Commentary

PER Diem: Feb. 24, 2009

How are the Lakers still behind the Celtics in the Power Rankings? Let me explain ...

Updated: February 24, 2009, 4:14 PM ET
By John Hollinger | ESPN.com

Kobe Bryant & Paul PierceBrian Babineau/NBAE/Getty ImagesThe Lakers are No. 1 in the standings, but the Celtics are No. 1 in Hollinger's Rankings.

Since the e-mails wondering how the Los Angeles Lakers aren't atop the Power Rankings keep coming in , let's run through this one more time.

The Lakers are 46-10, the Cleveland Cavaliers are 43-11, and the Boston Celtics are 46-12. This does not constitute concrete proof that the Lakers are better than the Cavs and Celtics -- not at all. Standings margins this small are almost always the product of how things swing in a few close games, rather than a beacon of a substantive difference in quality.

The Lakers are 4-0 head-to-head against Boston and Cleveland, but sorry, that isn't proof either. Not when there are more than 50 other games besides those on each team's profile.

And if you really want it to come down to that, what you are in effect saying is that the entire difference between the three teams comes down to a noncall on Ray Allen's last-second shot in overtime Feb. 5, and that we can disregard the rest of the season entirely. Because if that one-point Lakers win turned into a one-point Lakers loss, we'd have three 11-loss teams vying for the league's best record from here on out.

That's the thing about relying on win-loss records -- because every game is just a binary result, a 1 or a 0, a lot of information is lost. The margin between the two teams is distilled down to one win, one loss, regardless if the winning team prevailed by 40 points or won on a half-court shot at the buzzer.

That, in turn, is why I rely so heavily on scoring margin -- it's a better predictor of future success than win-loss record. Yes, believe it or not, it's even a better predictor of a team's future win-loss record.

Using scoring margin makes it easy to see why L.A. is still a close third behind Boston and Cleveland. Despite their recent success, the fact remains that the Lakers have the league's third-best scoring margin, and were fourth behind Orlando for some time; they still are more than a point per game behind Boston and Cleveland.

L.A. has also done this against a lighter schedule than Boston (.497 opponent winning percentage versus .512 for the Celtics), and with more home games (30 to 26). Based on that information, how could I possibly put the Lakers first?

Yet the e-mails keep coming. So perhaps it's time to address them point by point.

• "Can you possibly be any more biased toward Boston?" asks Tony in Syracuse.

Remember, people, the Power Rankings are 100 percent automated -- the little gizmo in Bristol updates them overnight while I'm sound asleep. Besides, if I'd had the foresight to develop a method over two years ago that would screw the Lakers in 2009, I'd be in Vegas bankrupting the sports books, not in my little home office writing basketball columns.

But it's comforting to know that the presumption that I'm biased against Team X is universal -- I got the same type of letters early in the season when I had the 27-2 Celtics ranked slightly behind Cleveland.

• "Either you really hate the Lakers or you rely on a computer to calculate your rankings like the BCS," begins Ciprian in Carbondale, Ill.

For some reason, I get e-mails like this a lot, even though the Power Rankings and BCS have virtually nothing in common and, in many ways, could not be more different.

The Power Rankings do rely on a computer, but that makes them unlike the BCS, which is mostly driven by human polls and win-loss records. Further, to the extent that the BCS does use computer rankings, they've been specifically instructed to ignore victory margin -- which is precisely the thing that my rankings use most heavily. Also, my rankings historically have been more favorable to teams from Utah.

• Ciprian's letter continues: "How far ahead of Cleveland and Boston does L.A. have to be to become #1? 10 games ahead? Or will your computer say that Cleveland has a better point differential so 10 games isn't good enough?"

Actually, that last part is correct. If the Lakers keep beating Minnesota by 3, and the Celtics keep beating Denver by 38, then the Celtics are going to keep being No. 1 regardless of who has the better record or by how much.

It's worked out that way before, actually -- two years ago San Antonio finished No. 1 despite being a whopping nine games behind Dallas in the standings. I got a lot of letters then, too, but it worked out OK in the end when the Spurs hoisted the championship trophy.

• "Your method is either useless or you are biased toward Boston," says Dexter from Michigan, which is a kinder letter than most -- it leaves open the possibility that my method might be useful, just as long as I'm biased toward Boston.

His letter continues: "The Celtics are neither playing the best basketball nor have the best record. They have been beaten at home and away to the Lakers, and the Spurs beat them recently on their home court. The Lakers are clearly the best team in the league."

I think this is the basic crux of the problem, and it's the same problem I had with Celtics fans earlier this season: Lakers fans are absolutely convinced that they're rooting for the best team, because L.A. beat the other good teams and has the best record. Anything that challenges this worldview inevitably creates problems.

But actually, it's the Celtics who are playing the best basketball of anyone right now -- they just happened to lose twice in front of large national TV audiences. In their past two games they beat two playoff-caliber teams on the road by 20 and 38, respectively, and since their midyear "What's wrong with the Celtics?" slump, they're 17-3; in that same period, the Lakers are 16-4 and the Cavs are 15-5.

Whether that continues remains to be seen, of course. As we know, both Boston and L.A. have important big men injured at the moment.

But if your starting point is trying to rank teams in a way that reflects likely future results, rather than just starting at "Lakers rule!" and working your way down from there, at this point the Celtics and Cavs simply have to rate ahead of the Lake Show.

John Hollinger writes for ESPN Insider. To e-mail him, click here.