Saturday, December 8, 2007

Hollinger’s Computer Hates the Nuggets

(Boulder-CO) I’ll be the first to admit that I am a little bit frustrated with the Denver Nuggets so far this season. John Hollinger’s new computer generated measuring stick seemingly just hates them. Just look at this bizarre, Vegas style odds article if you need some more proof.

In his most recent computer generated, and might I add cockamamie, measuring stick Hollinger has put his name by a .1% chance the Nuggets win the NBA championship. And while I don’t think, at this point at least, the Nuggets are destined for basketball supremacy; the other odds that the article provides are actually quite maddening. For instance, the Jazz are given a 98.5% chance of winning the Northwest Division with the Nuggets only garnering the other 1.5%. This is irritating because as of this article’s post the Nuggets (12-8) are only a single game behind the Jazz (13-7) with ¾ of the season still remaining. The Nuggets could get everyone healthy (finally!) and win 45 out of the remaining 62 games. The Jazz could run into injury problems and drop eight of their next ten games. You just never know, but I would think that with the way things look right now in the Northwest that something a little bit more like a 51-49% chance either way between Utah and Denver would be more accurate.

My point is anything could, and probably will, happen. And it is my humble opinion that this kind of analysis, while mildly entertaining, is absolutely moronic.

I mean, c’mon now!

How Hollinger could put his good name by anything that has three teams optimistically winning 70 games is absolutely preposterous. Furthermore, we might as well just give the Celtics the Larry O’Brien trophy because according to Hollinger’s computer generated odds they have a 71% chance of being crowned kings of the NBA.

The bottom line is the “Worldwide leader” definitely dropped the ball this time by publishing a piece that puts the NBA regular and post seasons on par with college football’s BCS computer rankings; as both take out reasonable human analysis and replace it with something as mindless and blind as a computer crunching numbers.

Additionally, in Hollinger’s explanatory paragraph it is stated, “Hollinger’s NBA Playoff Odds are based on the Hollinger Power Rankings, designed by ESPN’s John Hollinger. The Hollinger Power Rankings are a measure of each team’s performance in the season so far. Based upon those rankings, each day the computer plays out the remainder of the season 5,000 times to see the potential range of projected outcomes. The results reveal the most likely win-loss record for each team -- and how likely it is for each team to make the playoffs, win the NBA title, win the lottery, and so on.”
To which I reply, "Oh they do, do they?"
Furthermore, If I was writing the cleverly diguised disclaimer it would have read a little bit more like this:

“In reality, all this kind of thing does it attach some percentages to some kind of six month crystal ball reading that means nothing to those of us who watch, read, and study the NBA on a day-to-day basis. Enjoy and don't read to deep into anything!”
I hope the Nuggets players read this and use it as added motivation to win the divison that Hollinger's computer all but has them dead in the water after only 20 games.
1.5% chance of winning the divison my foot.
Go Nuggets!


ThaAnswer said...

I was checking out some of his ridiculous rankings last night. Some of which had us barely finishing 8th and another that had us not making the playoffs! I put this at the top of the "ESPN feels they need to provide us with some new BS" list. Nuggets Nation knows damn well we're makin the playoffs and hell, we might even cause some trouble in em.

Jonathon said...

The problem doesn't lie with Hollinger's system, it lies with your understanding of it. So many people (apparently yourself included) look at rankings like these and instantly dismiss them as useless because of a few perceived incongruities and maybe even what seems like an outrageous claim or two. The truth is, none of these systems (and this is something even Hollinger frequently notes) proclaim to be literal predictions for how the season will unfold--only useful ones. For instance, you have to know before looking at the chart that the machine has NO idea who is playing--it only takes into account team stats. Thus, Hollinger's predictor doesn't know that the Nugs are missing two key players in Nene and Atkins, just like it won't know it once they've returned. It does, however, know and point out several useful truths-- one being that, while the nugs are tied with the jazz in the standings, they have also played by far one of the easiest schedules in the league so far. Can the nuggets start consistently beating good teams? I hope so, but the computer isn't inclined to give them the benefit of the doubt, and I'm not sure I am either.

I agree that the nuggets aren't exactly the best candidates for statistical scrutiny, mainly because they have significant injuries and I (as well as any fan who's watched this team over the years) knew coming into the season that the nugs were going to have significant consistency issues at the beginning of the season, but that doesn't mean you wouldn't learn a few things about the boys in blue from a closer look. (A good start would be to realize that the system simply extrapolates an estimate of how a team is playing RIGHT NOW over the course of the season--do you think the nugs are playing championship caliber, or even division winning, basketball right now? Honestly? In my opinion, and the computers back me up, the nugs have a lot of growing up and healing to do before they're where they need to be. But mostly growing up.)

Jonathon said...

I checked Hollinger's rankings again, and up to this point the Nuggets haven't just had one of the easiest schedules in the league- they've had the league's easiest schedule. This doesn't take into account which of those games were home and which ones were away (although this factors into point differential, with the road team picking up three points), nor does it take into account the effect of back-to-backs (a factor that I think has helped the nugs more than hurt them), but it still means we've played more bad teams than anyone else in the league, and fewer good ones. I just wish we would have taken a little better advantage of it.