What Gonzaga’s Loss Doesn’t Tell Us

When the Gonzaga Bulldogs were eliminated from the NCAA tournament by the Wichita State Shockers last night, I was annoyed. In part, I was frustrated that my bracket was busted — I had Gonzaga reaching the Final Four — but the larger concern was how the Zags’ early exit validated criticisms of their No. 1 seed that I don’t find accurate.

On Selection Sunday, Kenpom.com ranked Gonzaga 4th in the country — exactly the same place the selection committee put the Bulldogs as the last No. 1 seed. Doubters might contend that this is a product of beating up on their weak conference schedule, and naturally ratings that don’t consider margin of victory had the Zags somewhat lower. They were sixth, for example, in RPI. However, there’s not exactly a history of Pomeroy and company overrating Gonzaga. Before this year, the last time the Zags lost to a lower-rated team in the NCAA tournament was 2004, when they were upset by Nevada in a 2-7 matchup. And the last time Gonzaga was this good, 2006, Kenpom.com seemed to underrate the Bulldogs, who finished 41st after losing to Final Four-bound UCLA in a Sweet 16 heartbreaker.

The legitimate criticism of Gonzaga’s No. 1 seed is that the team never was tested against other elite teams. In fact, Kenpom.com now ranks Wichita State as the best team the Bulldogs played all season. But that would be a more plausible explanation if Gonzaga’s tournament run came to an end at the hands of a top-10 foes. The Shockers are rated similarly to many of the tournament-bound teams the Zags did face, including Kansas State (a 4 seed), Oklahoma State (5) Illinois (7) and St. Mary’s (11).

Of course, Gonzaga lost at home to Illinois and at Butler. But when we break down the Bulldogs’ pre-tournament performance by quality of opponent, it becomes clear that their schedule had little to do with their rating. To do so, I switched to Sports-Reference.com’s Simple Rating System. (I don’t have rankings from before the tournament, but I believe SRS had Gonzaga fifth, behind the three three No. 1 seeds and Duke.) Using opponent ranking and adjusting for location, we can come up with how the Zags’ actual scoring differential in any game compared to expectation for an average team. Here’s how that works out against different groups of opponents:

Split                        SRS
--------------------------------
Total rating                20.9
WCC opponents               22.0
Tournament opponents        21.8
Non-WCC Tourney opponents   19.4

Yes, Gonzaga was at its best in conference play, but the Zags were almost equally good against teams that made the NCAA tournament. In part, that reflects how well Gonzaga played in three matchups against a quality St. Mary’s team (an average of +26.5 in three games, which would make the Zags the best team in the country). Even when you take the Gaels out, Gonzaga’s rating against tournament-bound foes is still commensurate with a top-10 team. Basically, there was nothing from the Bulldogs’ regular season that suggested they would struggle like they did during the NCAA tournament.

And struggle Gonzaga did. The Zags’ ratings from their games in Salt Lake City — +3.5 points above average against No. 16 Southern, and +8.2 points above average in the loss to Wichita State — were two of their four worst performances all season, along with the loss to the Illini and a two-point escape at San Diego in WCC play.

Why Gonzaga played so poorly in the NCAA tournament is a different issue. Was the no. 1 seed too much pressure, especially after an unexpected opening-round scare? Was it nothing more than 3-point defense/luck? I’m not sure. But I do know that what happened in Salt Lake City doesn’t disprove that the Zags were one of the nation’s best teams this season.

Advertisements

NIT Projections

With the NIT set to tip off shortly, here’s a statistical look at the probability of each school advancing and ultimately winning the title of the nation’s 69th-best team. This is in the same spirit as Ken Pomeroy’s log5 projections, though it does not actually utilize the log5 model. To incorporate home-court advantage, I use Jeff Sagarin’s predictor ratings, with the Pythagorean method giving a probability of each team winning based on rating plus any appropriate home-court advantage.

Here are the figures:

School         Sagarin   2R     QF     FF   Final  Champ
--------------------------------------------------------
Virginia         84.6   .969   .773   .538   .323   .179
Kentucky         85.0   .782   .642   .437   .288   .164
Baylor           85.0   .922   .764   .405   .266   .152
Iowa             85.0   .898   .606   .290   .179   .102
Southern Miss    81.8   .920   .729   .493   .200   .088
Alabama          81.9   .912   .583   .368   .155   .069
Maryland         82.5   .887   .598   .298   .133   .062
Tennessee        82.2   .831   .371   .183   .077   .035

School         Sagarin   2R     QF     FF   Final  Champ
--------------------------------------------------------
Stanford         82.7   .808   .357   .148   .067   .032
Denver           82.2   .752   .310   .152   .066   .030
BYU              80.5   .694   .422   .183   .065   .025
Providence       80.1   .828   .280   .078   .036   .013
Massachusetts    79.3   .648   .287   .083   .032   .011
Saint Joseph's   80.4   .749   .182   .058   .025   .010
Washington       79.2   .306   .169   .053   .016   .005
Arizona St.      78.7   .608   .130   .043   .017   .005

School         Sagarin   2R     QF     FF   Final  Champ
--------------------------------------------------------
Detroit          79.5   .392   .092   .025   .011   .004
Florida St.      77.3   .632   .161   .047   .011   .003
Ohio             79.1   .248   .076   .020   .006   .002
Stony Brook      79.2   .352   .074   .016   .006   .002
Louisiana Tech   77.6   .368   .097   .030   .008   .002
SFA              77.7   .192   .052   .012   .003   .001
St. John's       77.5   .251   .043   .010   .003   .001
Mercer           76.4   .169   .038   .008   .002   .000

School         Sagarin   2R     QF     FF   Final  Champ
--------------------------------------------------------
Indiana St.      75.7   .102   .033   .004   .001   .000
Charlotte        74.6   .172   .036   .005   .001   .000
Robert Morris    74.2   .218   .042   .004   .001   .000
Long Beach St.   74.1   .078   .015   .002   .000   .000
Niagara          74.1   .113   .015   .002   .000   .000
Charleston       71.6   .080   .012   .001   .000   .000
Northeastern     72.1   .088   .008   .001   .000   .000
Norfolk St.      68.8   .031   .002   .000   .000   .000

A few notable things:

– As compared to the four-letter tournament, seeding is even more important in the NIT because the first three rounds are played at host sites. So Iowa, despite being tied for the best rating, has only the fifth-best chance of reaching the semifinals at Madison Square Garden as a No. 3 seed.

Virginia got the most favorable draw of the top teams. The Cavaliers’ opening-round game against Norfolk State is close to a 1-16 matchup in terms of lopsidedness, and their second-round game should be relatively easy. It’s not until a potential quarterfinal matchup with Iowa that UVa will really be tested.

Baylor also got a tough break. The Bears are the second team with a power rating of 85.0, but happen to be in the same bracket as the third (Kentucky), and would have to play at Rupp Arena if both teams advance to the quarterfinals.

– Sagarin likes UK quite a bit more than Pomeroy, who has the ‘Cats in 41st, and especially BPI (52nd), which attempts to account for Nerlens Noel’s injury. So Kentucky’s chances of winning are surely overstated. In fact, UK isn’t a sure thing to get past tonight’s game at Robert Morris, played on the road because Rupp is busy preparing to host the NCAA tournament.

– The other higher seed we know will be affected by hosting other tournaments (the NCAA women, in this case) is Tennessee. Should the Volunteers beat Mercer, they’ll have to play at the winner of BYU and Washington, which will be a long haul either way. Due to that, and playing in the most wide-open bracket, the Huskies actually got an incredibly favorable draw for a sixth seed. If they can get past Brigham Young on the road, UW has a 1 in 6 shot at reaching Madison Square Garden.

Baylor and Maryland are also hosting the women’s tournament, and it’s unclear whether these teams might have to play on the road in the second and/or third rounds.

– The widest-open bracket is in the Alabama region, where four teams have a 14.8 percent chance or better of getting to the Final Four.

– Lastly, take these projections with a note of caution that the NIT is unlike other tournaments because motivation differs so widely among teams. Every year, we see one of the top seeds sleepwalk at home and stumble to a lesser foe. So I might be understating the variability in the projections.

Did the Huskies Underachieve?

Last night’s overtime loss to Oregon essentially put a bow on the Washington Huskies’ 2012-13 season. The Huskies still have a pretty good shot at an NIT berth, but are unlikely to make the kind of run they did to get to Madison Square Garden last year. They simply aren’t that good, though they can compete with basically anyone outside the country’s elite teams.

Clearly, Washington fans will look back on this season as a disappointment. The Huskies’ 9-9 record in conference play was their worst mark since 2007-08 and out of line for a program that has come to count on trips to the four-letter tournament in March. Whether this year’s UW team came up short depends on how expectations are framed. In the larger scheme of things, these Huskies underachieved. But in the context of a team that lost two stars to the NBA, I don’t think that’s nearly so clear.

Revisiting Preseason Expectations

Join me in traveling all the way back to October. What were the best guesses for what this Washington team might accomplish?

In College Basketball Prospectus 2012-13, I picked the Huskies seventh in the Pac-12 with a prediction of 9-9 in conference play. My conclusion? “The [new] coaches have their work cut out for them to help the Huskies overcome the talent drain and a short rotation. Consider this a rebuilding year for Washington, the most consistent team in the Pac-12 over the last decade.”

This year’s edition of College Basketball Prospectus also featured Dan Hanner’s comprehensive projection system. On the eve of the season, after factoring in late-breaking developments, Hanner’s system projected an 8-10 conference record for the Huskies, putting them seventh in the Pac-12.

I don’t have Ken Pomeroy‘s record projections archived, but I was able to go through his preseason ratings for each team (derived using a simpler method than Hanner’s). Pomeroy rated Washington 70th in the country and seventh in the Pac-12.

The most optimistic assessment came from the Pac-12 media poll, which pegged the Huskies for fifth place.

Taken together, these projections and predictions put UW in a band somewhere in the middle of the Pac-12 with a conference record near .500. And that’s exactly where the Huskies finished. The team’s inability to overcome injuries to beat lesser foes in non-conference play meant Washington went to Las Vegas with no real path to the NCAA tournament save winning the auto-bid, but beyond that the Huskies were who we thought they were.

Too Talented to Fail?

What about Washington’s talent? During the first two days of the Pac-12 tourney, Bill Walton has made a couple of references to the Huskies being as talented as Arizona and UCLA. I love Walton’s work, but this is crazy. Let’s take a look at the top-100 (and top-10, top-25 and top-50) recruits on each Pac-12 roster, per StatSheet.com:

School      10  25  50  100
---------------------------
Arizona      1   4   4   6
UCLA         2   3   3   5
Stanford     0   0   1   4
UW           0   1   1   2
USC          0   0   1   2
Colorado     0   0   1   2
Cal          0   0   0   2
Oregon       0   0   1   2
ASU          0   0   1   1
OSU          0   0   0   1
Utah         0   0   0   1
WSU          0   0   0   0

As unpredictable as recruiting can be at the individual level, on the team level it’s a pretty good measure of pure ability. Other than Oregon, which benefited from the arrival of non-recruit Arsalan Kazemi, this matches up closely with performance in Pac-12 play, if not necessarily the actual standings.

Technically, the Huskies were the only team outside the Arizona-UCLA axis with a top-25 recruit on the roster. Of course, as discussed last week, Abdul Gaddy’s rating is noteworthy for being misleading. Besides Gaddy, UW has one other consensus top-100 recruit on the roster (Scott Suggs, ranked 69th in 2008). That’s fewer top-100 recruits than Stanford and as many as most of the other mid-tier conference schools.

In terms of sheer talent, this Washington roster just doesn’t match up with its predecessors or the conference’s elite team. To the extent UW does have NBA potential on the roster (most notably C.J. Wilcox), it’s more a testament to the coaching staff’s eye for underrated talent than sheer recruiting prowess.

The Real Problem

Ideally, this season should have been the start of a two-year rebuilding process for the Huskies after a four-year run at the top of the conference. Next year’s recruiting class, with top-100 guard Nigel Williams-Goss, will help replenish the team’s talent. But unless Washington can land elite prospect Aaron Gordon, there’s unlikely to be so much quality on the roster to overcome the team’s youth.

A two-year down cycle would be about right for the Huskies. It’s similar, if softer, to the two seasons Washington spent outside the tournament in 2006-07 before rebounding with the arrival of Isaiah Thomas. Unfortunately, last year’s team did underachieve by missing the NCAA tournament despite winning the Pac-12, setting up Husky fans to go into this season already frustrated and putting more pressure on next year’s squad to get back to the tournament.

Whether disappointing or not, this season was undeniably frustrating. I think that’s the nature of a .500 team. Fans of Arizona State, Stanford and USC probably feel the same way. There are going to be good stretches and bad stretches, and it’s easy as a fan — or an analyst — to assume that the good stretches are legitimate and the bad stretches are the outlier. Really, both types of performance are equally valid. That goes double in this year’s Pac-12, where the thin margins between the conference’s best and worst teams meant many games — and basically every one so far during the Pac-12 Tournament — have been decided by the fickle vagaries of the last five minutes. Win a couple of those, as UW did at the start of the conference season, and you feel unbeable. Lose a couple and the sky starts falling.

Washington won one close game in Vegas and lost another, and that’s ultimately about what should have happened.

Senior Day: Don’t Blame Abdul Gaddy for Someone Else’s Mistake

“Whom the gods wish to destroy they first call promising.” – Cyril Connolly

During Abdul Gaddy‘s first two seasons at the University of Washington, no televised game went by without the broadcast crew making mention of the fact that seemed like it was part of his full name: In 2009, when Gaddy was coming out of Tacoma’s Bellarmine Prep, he was rated the No. 2 point guard in the nation behind John Wall.

That note still appears in the media guide, but as it became clear Gaddy was no Wall, those mentions became less frequent. The lofty ranking and the comparison with an NBA-bound star no longer felt like a source of pride but instead a cruel taunt.

There was always a false equivalence at play when Wall was referenced just because the two happened to play the same position. Wall was rated the No. 1 overall prospect in the country and needed only to avoid crashing and burning at Kentucky to be taken with the top pick of the 2010 NBA Draft. Gaddy was actually rated 11th by RSCIHoops.com’s consensus, putting him directly behind washout Tiny Gallon.

Still, Gaddy was a major prospect and the subject of a conference-wide recruiting battle that saw him initially sign with Arizona. I don’t know what the scouts saw when they watched Gaddy play in high school. I saw one of his games, during the annual King Holiday Hoopfest tournament at Hec Edmundson Pavilion, and came away somewhat underwhelmed. Gaddy was the best player on the court, certainly, but not the kind of singular talent the recruiting services suggested. I chalked that up to a mid-level high school game failing to showcase the court vision that was Gaddy’s strength, but when he arrived on campus, Gaddy showed little more in terms of high-level athleticism.

Maybe things would have been different had Gaddy not torn the ACL in his left knee. After all, he was 17 for most of his freshman season, making him the youngest player in the nation and a baby compared to some prep school products. (Teammate Shawn Kemp, Jr., for one, was 20 throughout his entire freshman campaign.) Before the untimely injury, Gaddy’s sophomore campaign was off to a solid start. He was making 55 percent of his twos and 40 percent of his threes, and while those numbers were due to come down against stiffer competition in conference play, Gaddy has never approached those shooting marks again.

For now, let’s stipulate that the scouts were in fact wrong about Gaddy. Here’s the thing: Nobody criticizes them for making a mistake, in part because of the overwhelming recruiting groupthink that makes it difficult if not impossible to single out any individual because of a bad evaluation. (No one had Gaddy ranked higher than 10th or lower than 16th.) More importantly, they’re not the ones out there running the point on a nightly basis. So all the blame has gone to the player. Gaddy has become the symbol for all the frustration Washington fans feel about the past four years, the subject of abuse if not scorn from the people who are supposed to be his fans.

There are few things more important to evaluating players than setting fair expectations. That goes double for amateur athletes, who suffer on-court scrutiny beyond their paygrade. What Gaddy owed the program, and fans, is the same thing any player does — working hard, representing UW well and giving his best effort. Has any of that ever been in question?

Gaddy worked his way back from one of the most devastating injuries an athlete can suffer. To the extent he struggled for reasons within his control, it was precisely because he lost confidence — in no small part because of the relentless criticism. If the goal of fandom is to see your team be as successful as possible, criticizing a player with a fragile psyche is overwhelmingly counter-productive.

Let’s talk about Gaddy’s performance. Here are the career stats for two point guards during the Lorenzo Romar era. Can you tell which of them is Gaddy?

            G    AST    TO   A/TO   PPG   APG   RPG
---------------------------------------------------
Player A   121   515   314   1.64   9.2   4.3   2.8
Player B   114   440   242   1.82   7.6   3.9   2.4

Player B is Gaddy and Player A is Will Conroy, one of the most popular Huskies in program history. Conroy is the better player, to be sure — he ultimately reached the NBA because he was a superior scorer and much better defender whose intangibles remain legendary. But it’s important to keep in mind the differences between the two players’ careers. Conroy arrived on campus as a walk-on, an unheralded recruit during a period when the entire Washington basketball program was an afterthought. By the time both he and the team were good enough to generate any expectations, Conroy was surrounded by talent like former Garfield teammate Brandon Roy, Nate Robinson, Bobby Jones and Tre Simmons. They grew together into a Pac-10 power.

By contrast, Gaddy’s college career has been much more uneven. After his injury, he had to find a way to coexist with the mercurial Tony Wroten as a junior. The departure of Wroten and Terrence Ross for the NBA last summer has forced Gaddy into the uncomfortable position of being the Huskies’ primary creator on offense. It’s also given him no choice but to be a leader, a role which he’s slowly embraced over the second half of this season. In another scenario, the areas in which Gaddy is not Conroy’s equal may never have been nearly so important.

On Saturday, Gaddy will be one of three seniors honored for their service to the University of Washington. Depending on where the Huskies land for postseason play, it might be the last time he takes the court at Hec Ed. I hope we don’t hear anything about John Wall, but instead about how Gaddy ranks third in school history in assists — and still has a chance to surpass Chester Dorsey for second by the end of the season. I hope Gaddy gets an enormous ovation. And I hope he finds some matter of satisfaction at the end of a career that should not be judged by his recruiting ranking.