free NBA basketball picks
Home | Teams/Players | Commentary | Premium Services | Contact Us

Valuing the Preseason

NBA Statistical Analyst Kevin Pelton

by Kevin Pelton, 10/31/05


The scouting process in particular and the process of learning about players and teams in general has always fascinated me. Lately, I've had this topic on my mind for a couple of reasons. One of them is that, as you've undoubtedly noticed if you're at this Web site, the NBA's preseason has just concluded. This means all sorts of observations about how teams and players have changed over the off-season. On top of this, I recently read The Jump, Ian O'Connor's excellent book detailing Sebastian Telfair's senior year of high school and his decision to go directly from Lincoln High School to the NBA.

If you've read my work before, you'll know that I am a passionate advocate of statistical analysis. In its own way, reading about the roller-coaster ride that was Telfair's draft stock over the year leading up to his selection in the 2003 NBA Draft makes an argument for my position. Telfair's underlying ability never really changed over the course of his senior year, I don't believe, but his performance did, affected by his mood (late in the season, Telfair was dealing with the stress of decisions about an agent and a shoe company, as well as his brother's arrest), his health (Telfair sprained an ankle early in the year) and the caliber of his opposition.

Most of the general managers and other decision-makers charged with deciding whether to stake their team's future on Telfair saw him only a handful of times over the course of his senior campaign. Depending on when they saw him, their opinions could vary widely.

In this specific case, stats wouldn't have been much of a help - I can't imagine high-school stats ever being looked at as a meaningful indicator - but in general, the beauty of statistics is that they "see" every game, good or bad, producing a more fair and balanced look at a player.

At the same time, it goes almost without saying that stats can never substitute for regular observation. Part of the reason why I consider stats so valuable in the case of college scouting is that it's simply unfeasible to watch every game played by a prospect. Once a player joins a team, however, it's easy to get a complete picture of his skills. I only get to watch a half-hour of practice, tops, in my role covering the Sonics, but I still felt like I got a fairly complete picture of the games of the players invited by the team to training camp this season.

To me, a fascinating case study in observation and a dynamic opinion of a player is second-year Sonics swingman Damien Wilkins. After an unspectacular college career, the son of long-time NBA veteran Gerald Wilkins and nephew of All-Star Dominique Wilkins put himself in draft consideration with an outstanding performance at the Chicago Pre-Draft Camp, where he averaged 16.7 points per game. Despite that performance, I still wasn't sold on Wilkins' potential.

"In my book, those three games pale in importance compared to the 30 Wilkins played for Georgia last season, and he simply wasn't that good in them," I wrote in my draft preview.

NBA front offices apparently agreed with me, as Wilkins went undrafted. After playing for the Milwaukee Bucks in the Minnesota Timberwolves Summer League, Wilkins showed up in Seattle to play for the Sonics entry in the Rocky Mountain Revue Summer League. He quickly established himself as a good player for the makeshift squad and averaged a team-high 15.3 ppg in Utah.

At this point, having still only seen Wilkins play a couple of times in the practices leading up to summer league, I thought he was definitely worth a training-camp invite but no lock to make the team.

Wilkins was indeed invited to camp, where he played smooth, intelligent basketball at the offensive end of the court and held his own on defense. Two weeks into camp, I chatted with some friends at the Sonics' open practice, and I distinctly remember telling them that Wilkins was quote, "an NBA player," which is high praise for an invitee.

Wilkins rewarded my faith with 17-point preseason outings in two of the next three games and made the team. When injuries gave him a chance to play regularly in March, Wilkins showed the world what those of us who had seen him in practice already knew - he was capable of being an NBA regular.

To me, Wilkins shows where preseason results have a lot of value. Starting at the 2003 Draft, I had a fairly low opinion of Wilkins, but very little certainty about that opinion, considering how little I had seen him play. I quickly discarded that initial impression when more credible evidence to the contrary came along.

Another interesting example is Marquis Daniels. The SEC isn't high on my college radar, so, like Wilkins, I hadn't seen a lot of Daniels going into the draft. In his case, however, I liked his numbers, and called him a sleeper. Undrafted, Daniels played well in summer league and preseason, then did the same in limited minutes early in his rookie season. As a result, it was no surprise to me when he played well in a regular role late in the year.

To me, preseason statistics have no meaning in the case of established players. For example, Peja Stojakovic shot just 37.0% from the field and 31.0% from downtown in the preseason. Should Kings fans be worried? Of course not. Stojakovic is widely regarded as one of the best pure shooters in the NBA, and eight games of statistics don't tell us anything (it could be a different story if there was an apparent injury, or Stojakovic's form had fallen apart over the summer). In the case of Stojakovic, this seems obvious, but the same courtesy should also be given to established players who don't quite reach All-Star caliber.

At the team level, the value of the preseason is limited. I did a quick study to compare last year's preseason results with how teams fared in the regular season. Though there actually turned out to be predictive value at the top and bottom - Phoenix was one of two teams that won seven games, while Atlanta was the only one to lose seven - the overall value is limited. I calculated the correlation, which measures the strength of a relationship with 1 (or -1) indicating a perfect one-to-one relationship and zero meaning no relationship at all. The correlation between preseason wins and regular-season wins was only .273.

What's interesting is that my host, Roland Beech, did find some predictive value in the NFL preseason in a study for 82games' sister site, TwoMinuteWarning.com. In theory, I'd think the NBA preseason would be more meaningful, as it lasts twice as long and starters play a higher percentage of minutes. Perhaps the relatively fierce competition caused by non-guaranteed contracts accounts for this distinction.

At the same time, it's easy to see why preseason results are meaningless. Teams don't scout during the preseason; they're worried about what they are doing, not what their opponents are doing. Most teams run base offenses and defenses, not wanting to tip their hand as to the more esoteric strategies they'll employ in the regular season. And rotations aren't usually established until the end of the preseason, it at all.

Still, in the face of all this evidence, there still are many fans and members of the media who treat preseason results as important. (It should be noted as a counter-point that nobody seems to have changed their championship pick away from the Spurs after San Antonio's 2-7 presesaon.)

At no point in this column did I wish to imply I had all the answers - or even any of them - to the questions I've posed about the meaning of the preseason or the process of re-evaluating players and teams. But I do think this is an important discussion, one well worth having. I'm hoping it can be had at the new CourtsideTimes.net, a Web site that opened today and touts itself as "Basketball analysis beyond conventional wisdom." I urge you to give it a look.

Kevin Pelton formerly wrote the "Page 23" column for Hoopsworld.com. He provides original content for both SUPERSONICS.COM and storm.wnba.com, where you can find more of his analysis of both the NBA and the WNBA.
He can be reached at kpelton@hoopsworld.com

Also see Kevin's previous columns for 82games.com:
The Year in Stats
Why I'm an APBRmetrician
Wanted: Open Minds
Investigating Dwyane Wade's Injury Risk
The Trade: Eddy Curry vs Mike Sweetney
Rating the Rookies: Projected Fantasy Stats


Rate this Feature
Poor   Fair   Good   Excellent

Enter your comments in the box

Email (optional)

We want your feedback! Tell us your thoughts

Copyright 2005 by 82games.com, All Rights Reserved