Friday, September 23, 2005
Steal of a Pick?
Danny Granger of New Mexico was my favorite player that I never watched last year. That's right - I didn't see him play a single minute, but I'm still fascinated with his game. Lucky for me, several scorekeepers and statisticians did see Granger play and they kept a log of his performance. The picture they paint is certainly an entertaining one.
I'll throw a couple stat lines out so you can see what I mean. First are the standard per game averages.
If you were picking a Mountain West Conference fantasy basketball team last year, I think you could make a good argument for taking Granger ahead of Andrew Bogut. Granger's production was just solid across the board. That's even cooler from an efficiency standpoint because he compiled those numbers in only 30 mpg for a slower-paced team. Ken Pomeroy's stats page estimates that New Mexico's possessions per 40 minutes ranked 211th out of 330 D-I teams last year.
Next are the numbers that I spend a lot of time with, and are the reasons why Granger impresses me so much.
Explanations for these stats can be found in the Stats Primer. Steal % is one I haven't used much - it estimates the percentage of a team's possessions that a player gets a steal on, while he's in the game. Over 3 % seems good, and being over 4 % is pretty select company, though guys like Eddie Basden and Rajon Rondo were in the 5-6 % range last year.
Granger was the primary offensive weapon for a New Mexico team that won the MWC tournament and went 2-1 against Utah (Granger missed the game they lost) and came close to erasing a huge first half deficit before losing to Villanova in the first round of the NCAA tourny. His possession usage was about as high as you'll see in the NCAA, but he still managed to shoot very well, including 43% from downtown, without coughing the ball up much. His rebounding was also excellent, and included a 24.7% rate at the defensive end, which means he rebounded 1 in 4 of his opponents' missed shots when he was on the court.
What really grabs my attention, though, is that Granger averaged over 2 steals and 2 blocks per game. I don't know how accurate this is, but it seems to me that college players who average over a steal and a block per game generally turn out to be decent defenders when they go pro. I guess it suggests that they possess the jumping, timing, quickness, etc to make it in the NBA. In short, they're great athletes.
Dan Rosenbaum has done some interesting research on individual defense in the NBA, and in the past he stated that the effects of steals and blocks (and players who accumulate them) tend to be underrated. In one paper, "Measuring How NBA Players Help Their Teams Win," he wrote -
Perhaps more importantly, with these adjusted plus/minus ratings I am able toSteals - check. Blocks - check. Avoid turnovers - check. Do it all - check.
estimate what game statistics predict better performance on the court; these
results help explain why certain players have such high adjusted plus/minus
ratings. It appears that rebounds are less valuable than typically assumed
and steals, blocks, and avoiding turnovers are more valuable. It also
appears that having three point shooters on the floor helps teams and that
players that can do it all – score, rebound, and assist – are more valuable than
simply the sum of those game statistics.
I found it interesting how much New Mexico missed Granger when he was hurt for three games. Against Air Force, BYU, and Utah, the Lobos allowed 114 points per 100 possessions (PPP). When they later played the same three teams with Granger in the lineup, they allowed only 100 PPP. It's clearly a small sample size to draw conclusions from, but Granger appeared to have a big effect on the defense.
Before you say that Granger's numbers need to be downgraded because he played in the MWC, I should point out that the conference was #7 in Ken's rankings last year, and each of the eight teams finished in the top 150. The MWC might not be elite, but it's certainly not a mid-major, either.
To wrap up, I want to compare Granger to a similar player who made the jump to the NBA from the same conference (technically it was the WAC back then), and has made a name for himself on a great team.
|Player X, Jr||20||32.9||22.8||11.3||3.1||2.3||55.3|
Player X spent his first two years at a junior college before transferring to UNLV for the 1998-99 season. Granger played a couple seasons at Bradley (MVC) before heading to New Mexico.
Think you got it?
* * *
* * *
* * *
I can't fool you - Player X is Shawn Marion, a starter for the Phoenix Suns. He averaged 19 pts, 11 reb, 2 stl and 1.5 blk per game on a team that went 62-20 last year. It's interesting that neither player shot many threes as a junior. Marion went 20-67 and Granger was 24-72. When he got to the NBA, however, Marion developed a nice outside shot, and has hit well over 1 three per game for the last three seasons. Granger is already starting to show a nice jumper, as he hit 45 of 104 (43%) last year.
Marion was already pretty good his rookie season with Phoenix, but he didn't get starter-minutes until his second year in the league. Granger stayed an extra year in college, and improved nicely on his junior season, so perhaps he'll be ready to break out for the Indiana Pacers, who took him 17th in this summer's draft. If his career turns out anything like Marion's, he'll be the steal of the draft. Check back in two years.
Update - Now I have seen Granger play. New Mexico's website has a sweet highlight video from last year. [Granger is #33. Or you could just look for the guy dunking on people's heads, nailing contested threes, and swatting shots off the backboard.] Why don't more schools do this? Tell your local SID to add one to your school's website!
Thursday, September 22, 2005
The Overhyped Club Is Now Seeking Members
You know what really grinds my gears, America? Players and teams that get way more hype than they deserve. Any discussion of that topic usually centers on Red Sox and Yankees coverage, but college basketball is certainly not immune. Dick Vitale's passion for the ACC is borderline obsessive, and Duke is always a lock to have an ESPN game every week.
My current target for the Overhyped Club is Syracuse's Gerry McNamara. This otherwise cool photo series by SI.com that listed him as the country's #4 point guard probably set me off, but it's been building for a while. Remember the big debate last year about whether Duke's J.J. Redick or Arizona's Salim Stoudamire was the better shooter, and how Dickie V kept trying to include McNamara in the discussion? I was as baffled as anyone.
Don't get me wrong, McNamara is good, and he's certainly had his moments. He really made a name for himself when he hit the six threes to help Syracuse win the NCAA title his freshman year, and that 40+ point tournament game as a sophomore didn't hurt his reputation, either. He's just not great, or at least not as great as so many people seem to think.
Yoni Cohen, whose site is the center of the college basketball blogging universe and whose writing now appears at Fox Sports, recently called McNamara "as a good a shooter as you'll find north of Durham, North Carolina." Dickie V thought he was one of the three best shooters in the country last year. And SI tells us he's the fourth best PG in college basketball? I'm not buying it.
First off, here's a quick comparison of Redick, Stoudamire, and McNamara from last season.
|Player||O Rtg||%Poss||3 pt%||3A / FGA||eFG%|
There's a little unfairness in comparing McNamara to Redick, since Gerry handled the ball a lot more, and the resulting TOs brought down his Offensive Rating, but I think the shooting numbers speak for themselves. His effective FG% (traditional FG% adjusted for the extra points from threes) is very pedestrian. Redick's was pretty plain, too, but his knack for drawing a ton of fouls (and shooting 94% at the line) created a very nice offensive rating. For my money, Stoudamire was the best shooter anywhere last year, and one of the most exciting players to watch, too.
So what justification does SI have for ranking McNamara at number 4 in their point guard list? They say
"McNamara started all 34 games a year ago and all 100 in his career. He averagedThey also mention that he was all-Big East, but that's subjective anyway so I'll toss that aside for now.
15.8 points per game last season and had 168 assists, by far the most on the
team. He's the Orange's all-time leader in 3-point attempts with 823 and
3-pointers made with 297."
Can you think of a shooting point guard who (1) started all of his team's games last year, (2) had a decent scoring average, (3) led his team in assists, and (4) will soon break his school's records for threes made and attempted?
Hmmmmmmm. *Deep thinking...........*
1 - started
2 - 14.0 ppg on the season, but he averaged 14.8 in the 13 games after Pierce left
3 - passed his way to 180 assists
4 - is not far from breaking Chris Kingsbury's old shooting records
I'm not going to suggest that Horner is better than McNamara, or even that he's quite on the same level. But I don't think they're that far apart. Consider this........
Some quick notes - that "Pass" rating is a very rough estimate of a player's points produced through assists, per 100 possessions. It can fluctuate considerably when players are put in different roles, but Horner and McNamara played similar roles for their teams, so I threw it in. "FT/FG" is actually free throw attempts per field goal attempt, which is one way of measuring how often a player gets to the line.
I find it quite interesting that McNamara has never broken 40% from downtown, despite his reputation as a sharp shooter. His eFG% has been very normal, too. The national average is something around 50%. If there's one thing you can say for him, though, he has been very consistent. He had a very good year for a freshman, but he hasn't progressed much since then. He's more or less replicated the same production each year.
Horner, like McNamara, has played a ton of minutes ever since his freshmen year (90% would be 36 min in a regulation game). He had a rough going his freshman year, but since then he's found his shooting touch and cut down his turnovers each season. He also grabs considerably more rebounds than McNamara, both offensively and defensively.
One big difference you might have noticed is McNamara's edge in possession usage. Horner has been as low as Iowa's fourth scoring option, with Pierce, Brunner, and Haluska all available to take shots, while McNamara has been Syracuse's second-leading scorer behind Hakim Warrick for the past two years. It's generally considered tougher to maintain a higher efficiency level when using more possessions, so you have to give Gerry some credit there.
Can Horner be effective when using a higher percentage of possesions. I think so - we got a little taste of that after Pierce was kicked off the team last year. In the 12 games against Big Ten opponents sans Pierre, Horner used 24.1% of the possessions with an offensive rating of 108. Pretty decent. I think that'll look even better this year when it's balanced out over a full season schedule that includes some OOC cupcakes.
So all in all, I'll give McNamara the slightest of edges for last season, but if I had to choose a floor general for this year, I'll stay loyal to Horner for sure. And if you're going to rank McNamara as the fourth best point guard in the country, at least give our guy Jeff some recognition too.
Wednesday, September 21, 2005
Last week I started speculating about the importance of having tough post players to anchor a team's defense. It seemed to me that last season's leaders in overall defense (ranked by points allowed per possession) had imposing big men who were tall and could block shots and/or rebound well. No, it's surely not rocket science to those of you who have watched basketball for a few years - good defensive centers stop easy baskets near the hoop, which makes it hell for the offense to score points (see: Iowa vs Cincinnati last year).
Basketball minds like Dean Oliver and Dan Rosenbaum have agreed with the theory, at least as a general principle. In a recent article on individual defense in the NBA, Rosenbaum concluded -
Defensive ratings, on average, are highest for centers, then power forwards,If post players are generally the most important players to the defense, I want to explore some of the better frontcourt defenders from last year, and get an idea for which teams will be strong this year. There are of course no publicly available plus/minus data available for college teams, so Rosenbaum's technique is out (not that I'd go that in-depth). Instead, I'll take a far simpler approach. I'm just going to look at how well each team prevented its opponents from shooting on 2-point field goals, relative to each team's conference.
then small forwards, then shooting guards, and then point guards.
Offensive ratings go the other way. This squares with the conventional
wisdom that defense is anchored by big guys.
Clearly, this is no exact measure of the strength of a team's defensive frontcourt. Perimeter defenders can prevent and allow 2-point baskets as well. Not every 2-point attempt comes near the lane where a center might affect the shot. Likewise, some teams allow a lot of fast break points that are contested by no one.
In general, though, having good post players should lower the 2-point % a team allows, just as poor players would raise it. With zone defenses allowed, centers and big forwards spend a lot of their time on defense near the basket. They guard opposing centers when they post up, they can prevent the other center from getting good enough position for an entry pass in the first place, they help out when perimeter defenders get beat, and they challenge the shots of opposing players who grab offensive rebounds. Also, they prevent opponents from getting easy putbacks by boxing out and grabbing defensive rebounds.
For this little project I recorded each major conference's shooting percentage on 2-point field goals, then compared each team's 2-pt% allowed to their conference average (conference games only). The following table lists each conference's 2 -pt%, its leader in 2-pt% defense, and any other team that held its opponents at least 5% below the conference average.
Note - I often like to restrict these data sets to conference games so I can feel confident when comparing teams within each conference. When comparing teams from different conferences, be aware that not all are equal. Holding opponents to 40% shooting is probably easier if you play in the MAC than in the ACC.
|Conference||2-pt% Avg||Team||2-pt% Def||Vs Avg|
Does this pass the laugh test? I'd say so. Each of these defenses was anchored by a center or power forward (or both) with a national reputation for defense. There's at least 12-15 current/future NBA players who manned the post for those squads.
The more I watch basketball and follow the stats, the more I agree with Dean Oliver, author of Basketball on Paper, that basketball isn't composed of offense, defense and rebounding, but rather just offense and defense. He wrote that offensive and defensive rebounding are two different skills, and I agree. An offensive rebound is a contribution to the offense because it keeps a possesion alive, while a defensive rebound prevents scoring by ending a possession. Dean's idea became a little clearer to me when I realized that most team's that keep other teams' 2-pt% low also have good defensive rebounders. Additionally, post players make opponents miss by simply blocking their shots.
Here's some key contributors for each team, including their playing time, blocks per 40 minutes played, and defensive rebounding percentage. Asterisks (*) denote returning players.
Note - If the average team rebounds about 2/3 of the other team's missed shots, then each of the five players on that team would have a defensive rebound % of 13.4, if they rebounded equally.
James Augustine*, 26.6 mpg, 1.8 bp40, 22.4 dRb%
Big Ten fans hoping for Illinois to disappear this year will likely be disappointed. Not only were the Illini the conference's best offense last year, they were also its second best defense, allowing only 95 PPP (points per 100 possessions). Augustine was overshadowed by the NBA-quality guards around him, but his presence in the middle solidified a tough defense. Illinois won't repeat last year's offensive showcase, but the return of Augustine and last year's Big Ten Defensive Player of the Year Dee Brown will make scoring difficult for anyone else.
Sean May, 26.8 mpg, 1.5 bp40, 26.3 dRb%
Marvin Williams, 22.2 mpg, 0.9 bp40, 21.8 dRb%
Building a college defense around two NBA lottery picks sounds like the way to go. (By the way, if was a GM with a high pick, there's no way Charlie Villanueva comes before May. He was simply a phenomenal rebounder last year.) It's hard to predict UNC's defense for next year with all the new faces, obviously.
Shelden Williams*, 33.6 mpg, 4.4 bp40, 22.1 dRb%
Shavlik Randolph, 18.9 mpg, 3.2 bp40
It's a testament to Williams's defensive ability that Duke's defense was this good, considering that J.J. Redick finished second on the team in total defensive rebounds. Adding the 6-10 McD's All-American Josh McRoberts should help considerably. Duke could easily be the nation's toughest defense, even with two freshmen seeing big minutes.
Josh Boone*, 29.5 mpg, 3.9 bp40, 15.0 dRb%
Charlie Villanueva, 25.8 mpg, 2.9 bp40, 20.3 dRb%
Rudy Gay*, 28.8 mpg, 2.6 bp40
Hilton Armstrong*, 12.4 mpg, 3.9 bp40
Hmm, I had no idea Gay had so many blocks last year. Look for him, Boone, and Armstrong to put the Huskies among the country's leaders in blocks once again.
Brad Buckman*, 25.5 mpg, 2.5 bp40, 22.4 dRb%
Lamarcus Aldridge*, 22.2 mpg, 2.7 bp40, 15.4 dRb%
P.J. Tucker*, 29.4 mpg, 16.3 dRb%
Jason Klotz, 28.9 mpg, 1.5 bp40
Luke Winn aptly pointed out how important Aldridge and Tucker were to the Longhorns last year. Lining them up beside Buckman will make Texas one of the country's toughest defenses, as well as one of its better rebounding teams.
Wayne Simien, 34.3 mpg, 23.8 dRb%
Simien's injury history probably knocked him down a few pegs in the draft, but he was still one of the NCAA's best rebounders and a very efficient scorer. Did he really get picked after Linas Kleiza?
Jason Maxiell, 31.4 mpg, 3.5 bp40, 13.9 dRb%
Eric Hicks*, 31.4 mpg, 2.9 bp40, 16.6 dRb%
I doubt one of these guys would seem all that tough alone, but in tandem they make for a ferocious defense, as Iowa found out in the first round of the tournament. It's interesting that neither was a big defensive rebounder, but they each blocked a lot of shots while committing relatively few fouls, which is probably an underrated skill. In fact, each their blocks/foul ratios are probably among the highest that I've seen, along with Shelden Williams and Josh Boone.
Duane Erwin, 26.5 mpg, 2.3 bp40, 16.6 dRb%
Joey Dorsey*, 15.1 mpg, 3.1 bp40, 24.5 dRb%
Rodney Carney*, 29.8 mpg, 1.0 bp40
Memphis seems to be the hot pick to climb into next year's top ten. A defense built around Dorsey would be a good start, if he can bring down his astronomical 8.1 fouls per 40 minutes. His rebounding numbers were terriffic at both ends of the court - 18.6 oRb%, 24.5 dRb %, 21.6 Rb%. That's Sean May territory.
Matt Haryasz*, 31.4 mpg, 1.5 bp40, 21.8 dRb%
Rob Little, 24.0 mpg, 17.6 dRb%
Did you know that Stanford had the second best PPP defense in Pac-10 play? Neither did I. They bring back their top three and seven of their nine top scorers.
Jermareo Davidson*, 25.9 mpg, 2.3 bp40, 23.3 dRb%
Chuck Davis*, 33.0 mpg, 2.4 bp40, 14.5 dRb%
Neither of these guys were too foul-prone, which allowed 'Bama's defense to lead the SEC in opponents' eFG% and FTA/FGA. Only Kentucky allowed fewer PPP. Alabama didn't force many turnovers, so I'm thinking the defense won't miss Kennedy Winston that much (not that he had any steals anyway).
Nick Fazekas*, 31.5 mpg, 2.0 bp40, 24.8 dRb%
Kevinn Pinkney, 29.2 mpg, 1.1 bp40, 18.7 dRb%
The defense should be solid again, with two seven-footers fighting for Pinkney's minutes. But who will score this year? Nevada's offense was only average in the WAC last year (102 PPP), and Fazekas is the only returning player with an offensive rating over 100 and at least 30 shots last year.
Ronny Turiaf, 30.8 mpg, 2.4 bp40, 22.1 dRb%
Just the opposite of Nevada. The offense should be excellent, again, but it'll be interesting to see if they can improve on last year's league average defense (102 PP in the WCC) after losing Turiaf. Might be the most overrated team (top 5, really?) this side of West Virginia.
Well there you have it - some of the best frontcourt defense in the country from a year ago. Please email me if you have any questions, or just want to tell me I went about this all wrong. I'm stopping a little abruptly, but I'm sure this won't be the last time I pick up this topic.
Tuesday, September 20, 2005
Maybe The Sky Isn't Falling
Iowa got a verbal commitment from Jamie Vanderbeken, the Canadian center who visited campus last weekend. Read about him from Blair Sanderson and Van Coleman.
Sanderson also reports that Iowa has a good chance to land Jamarcus Ellis, a 6-5 wing player from a Florida junior college. He's originally from Chicago and would have three years of eligibility remaining.
And no, I'm not turning into one of those bloggers who just posts a smorgasboard of links - I'll get something original up this week.
Sunday, September 18, 2005
Tom Kakert of HawkeyeReport.com has an article up about two basketball players who visited Iowa City this weekend. One was current verbal commit Cyrus Tate, the other was Jamie Vanderbeken out of Canada. It sounds like Vanderbeken liked what he saw at Iowa -
"[M]y impression thus far is very good. I came here with an open mind and reallyThe 6'10" center will choose to play at Iowa, Oregon State, or UCLA (which has yet to offer a scholarship).
did not know what to expect. Thus far it has exceeded my expectations."
Tate background (April 26)
Vanderbeken ranked 13th in Canada by Hoop Life (December 31)