Sunday, October 28, 2007

Blue-White rundown

The season kicked off with the Blue-White scrimmage last night with the Blue squad winning 69-64 behind 21 points from Gerald Henderson and 19 from Taylor King. I didn't have a chance to do defensive stats for the game (and probably won't be able to do any until the real games start against NC Central), but I did have a chance to look at the tempo-free stats, as valuable as 30 minutes of intra-squad scrimmaging can be.

The efficiency and usage stats in particular I found to be pretty useful because they sort of jived with performances from last year, and also sort of fell in line with expectations for most Duke fans.

Blue


Player Efficiency
Usage
Gerald Henderson 131.73
27.78%
Taylor King 129.81
20.22%
Brian Zoubek 104.56
17.50%
Greg Paulus 115.36
14.32%
DeMarcus Nelson 108.84
19.77%




White


Player Efficiency
Usage
Kyle Singler 120.79
30.28%
Lance Thomas 92.73
16.30%
Nolan Smith 91.83
29.72%
Jon Scheyer 112.74
14.60%
Jordan Davidson 97.15
10.90%

Blue was a better team and these numbers sort of reflect that pretty obviously. They shot 10 for 19 from behind the arc, overall had an eFG% of around 65%. White stayed close behind Singler's incredible game (23 points on 9 for 17 shooting, 2 assists, 6 ORBs, 3 DRBs) and by causing 15 turnovers (25.9%), but Blue was ultimately too much.

Overall as a team (both sides), we shot over 40% from behind the arc, and the game was played at a breakneck pace - 77-80 poss per 40 minutes. It's doubtful that we'll be able to keep a pace like that all season - most teams play much slower these days and fast and slow teams tend to average out to medium speed when they play - but it shows that we're going to try to push the ball more this year, which is a welcome change from the standpoint of the fan and concerns for recruiting. Players love to play a fast pace and it's always been a drawing point for Duke: we play fast.

At any rate, it's hard to glean too much from the Blue-White game. Thursday we'll play Shaw University in an Exhibition. Most likely we'll win by 40. Again it will be hard to glean too much. As I recall, last year through the exhibitions and the first few games of the year, Zoubek was actually our star offensively. Nevertheless, our performance against Shaw and Barton will yield some useful information and start to answer some questions about this team.

  • Does Paulus make more mature decisions this year?
  • Will our freshman be able to contribute more immediately than our freshmen did last year? Is Kyle Singler the best freshman in the ACC?
  • Will Gerald Henderson live up to his potential?
  • Can we get our offense back to Duke levels? Can we reduce turnovers and take better shots?
  • Can we defend the post?
To me those are the chief questions for this team. If the answer is yes to them all, I think we contend for the ACC title and can go deep in the tournament, though that might be a little optimistic. Hell, it's the preseason - when is optimism more warranted for any fan of any team?

Thursday, October 25, 2007

A Tale of Two Players

Ken Pomeroy recently wrote an interesting piece about what constitutes similarity from one player to another. His ultimate thesis was that mainstream media fixates on a pair of players, usually because of superficial similarities, but rarely delves any deeper into the subject to find players that produce similarly on the court. I have long believed that Mike Dunleavy was the 3rd pick in the 2002 draft mainly because he was white. While Dunleavy was a fantastic player at Duke, I also have no doubt that the ceaseless Larry Bird comparisons would not have existed if Dunleavy wasn't white. At any rate, that's neither here nor there. The point of this post is to look at the careers of two Duke players removing from the analysis all mention of names, years they played, who their teammates were, and so forth. Unfortunately, unlike the comparison that Pomeroy fixates on at the end of his article (Kevin Durant and Luke Harangody), the two players in this comparison are both of the same race. And they play(ed) the same position. And they play(ed) for the same school. Comparisons would seem to be inevitable... but between these two players they almost never happen.

To start, I don't know Pomeroy's system of creating "similarity scores." That's fine. I'd rather just list important stats for both players to get a more accurate picture of what each player brought to the table. We'll start with his freshman year.

Player 3P % 2P % eFG% FT% TO% FT Rate Ast% Stl% PPG ORating %TMPoss
Player A 35.6% 34.7% 42.9% 76.9% 35.1% 54.6% 33.0% 2.7% 11.6 93.4 19.1%
Player B 31.4% 40.7% 43.0% 78.4% 33.0% 50.3% 26.1% 2.7% 9.2 92.6 17.2%


One quick word on the stats chosen. I originally also looked at offensive and defensive rebounding rates, and block rates, but I don't find either very illustrative. Both players had insignificant block rates and rather low rebounding rates. It's hardly important. I also want to point out that "PPG" is not points scored per game, but points produced per game, which is a different calculation involving an estimate of how many points this player individually created/produced for the team. It takes into account, for instance, assists (as are most important for this comparison). It also takes into account total team points scored, which can vary from team to team based on tempo and overall offensive efficiency.

At any rate, it's very plain to see that these two players had shockingly similar freshman seasons at Duke. They shot a nearly identical eFG%, predicated by one's ability to shoot a bit better inside the arc and the other's ability to shoot 3s slightly better. They shot a nearly identical percentage from the line. They both had tremendous turnover problems. Player A got to the line a little more often, but nothing incredibly significant. They had nearly an identical rate of stealing the ball. However, Player A assisted many more baskets while on the court, and not shockingly had a higher points produced per game, even though both players operated at about the same individual offensive efficiency. Player A also used more possessions - another contribution to his greater points produced.

Player A probably had a better season, but the difference is not compelling. Now we move on to the second year.

Player 3P% 2P% eFG% FT% TO% FT Rate Ast% Stl% PPG ORating %TMPoss
Player A 40.4% 44.8% 53.8% 72.8% 29.8% 34.2% 31.5% 1.9% 13.6 104.8 19.3%
Player B 45.0% 46.3% 57.5% 75.3% 27.2% 28.4% 23.5% 2.1% 11.9 105.7 21.0%

Both players have improved dramatically, but still remain pretty similar. It's now evident that Player B has improved his shot a great deal and outshoots Player A convincingly. Again they have very similar FT percentages and turnover rates - though both improved. It seems Player A remains able to get to the line more often than B, though both saw their free throw rate drop noticeably. Both suffered a dip in their assist rates (1.5% for Player A, and 2.6% for Player B). Again they have very similar steal rates (though both saw it drop from their freshman numbers). At any rate, at the end of the day, they produce about the same points per game yet again (the difference mainly in the assist disparity). Both improved their offensive efficiency - and significantly, Player B did so while also increasing his individual importance in the offense. As a freshman, he only used 17.2%, but as a sophomore he increased that number to 21%. As a side note, it is more rare to see a player both improve his efficiency and improve his usage - it reflects a great deal of improvement. As Pomeroy notes here, a player with a usage of about 17.1% his freshman year will most likely end up with a usage of about 18.2%. Player B here increased his to 21%, which is somewhere in the 80-90th percentile I would imagine. A substantial improvement.

Well by this point, it's probably no surprise who these players are. Player A is none other than NCAA All-Time leading assist man, and starting point guard for back-to-back national champions, Bobby Hurley. And Player B, of course, is none other than perhaps the single most criticized and maligned Duke point guard in the K era, Greg Paulus.

Seeing as how the two players followed a pretty similar course of improvement over their first two years, I think it's at least moderately illustrative to look at Bobby Hurley's Junior campaign as, perhaps, a preview of things to come.

Player 3P% 2P% eFG% FT% TO% FT Rate Ast% Stl% PPG ORating %TMPoss
Bobby Hurley 42.1% 44.4% 53.7% 78.9% 25.8% 46.8% 35.1% 1.9% 15.5 113.7 22.4%

As can plainly be seen, after his sophomore year, Hurley saw another year of improvement across the board. He shot better from the field, he turned the ball over less often, had a higher assist rate than he ever had (after having a slightly down year his sophomore year), produced more points, and at a higher efficiency, while carrying a larger offensive burden. Hurley is absolutely beloved by Duke fans - considered by many of them as the greatest point guard in NCAA history, and at any rate the greatest point guard in Duke history. I think what this demonstrates is in that any accurate and demonstrable way, Paulus and Hurley had at least similar freshman and sophomore years, and both improved nearly across the board in every major stat associated with their position. I suppose that's the difference between Hurley having Grant Hill and Christian Laettner and Greg Paulus having... who exactly in 06-07?

I can also use this time to dispel some myth that Bobby Hurley was a tremendous defender and Paulus is an atrocious defender. They had an identical rate of stealing the ball their first year (and Paulus actually had a higher defensive rebounding rate both years for whatever that's worth - by about 2%) and Paulus actually had a higher rate his second year. Hurley was chiefly an offensive player in nature, which worked just fine considering how absolutely good his teams were offensively (the 1991-92 team is one of 3 Duke teams to crack an efficiency of 1.2 points per possession). What is sometimes overlooked is that the 1991-92 Duke team is (shockingly) the third worst defensive team in terms of raw efficiency that Duke has fielded in the last 20 years, giving up .9987 points per possession (most Duke teams give up somewhere between .9 and .95, and the awful 95 and 96 teams gave up 1.06 and 1.02 respectively). If anything, Paulus is probably a better defender than Hurley was. And he did it with a broken bone in his foot that needed to be removed. As an aside, do people still believe that defense wins championships? After Florida's mediocre defense in the tournament last year, I hope that myth has been dispelled. The 34-2 Duke squad in 92 is pretty persuasive evidence against that myth too.

Ultimately my point isn't that Greg Paulus is as good as Bobby Hurley. That argument would never fly with anyone - where are Paulus' rings? Where are his endless accolades? I think what the comparison does show, though, is that perhaps we, as Duke fans, should be patient with Paulus. His talent is really awfully similar to Hurley's talent - and neither were in the same stratosphere as Jason (Jay) Williams who developed into a monster as a sophomore. The biggest difference is the level of talent surrounding Hurley vis-a-vis Paulus. This year, most Duke fans are expecting our team to be much improved offensively. Whether that comes true or not, I think we can very reasonably assume that Greg Paulus will have a very successful junior season offensively.

Saturday, October 20, 2007

Durant/Oden and Practice

It's late so I'm keeping this (relatively) short. Reading back over some of my earlier posts, I suppose I ought to feel a bit silly advocating so strongly for Oden now that he appears to be quite injury prone. Oh well, he's still young and surgeons are insane these days. I just hope he comes back and is able to compete at the level we know he could when we saw him play in the Final Four. He seems like a really good kid who doesn't deserve this kind of bad luck.

As for actual Duke basketball, there's not much to say. I got to see a team practice last Wednesday and that was exciting. It looked like everyone who returns had improved, and the freshmen look competitive (Singler looked outstanding). Henderson and Singler will likely be All-ACC at the very least. The style the coaches were advocating was a real up-tempo Phoenix-like game, which perfectly suits a team without a true big guy but with a great number of athletic wings and guards.

My short preview for the season: Duke will win a lot of games, some of which they maybe shouldn't. However, some games we'll run into foul trouble guarding big guys and have issues defending the paint and lose games we maybe should win. I also think we'll be fine rebounding and be one of the best perimeter defensive teams in the country (up there with Kansas and UCLA), as well as a very good 3 point shooting team. Such a team is volatile and unpredictable in some ways because if shots are dropping, they can seem unbeatable. I can see a team like this losing in the first round or going to the Final Four. Either way, I'm excited just to see what the hell we're going to look like this year.