In honor of the on-going writer’s strike, today’s column is a re-run. Okay, it’s not word-for-word something posted before. But the Chris Paul and Deron Williams story has been told before in this forum. What I am doing in this column is updating this tale with data from the first half of the 2007-08 season. In the course of telling the story I am also going to note (and I think I have said this before also) that the Hornets are not quite the surprise this year that people seem to believe. Before I get to that point, though, here is the Paul-Williams story (again).
Updating the Chris Paul-Deron Williams Story
Williams was selected with the third choice in the 2005 NBA draft. Paul was taken with the very next choice. So if we go back about 30 months or so, Williams was believed to be – at least by the Utah Jazz – the better player. However, when we look at the college numbers of each player – reported in Table One (originally posted HERE) — it was clear that Paul was the more productive collegian.
As rookies it was also quite clear that Paul was the more productive player. In 2005-06, Williams posted a 0.020 WP48 [Wins Produced per 48 minutes]. Meanwhile, Paul’s mark of 0.308 easily led all rookies. When the season was over Paul was named Rookie of the Year and people were wondering what the Jazz were thinking on draft night in 2005.
Last year, though, Utah’s fans might have thought the Jazz had been vindicated. Williams and the Jazz advanced to the Western Conference Finals. Paul and the Hornets literally didn’t show up in the playoffs. After seeing Williams in the playoffs for three rounds, and not seeing Paul play at all, many observers began to believe that Utah had made the right choice.
The numbers, though, said otherwise. Table Two reports what each player did during the 2006-07 regular season.
Table Two reports that Williams improved his sophomore year. Paul, meanwhile, was a bit worse. Despite such movements, though, Paul was still a much more productive player last year.
So here is the story I have told about Chris Paul and Deron Williams:
Chris Paul was the more productive player in college.
Chris Paul was the more productive player during each player’s rookie season.
Chris Paul was the more productive player in 2006-07.
At the beginning of this year (in the following columns), I again noted that Paul was the more productive player.
When we look at where the Hornets and Jazz were at earlier this season, and if we consider the 2007 NBA playoffs, it seemed like the Jazz were the better team. And consequently, for many people, Williams had to be the better player. Hence, many people read these columns and politely (or not so politely) offered an opposing viewpoint.
But at the midpoint of the 2007-08 campaign, the Hornets have the best record in the Western Conference. And the Jazz are just barely in the playoff field. Suddenly, as the team results have changed, the perception of Paul and Williams has again shifted. Now Paul is being referred to as a possible MVP candidate. And Williams is now just a very good point guard.
When we turn to the numbers each player has generated this season, we see that the general perception of each player is now consistent with the data.
As Paul has done in the past, he bests Williams with respect to rebounds, steals, and turnovers. In fact, his value with respect to net possessions (rebounds + steals – turnovers) is worth more than eight additional wins across an 82 game season. When you add in Paul’s advantage in assists and personal fouls, you can see that even though Williams is the more efficient shooter (and slightly better at blocking shots), the level of productivity offered by Williams does not come close to what we see from Paul.
Let me repeat something I have said in the past: Statistics are tracked to separate the player from the team. When we look at the numbers, we can see that Paul has consistently offered more than Williams. Although the team results differed, the numbers alwasys told the same story. Too often, though, people ignore (or don’t understand) the numbers. And when that happens, the team result ends up driving the evaluation of the player.
Now let me emphasize the story the numbers tell. Williams is most definitely an above average point guard. Paul, though, is currently the most productive point guard in the league. And one could argue, a good choice for league MVP.
The Hornets Surprise?
One of the reasons Paul is considered an MVP candidate is that few people believed the Hornets would be this good.
So how good is the team from New Orleans?
After 41 games (yes, I know they have played 44 games but I am working off my mid-season data base), the Hornets had an efficiency differential – offensive efficiency minus defensive efficiency – of 6.2. This projects out to about 56 wins across an 82 game season.
What if each player on the Hornets (except the rookies) played as well as they did in 2006-07? The answer is in Table Four.
Table Four tells us that the Hornets, if each player did what he did last year, should expect to be on pace to win between 50 and 51 games in 2007-08. So New Orleans has improved, but only by about five games. In sum, the Hornets are not really much of a surprise. Given what Chris Paul and Tyson Chandler have done in the past, we should have expected this team to be competitive in the West. Yes, the team is improved – primarily due to even more production from Paul – but we should not be surprised.
Once again, if this entire column read like a re-run, that’s because I have said all this before. Even the stuff about Paul and Chandler was said in the following post:
For those looking for new stuff, let me note that on Tuesday night the Hawks finally managed to play their 41st game of the season. So I can now complete the mid-season database. This means that I should be able to offer some original material (or not so original stuff) on the first half of the 2007-08 season. Look for this stuff to be posted soon.
Our research on the NBA was summarized HERE.
Wins Produced, Win Score, and PAWSmin are also discussed in the following posts:
Finally, A Guide to Evaluating Models contains useful hints on how to interpret and evaluate statistical models.