For those who are not Portland fans this review might be too long to bother reading. In this essay, though, why losing teams should change their roster is explained. A potential rookie of the year is identified and the question marks surrounding LeMarcus Aldridge are discussed. Finally, the difference between NBA Efficiency and Wins Produced is once again offered. So even if you do not care about Portland, this might be an essay worth reading.
The New York Knicks and Los Angeles Lakers – the teams representing the largest markets in the NBA — failed to have the level of success last year that their fan base would have liked. In the summer of 2006 each team followed a fairly similar path to improve the happiness of their supporters. In essence, each team decided to bring back basically the same cast of characters that failed last year and see if the results would be different.
Recent history tells us that this is not generally a successful strategy. Over the past ten years 69 teams returned at least ten players from the previous season’s roster. Of these, only seven managed to add ten or more wins to their final record. In other words, only 10% of teams returning basically the same roster managed to dramatically change their outcomes.
As one can see with the review of Portland in 2005-06, no team failed more than Portland last year. And apparently the odds of improving with the same cast of characters seemed too long. Last year the Blazers had twelve players who played at least 1,000 minutes. Of these, only seven will suit up for Portland this year. Steve Blake, Ruben Patterson, Viktor Khryapa, Theo Ratliff, and Sebastian Telfair all played significant minutes for Portland last year but will be playing elsewhere in 2006-07.
The list of new players is led by rookie Brandon Roy, who is penciled in as the starting shooting guard. Of all shooting guards taken in the draft who played college basketball, Roy posted the best college numbers. If Roy does get the minutes of a starter with Portland then he has to be one of the favorites to win Rookie of the Year.
On draft night the Blazers also added LeMarcus Aldridge. In college last year Aldridge averaged 15 points and 9.2 rebounds per contest. These would be respectable numbers in the NBA, but for a player taken with the second pick overall in the draft, such numbers are a bit disappointing. This is especially true of the rebounding numbers. On a per-minute basis Aldridge grabbed 0.27 rebounds. To put that performance in perspective, Tyrus Thomas, the player taken with the 4th pick, posted a per-minute mark of 0.36 in college last year. In general college numbers are higher than NBA numbers, so one should expect Aldridge and Thomas to capture fewer rebounds as a professional. Unfortunately for Portland, if Aldridge captures fewer rebounds in the NBA he will struggle to offer the level of productivity one might expect of a front court player taken with the second choice in the NBA draft.
It is possible that we will not find out if Aldridge is worthy of his draft status his rookie season. Currently he is hurt. Beyond his injury, Portland also added two veteran big-men — Jamaal Magloire and Raef LaFrentz – who will demand minutes in the front court.
Each of these veterans posted numbers last year below their career averages. Magloire had a Wins Produced per 48 minutes [WP48] last year that was just about average (average WP48 is 0.100). For his career, though, his WP48 has been 0.142. So he is capable of being good. LaFrentz was also below average last year but has been much closer to average across his career. Along with the re-signing of Joel Pryzbilla – Portland’s most productive player last year – the Blazers might have a fairly productive front court even if Aldridge can’t produce.
And this is even more likely to be true if Zach Randolph makes a positive contribution to this team. As one can see in the review of Portland last year, Randolph did not help last year.
One might wonder how that is possible. Last year Randolph averaged 18 points and 8 rebounds per game. If we look at NBA Efficiency we see that Randolph ranked 65th in the league, or in the top 15% of NBA players. If we look at this on a per-minute basis, of the 118 players who played at least 2,000 minutes last year, Randolph ranked 57th in per-minute NBA Efficiency. In other words, he was a bit above average.
According to WP48, though, Randolph was ranked 112th out of 118. In other words, Randolph had a horrible season last year if we focus on Wins Production.
Why the difference between NBA Efficiency and Wins Produced? Some might argue that the key is rebounding. It is true that Randolph had below average rebounding numbers last year. Yes, he averaged 8 per contest. But on a per-minute basis he was below average for his position.
Although rebounding was an issue, that is not he primary difference between how we evaluate players and the NBA Efficiency approach. In fact, Win Score and NBA Efficiency employ the same weight for rebounds. The difference between the metrics, and this is spelled forth clearly on page 118, [if you do not have the book yet, one can click HERE to read the page in question] is the importance of shooting efficiency. NBA Efficiency, oddly enough given the name, does not value shooting efficiency. A player can break even on his NBA Efficiency value if he hits 33% of his two point shot attempts. From three point range the break-even point is 25%. Any player who exceeds these thresholds will increase his NBA Efficiency the more shots he takes. So inefficient scorers can post very respectable numbers according to the NBA’s metric.
Given these break-even points a player like Randolph, who shot 44% from two point range and 29% from beyond the arc – does not look too bad in NBA Efficiency. Again, although he shot badly, the more he shot the better his NBA Efficiency value. But in Wins Produced, these dreadful shooting percentages are valued in terms of their impact on wins. And when one takes that approach, Randolph is found to be quite unproductive.
Now Randolph has been efficient in the past. So if Randolph can once again hit his shots, or at least stop taking three pointers (why he took 55 of these shots last year is hard to explain), then the Blazers do indeed have the potential of a productive front court.
Even if this happens, though, questions remain at small forward and point guard. Martell Webster and Jarrett Jack were both below average as rookies. If these players each improve in their sophomore years… okay, now we are looking at way too many “ifs.”
Basically the Blazers need quite a few players to step up this season. Roy has to play as well as advertised. Magloire and LaFrentz have to return to form. And now Webster and Jack have to improve. And I haven’t even mentioned Darius Miles the player who will probably start at small forward. He didn’t play well last year either.
In a very deep Western Conference all these questions marks are not a good sign. Still, there are some pieces in place on this team and it does look like Portland will at least improve upon its dreadful 2005-06 campaign. In fact this team might improve 10 games in the standings, a result that probably would not have happened had Portland returned the same cast of characters.
– DJ
Our research on the NBA was summarized HERE.
Wins Produced and Win Score are Discussed in the Following Posts
Simple Models of Player Performance
What Wins Produced Says and What It Does Not Say
Wages of Wins Journal 
KC
October 24, 2006
So this morning I’m reading my NBA feed when I stumble on this article: If computers ran the NBA, Shaq would be benched
Since the article talks about some statistical metric of player performance, I ran over here to see if you’ve noted it yet. No, but this article mentions the NBA efficiency stat — and maybe that is the metric they’re using to diss Shaq???
Any info on this?
dberri
October 24, 2006
KC,
From the article, Lenova — the computer manufacturer — is providing the NBA with a statistic, which the manufacturer claims is new. All I know about this is from the article. Still, it looks like this statistic is quite similar to the Winston-Sagarin method — or a more sophisticated version of plus-minus. So I am not sure this is new.
We talk about plus-minus in the book, where we discuss the pluses and minuses (sorry about that) of this approach. There are people who do this analysis for specific NBA teams – like Winston and Sagarin – and they get paid for their work. Lenova is providing this to every NBA team as part of their marketing agreement. Here is the question: Will the people who do this for specific teams still get paid in the future?
About Shaq vs. Mourning – in the playoffs Mourning was slightly more productive than Shaq. Not sure that the difference would persuade me to play Mourning ahead of Shaq. Across their careers Shaq has been quite a bit more productive than Mourning.