process that should limit subjectivity as much as possible. And while the Selection Committee is subjective, it’s subjective within objective boundaries. By this I mean the Committee applies a subjective weight to objective measures. They may prioritize overall and conference records differently, but those records are set. It’s not like someone’s saying, “Well, Seton Hall looks more like a 22-9 team instead of 19-12 in my book.”
The Eye Test is what leads to drafting Darko Milicic and Marvin Williams second overall in basketball and the overemphasis on the Scouting Combine in football.* It’s another name for the old guard in baseball, the pre-Moneyball emphasis on in-person, small-sample scouting instead of long-term statistical analysis. If I used the Eye Test for this college basketball season, Georgetown would be in the top five and J.P. Prince would be an All-American.**
*The NFL’s Scouting Combine is the dumbest thing in professional sports.
**I have never seen J.P. Prince play a bad game. A 6-foot-4 lefty point guard that can handle, pass, and finish in traffic? How was he not one of 16 players selected for SEC first- or second-team?
My biggest issue with the Eye Test, however, is in its practice. You can’t see enough games of the teams in question to legitimately claim one is better than the other on looks. Most Eye Test judgments—“This team just doesn’t look like an NCAA team”—are based on 1-2 game samples. The same goes for so many of these bubble teams because inconsistency is an inherent trait for teams on the bubble. Georgia Tech, Illinois, Memphis, UAB, the Mississippis—all these bubble teams have been wildly inconsistent. And if you base a judgment on the Eye Test, it comes down to whether you saw them on a good day or a bad day. If you watched South Florida play Georgetown the first time but not yesterday, they pass the Eye Test. If you watched the Bulls yesterday against the Hoyas but not the first time, they fail it. If you saw them both times, your results are inconclusive.*
*Something else that grinds my gears: Last night, Joe Lunardi said Seton Hall did not pass the Eye Test in sneaking by Providence and losing to Notre Dame. This was not an instance of the Eye Test. Joe himself could not have possibly seen Seton Hall’s game with Notre Dame; he was covering Robert Morris and Quinnipiac at the same time. ESPN made a point of showing him doing this. The Eye Test does not have anything to do with margin of victory/defeat.
The other issue is that the Eye Test favors certain styles of basketball, certain programs, and certain conferences. Let’s look at last season, when the Eye Test was likely the defining factor in determining the final at-large berth last season, which came down to Arizona, Penn State, and Saint Mary’s. The Wildcats clearly had more talent with two first-round draft picks on their roster. The Gaels were on the other end of that spectrum; even with the talented Patty Mills, Saint Mary’s played a much slower, more defensive, less athletic style.* Arizona looked like a good basketball team even if it didn’t often play like one. Saint Mary’s—a team that got the most out of its talent—won a lot of games ugly. It also happened to lose one game very ugly: Its 25-point loss in the WCC Tournament Final against Gonzaga stuck in a lot of people’s minds—probably because it was one of the few times they had seen the Gaels play. Furthermore, Saint Mary’s was up against a power-conference team** that had made the Tournament 24 consecutive years. It’s thus a little easier to say, “Sure, they’re more inconsistent than usual, but it’s still Arizona.” Saint Mary’s, on the other hand, is a relatively unknown quantity.
*Penn State lies somewhere on the middle of this spectrum.
**The PAC-10 was once considered a power conference.
Back in 2006, the Committee even admitted that one of the main reasons it included the worst-ever at-large team, Air Force, was that the Falcons passed the Eye Test. It didn’t matter that they hadn’t beaten a single RPI top-50 team; they looked like an NCAA team. Another team the Committee chose that year, Utah State, also gained entry because of the Eye Test—and not for the whole season, but for its performance in the WAC Championship against Nevada (frighteningly overseeded that season as a 5) the night before Selection Sunday. Committee chair Craig Littlepage admitted these things the next day.
Isn’t this reprehensible?
First, a brief defense of the Eye Test, or at least how the Eye Test has been used by the Committee. The Committee’s goal is to determine which teams deserve to play in the NCAA Tournament. Therefore, it considers a team’s entire body of work. But certainly it ought to weigh certain things more than others. The Saint Mary’s team of last year, for example, was clearly in a different situation at the end of the year, with their best player in Patty Mills coming off an injury, than they were earlier. So going by the “Eye Test” in that situation—when it’s an attempt to gauge how a team will play without its best player at full strength—makes sense. Similarly, I think you can go by the Eye Test if a team has, for any real reason, been playing significantly worse, or better, in the days and weeks before the Tournament, as the team did with Utah State. I don’t think it’s totally justified every time, but it does make some sense to place emphasis on how a team has “looked” recently.
Having Said That, the Eye Test is, more often than not, a way to reaffirm preconceptions and reward teams for underperforming. There’s all kinds of research on this phenomenon: When you look at something with a preconceived notion, you are far more likely to see things that uphold those notions than things that disprove them. And people who watch these teams play—particularly “experts” like TV analysts and the people who sit on the Committee—are bound to have preconceived notions about this team. So there is naturally a bias towards teams that are typically in the Tournament—like Arizona—and teams with a lot of talent.
The most disturbing thing about the Eye Test is how it excuses bias under the guise of subjectivity or, worse, expertise. If someone on ESPN or on the Committee says that a team “just doesn’t look like a Tournament team,” then they’re really making an unsubstantiated claim, but one that carries more weight than it should because of the person’s position. This is not the same as subjectivity, as you hinted at. If I think a team’s RPI is inflated because it doesn’t take into account the fact that the team’s biggest games have all come at home, or that a big win came when their opponent wasn’t playing well, then that’s a legitimate opinion. It may or may not be convincing, but at least it’s a reasonable position. Saying a team passes the Eye Test, though, has no supporting arguments—it’s stated almost as a matter of taste. And choosing who gets into the field of 65 shouldn’t be left to someone’s taste.
Oh, and note to Etan Kaplan of Bleacher Report: Blind resumes are not the Eye Test.*
*A little bit of a P.S. I just listened to Doug Gottlieb’s excellent weekly podcast on ESPN, and he made a very similar point about the Eye Test.
Is it possible to have a more exciting season than Marquette this year? Twelve of the Golden Eagles’ 19 Big East games were decided by three points or fewer. They’re 1-2 in one-point games, 3-2 in two-point games, and 3-1 in three-point games. Four of those went to overtime. Expect their game this afternoon with Villanova to be another nailbiter: In their two prior meetings, the Wildcats beat Marquette by two points each time.
Speaking of, Lazar Hayward is my off-the-board guess at a college guy who sticks in the NBA as an off-the-bench three-point shooter and defender. Of course, I said much the same about Marquette’s Jerel McNeal last year. (I totally should have gone with Wes Matthews.)
And I guess we can officially say that Notre Dame is NOT better without Luke Harangody. He finished with 20 points and 10 rebounds in his return as the Fighting Irish beat Seton Hall by a dozen.
The New York Times had a pretty condemnatory article on Seton Hall coach Bobby Gonzalez this week. Citing several former players and administrators at Manhattan, Kevin Armstrong and Pete Thamel paint Gonzalez as a borderline insane coach bent on winning at all costs. Thamel, of course, was behind a similar story last year uncovering that kind of mindset at Binghamton—a school that made the NCAA Tournament for the first time in its history last season only to see several of its players arrested, its coach fired, and the program withdraw from its conference tournament this year. Could Seton Hall be on the same path with Gonzalez, proving more and more that firing Louis Orr after two Tourney appearances in five years was a massive mistake.
Firing Norm Roberts at Saint John’s would also be a mistake. Roberts has shown he can coach that team; it’s a matter of recruiting there, which requires some more time for Norm to build up the brand following the scandals at the end of Mike Jarvis’ time.
We finally got some conference tournament magic from Montana’s Anthony Johnson last night. Johnson did his best LeBron James/Wally Szczerbiak, scoring 42 of the Grizzlies’ 66 points. He had eight of their 20 at halftime (when the Griz trailed 40-20), and then 34 of their 46 after intermission, including the last 21. Anybody want to throw a double/triple/quadruple team at him?
It was the most dominant half of basketball by one player I’ve seen since last season, when Tyrese Rice scored 34 in the first half against UNC. BC still lost.
I think it’s about time we officially call Doris Burke one of the two or three best analysts in college basketball. She brings the same insight as Bobby Knight without being repetitive. The trio of Dave Pasch, Fran Frachilla, and Burke this week in the Big East Tournament has been almost as good as McDonough, Raftery, and Bilas—and that’s the highest of praise from me.
Speaking of women, did you see Marion Jones signed with a WNBA team? And even more shocking is that the team (the Tulsa Shock) is coached by NOLAN RICHARDSON! What the hell? That guy won a National Championship and he can’t get a better job than the WNBA? Even Isiah Thomas has a job coaching men’s basketball. I know Richardson got into some hot water with Arkansas, but you’d think somebody would give him another shot.
72 in a row! Yeah!
Oh wait, you’re talking about a different women’s story?
(Come on, John, MJ played at UNC when they won the national title in 1994* and has her jersey hanging in the rafters of Carmichael Auditorium. Catch up.)
*This was arguably the greatest women’s college basketball championship game of all-time, as Charlotte Smith took an inbounds pass with 0.7 left and the Heels down two and nailed a buzzer-beating game-winning three. Someway, somehow, I saw this live.
(And they’re keeping the name “Shock”? Bad move in Tulsa, where they should join up with their brethren in OKC and go by the Thunder.)
As usual, the conference tournaments carry some interesting implications for seeding and at-large bids, in addition to the automatic bids that get handed out. In the Big Ten Tournament, for example, Illinois probably needs to make some noise to get in. More importantly, though, are the implications in that conference for the last remaining #1 seed. Tim mentioned that Purdue could steal it if they win and Duke loses in the ACC, but let’s say Ohio State wins it, beating Wisconsin and Purdue in the process, or Michigan State takes the title with Ws against Purdue and Ohio State. Could either of those seven-loss schools steal a #1 on the strength of those RPI Top 50 wins?
OSU has an outside shot; wins over Wisconsin and Purdue/Michigan State would put them at 7-4 against the RPI top 50 and 10-6 v. the top 100. Duke is currently 8-4 and 15-4 by those metrics and would have to lose another in your scenario. I think the only real chance the Buckeyes have to swoop in as a No. 1 is if Duke loses to BC or Virginia on Friday, Syracuse dominates the Big East, and Ohio State wins its games convincingly.
Michigan State, with only three top-50 wins right now, has no shot. If the Spartans don’t beat Purdue, they may not get a protected seed.
Games of the Week: Saturday is that glorious day where ESPN and ESPN2 stagger their games, so you can watch second half after second half all day. It will be awesome.
Upset of the Week: Syracuse will not win the Big East Tourney.
I’m sorry, is that really an upset? Cuse may be the top seed, but they may have to beat Georgetown, Villanova, and Pitt/West Virginia. Some Final Four teams won’t face a stretch that tough. Way to go out on a limb and say they won’t win all three of those games. I’m picking Georgia Tech to make it to the ACC Championship game, upsetting Maryland and Florida State in the process.
In the week’s most exciting news, get pumped for extra editions of The Double Bonus as we head into the NCAA Tournament. I’ll have my final bracket posted before the Selection Show on Sunday, and we’ll have our first-round picks to you by Wednesday–or just in time to screw up your bracket. Also be on the lookout for the first-ever NPI podcast breaking down the entire Tournament on Monday afternoon. I plan to start it with a long, caustic rant chastising the Committee for their many errors on Sunday.