ESPN NBA

Updated: January 24, 2007, 5:24 PM ET

Hollinger's Power Rankings FAQ

Email Share

By John Hollinger
ESPN Insider
Archive

As part of the enormous response to our new daily Power Rankings, we've received more than 14,000 comments and questions in the last week.

In going through the mailbag, several themes have emerged. Let's look at some of the most common comments and questions, followed by my replies.

HEAD-TO-HEAD

Reed (Dallas): Does head-to-head competition not count?

Brad: Take into account teams sweeping another team.

Philip: How are the Mavs behind the Suns. This should be your formula: Better record? Head to head? Record against the rest of the "elite teams"? Other than have a longer current winning streak, what have the Suns done to prove they are better?

Beemer (Allen): This is very interesting. Can Phoenix be ranked No. 1 and Dallas No. 2 when (a) Dallas beat Phoenix in both meetings this year and (b) Dallas has had a tougher schedule so far, but still a better record? Also, (c) Dallas has a better record against teams .500 and above at 15-4 to Phoenix 's 13-6. And (d) Dallas also has a better record than Phoenix against the top five teams in this ranking -- Dallas 6-2 vs. Phoenix 3-3.

Brent (Dallas): I'd like to see a component of "quality wins" in your equation, not just strength of schedule, but nevertheless I'm a big fan already after one day of these.

HOLLINGER: Probably the biggest question so far is how Phoenix can be ranked ahead of Dallas when Dallas beat Phoenix twice.

The problem with this logic is it asks us to ignore the other 40 games and focus, essentially, on the final six seconds of one game on Dec. 28, when Dirk Nowitzki made a 19-foot jumper to beat the Suns. That game was in Dallas, by the way, so once you adjust for the 3.5-point home-court advantage it comes out as a Phoenix "win" for the purposes of the rankings.

I should also point out that the Suns would have a 31-game winning streak right now if not for that game and an overtime loss to Washington. I think Phoenix is fairly obviously the best team in the league right now.

Along the same lines, the other common refrain emanating from north Texas has been to penalize Phoenix for its poor record against the West's elite, but that's a red herring. Phoenix played all those teams the first two weeks of the season when it struggled out of the gate.

WINNING CLOSE GAMES

Graham (New Haven): Your formula has allowances only for strength of schedule and winning margin. However, one key component to being a good NBA team is the ability to win close games. Please include something useful in your formula (like winning percentage).

HOLLINGER: This is commonly thought to be true, but it isn't.

The real mark of a team is its record in games that aren't close -- like Phoenix's evisceration of Washington last night.

Teams' records in close games tends to vary wildly from year to year with no rhyme or reason, even when their personnel and talent level remain the same. Which suggests that it isn't much of an "ability" at all, but subject heavily to the whims of chance.

MARGIN OF VICTORY

Derek (Derwood, MD): Margin of victory values should be capped for the stat to be meaningful. For very good teams and very bad teams, you end up looking at how well they perform in garbage time as much as anything.

HOLLINGER: A lot of people have made similar points, but this is much less a factor than you might think. NBA coaches tend to play their best players most of the fourth quarter as long as the margin is under 20, and as a result, even for the best teams only a small portion of their games are so one-sided that the starters can spend the second half yukking it up on the bench.

Phoenix is a good example -- even with all the one-sided wins, Steve Nash is playing a career-high 35.7 minutes per game.

Trey (Dallas): I have a question/concern about your formula. A team could go 65-17, have the best record in the league and have a negative victory margin (unlikely, but possible).

The formula needs to take into account winning percentage, or getting a point added for every victory. Ask the 1960 Yankees about victory margin (outscored the Pirates 55-27, but lost the series 3-4).

I love going strictly by the numbers, but winning percentage or the number of victories needs to be included to truly determine the best team. Any team would gladly have a smaller margin of victory if it guaranteed them the title.

HOLLINGER: Scoring margin is a better predictor of future success than win-loss record, and your example unwittingly proves my point. A year later the Yankees won the World Series while the Pirates went 75-79.

HOME AND AWAY

CT: Home games and away games ... are they weighed differently?

HOLLINGER: Yes, they are. The road team gets a 3.5-point bonus, since that's about the standard home-court advantage in NBA games.

RECENT GAMES

Ben: The formula you're using seems to overemphasize their latest 10 games as an indication of power instead of looking at the whole season.

HOLLINGER: That's by design -- we're trying to get a sense of where teams are at right now, and we do that by placing greater weight on the most recent 10 games (which expanded to 25 percent of games after the halfway point of the regular season).

THE INJURY FACTOR

Kevin (Santa Clarita, CA): How about a team playing thru injuries and dominating plus-.500 teams?

HOLLINGER: The injury factor is something a computerized ranking just can't account for. A number of Laker fans want some kind of bonus for injuries, apparently not realizing that their injury problems pale besides those of many other teams.

RECORD VS. GOOD TEAMS

Jeremy (Lincoln): When the teams have played each other it would make sense to use that as a gauge of who's the better team. Based on record alone, the Lakers should be 4th, but based on the fact that San Antonio is yet to beat L.A. this year, the Lakers should jump ahead of them even though they are a half game behind.

HOLLINGER: Lakers fans also want a bonus for playing well against good teams, since L.A. has fared well against the big boys. But that's not more relevant or indicative than stinking it up against bad teams, which the Lakers have also done in spades.

EAST VS. WEST

Alex (L.A.): Only one quibble with the system so far. Looking at the top 15 teams, eight are from the East. Impossible. In considering records, there is no accounting in your formula for the strength of conference, i.e. a win against an East team is not worth the same value as a win against a West team.

So I suggest taking the entire league interconference record and using that percentage to adjust for the discrepancy. This should push a few more of the West teams into the top half of the league and more adequately represent that the top six or seven teams in the West could take out any team from the East. Thoughts?

HOLLINGER: This has been a common complaint. Let me offer three thoughts here for some context.

First, one of the big differences between the two conferences is that nobody in the East can touch the West's top three, which you can see by the wide margin in "Rating" between the two groups.

Second, remember that recent play is an important factor, and that although the Western teams have clearly been better for the season as a whole, several Eastern teams have been much stronger of late.

And finally, the rankings are not as anomalous as you make them out to be. Seven of the top 11 teams are from the West. It's only when you get past that group and into the Minnesotas and L.A. Clippers of the world that the Eastern teams start turning up in large numbers.

HIBACHI!

Chris (Washington, D.C.): The Hollinger system does not factor in the "Hibachi" effect. The margin of victory stat stands in direct conflict with the Hibachi effect. There can be no "Hibachi!!!" without a close game coming down to the final shot.

HOLLINGER: Admittedly, this was an oversight.

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