Are You Experienced?

Charles Barkley guaranteed on Thursday that Denver would beat New Orleans in the playoffs (making the assumption that the West standings don’t change). Guaranteed. In fact, he called it his only guarantee of all first round series. Not San Antonio over Dallas, not L.A. over the T-Macs, but Denver over New Orleans. Now obviously, this makes me mad as hell. The Hornets annihilate Denver offensively (4th to 11th in efficiency), edge them defensively (7th to 9th), obliterate them turnover rate wise (2nd to 8th), kill them on the offensive glass (11th to 20th), and devastatingly humiliate (I need to write a thesaurus) them on the defensive glass (3rd to 25th!). So, I’m going to go out on a limb and say Barkley’s guarantee relies entirely on vague “playoff experience” conjecture and throws the stats out the window.

But wait! Could “experience” actually mean something? After all, you hear so many former players talk about how overwhelming the playoffs are, Barkley included. Time to head out into the field, and do some investigative journalism; maybe I’ll be caught in sniper fire.

The main gripe against New O is lack of experience as a “unit.” This is an easy way for the media to ignore Peja’s Sacramento Kings runs (and Bonzi’s), the near double-double Tyson Chandler averaged in his first 6 playoff games, and Morris Peterson’s 19 playoff contests. But whatever, we’ll fly with it. First off, let me briefly lay out what I’m going to be doing before you see all the numbers and experience arithmophobia.

  1. Find data on every Western playoff team, 2003-2007 (5 years worth, and the New Orleans years)
  2. Assign years of playoff experience. If a team returned at least two big parts of its core (ie, Kobe and Shaq), I counted that as a continuation. Of course, Stockton and Malone presented a challenge since 2003 was their last of 16 seasons together. More on that later.
  3. Graph playoff experience versus playoff winning percentage (season by season)
  4. Graph Pythagorean W-L (the sum of offensive and defensive efficiencies) versus playoff winning percentage (season by season)
  5. Calculate the correlations for both graphs

As far as step one and two go, the data tables are too large for me to actually place here. If you want to review them (I got all data from b-r) or want them to conduct your own calculations, feel free to ask and I’ll send them right over. They’re 40 by 5 (40 playoff teams by 5 stats (team, playoff experience of core, Pythagorean W-L, Opponent Pythagorean W-L, and Playoff Winning Percentage).

Now, let’s switch gears back to the Stockton and Malone thing for a sec. Putting in that specific data set (2003 Utah Jazz) hinders the “playoff experience matters” argument since that would give Utah 16 years of experience without that great a winning percentage to show for it. So, being the gracious guy I am, this following experience graph doesn’t include that set. Here are two graphs- the first is experience vs. playoff winning percentage, and the second is the difference in regular season Pythagorean Winning Percentages (own percentage minus cumulative opponent percentage) vs. playoff winning percentage.

CLICK FOR LARGER:

As you can tell from the first graph, there’s been almost no correlation between playoff experience and playoff winning percentage the last 5 seasons. The r2 value of 0.066 indicates almost no tendency of winning percentage to follow experience (the closer r2 is to 1, the more dependent they are). This extreme lack of correlation was quite surprising to me, especially when you factor in the fact that both the Spurs’ dynasty and the last two years (and thus, most experienced) of the Lakers’ dynasty are both covered in the graph. By the way, if you include Stockton and Malone’s final season in Salt Lake, the r2 value drops from 0.066 to 0.000. As in Zero.

On the other hand, there’s a nice correlation between regular season success and playoff success. The one obvious outlier on Graph 2 is last year’s Dallas Mavericks (lower, right hand corner), but generally the good teams tend to do well in the playoffs. And comparing the relative correlations, these two graphs tell us that regular season winning percentage is approximately 400% more likely to affect a playoff game than experience.

(If you’re more statistically inclined: the null hypothesis for the second case is “regular season winning percentage does not correlate well with playoff winning percentage.” The degree of freedom would be 38, and we would assume an alpha level of 0.05. Consulting a critical value chart, the intersection of our degree of freedom with our alpha level is at 0.32. This value is vastly superseded by our correlation coefficient, and thus we can reject the null hypothesis, meaning we can be 95% sure of a statistically significant relationship.)

So the next time somebody pulls out the experience card to put down the Hornets, you can snicker a little bit. (Or punch them in the face, that works too). You could bust out this little factoid I uncovered along the way: since the Hornets moved to New Orleans, only three teams with playoff inexperienced “cores” have posted a better Pythagorean percentage than the Hornets have now. Those would be the 2004 Minnesota Timberwolves (adding Cassell and Sprewell), the 2005 Phoenix Suns (adding Steve Nash), and the 2005 Dallas Mavericks (replacing Nash with Jason Terry et al.). Those three teams won an average 8 playoff games apiece, and each won at least a playoff series, with Minnesota and Phoenix winning two.

At the end of the day, I’m not guaranteeing we will beat Denver. There are no guarantees in sports. The Nuggets are an excellent squad that produces mismatches offensively. But guaranteeing Denver will beat New Orleans is totally ridiculous. Dismissing New Orleans as inexperienced is equally insane. People need to start calling out the media for making sweeping statements like this one with no evidence as back up.. This post is just a start.

About these ads

3 Responses

  1. Wow. Your ability to pull numbers and stats is amazing.

  2. Thanks a lot mW! This post took me around 4 hours to do, hahah.. but I think the end result was worth it. When I finished the first season’s data, there was a bigger correlation between experience and playoff ability than regular season ability and playoff ability, which made me almost quit.

    But the more seasons you add, it seems like this trend just gets more and more obvious.

  3. Great number crunching! I’m glad that the Hornet’s overwhelming inexperience in the playoffs doesn’t mean anything!

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Follow

Get every new post delivered to your Inbox.

%d bloggers like this: