Page 1 of 1
Hollinger Team Power Rankings
Posted: Fri Jan 13, 2012 2:39 am
by Contrarion
Hey guys...been trying to replicate Hollinger's Team Power Ratings in my own Excel doc, but coming across an issue on the first metric I am trying to calculate.
SOS: Average winning percentage of opponents played
I am looking at the SOS number for Philly: .478
No idea how he derives this number.
Their opponents to date:
Por 0.700
Pho 0.444
Uta 0.600
Gol 0.333
Nor 0.300
Det 0.200
Tor 0.273
Ind 0.700
Sac 0.364
Nyk 0.600
Average of these = .451
If I take the Total Wins/Losses combined to get a Win % I get 45-55 or .450
Any ideas? Am I missing something? I e-mailed him directly, but who knows if I will hear back.
Thanks
Re: Hollinger Team Power Rankings
Posted: Fri Jan 13, 2012 2:51 am
by Crow
He might adjust for overall strength of schedule for each of the opponents Philly faced. 6 of the 10 Philly opponents had stronger than average SOS so if such an adjustment was made I'd expect the complex SOS to be somewhat higher than the simple version. He could account for home and away affects on SOS too. Philly was only had 4 home games, 6 away.
John very occasionally answers questions here. Maybe you will get lucky and we'll see if either or both of the things I mentioned are involved. I have a vague recollection that he does these things from dialog from a season past but I could be wrong.
Re: Hollinger Team Power Rankings
Posted: Fri Jan 13, 2012 11:38 am
by Mike G
If Philly opponents have a record of 45-55, and the Sixers are 7-3,
then these teams are 42-48 against teams other than Philly.
So, against the rest of the league, they're 42/90 = .467
This is closer, but not all the way there. It's possible he's using their Pythagorean W%, which is expected W/G based on their point differentials.
You can calculate that or get a rough estimate from the PW and PL columns in Miscellaneous Statistics here:
http://www.basketball-reference.com/lea ... _2012.html
Edit: Come to think of it, I'm pretty sure he weighs recent games more heavily; and then I don't know if the analysis changes as previous opponents' ratings change, or if he sticks with their rating at the time of the game.
And what Crow said -- home and away adjustments? opponents' SOS? -- both make sense, but I dunno.
Re: Hollinger Team Power Rankings
Posted: Fri Jan 13, 2012 2:07 pm
by Contrarion
Thanks Crow and Mike G...
I am familiar with Pyth W%, I might give that a try and see if I can get any closer. Nonetheless, I can still collect his ratings on a daily basis...was just making a project out of some free time.
Re: Hollinger Team Power Rankings
Posted: Fri Jan 13, 2012 2:36 pm
by Contrarion
From his more detailed explanation:
Scoring margin
One of my goals was to create a system that told us more about a team's quality than the standings do.
So instead of winning percentage, these rankings use points scored and points allowed, which are better indicators of a team's quality than wins and losses.
This might not sound right at first, but studies have shown scoring margin to be a better predictor of future success than a team's win-loss record. Thus, scoring margin is a more accurate sign of a team's quality.
That explains why, for instance, five seasons ago the Spurs ranked ahead of the Mavericks even though they had won nine fewer games -- San Antonio's scoring margin was superior. That ultimately proved to be prophetic, as Dallas lost in the first round of the playoffs, while the Spurs won the championship.
This definitely makes me think he is using Pyth W%.
Recent performance
Another key variable in the formula is recent performance, which I included for two reasons.
First, it stands to reason that more recent games are more valid indicators of how strong a team is currently.
Second, I wanted these rankings to follow the model of Marc Stein's "human" power rankings, on the site each Monday, in which a team's recent play is a huge factor.
To accomplish this, I weigh a team's full-season results by two-thirds and its most recent games by another one-third, so the overall ranking gives greater weight to recent games.
You're probably wondering at this point what I mean by "recent." It varies depending on where we are in the season.
For the first 33 games of the season, it means a team's past 10 games.
From that point forward, however, it means the most recent 31.8 percent of a team's schedule. The net result is that, after the first 33 games, a team's most recent 31.8 percent of its schedule will account for 40 percent of its ranking.
Since we are in the first 10 games (at least for Philly example), there would be no weighting involved for recency.
Home and road
The final variable here is home and road games.
In each game, a team's scoring margin is adjusted by the 3.5-point advantage we (and by "we," I mean the Vegas books, of course) expect the home team to have in a game between otherwise equal opponents.
Not sure exactly what he does here. Does he subtract 1.75 from the Home Team PF and add 1.75 to the Road Team PF to adjust for the 3.5 margin, or does he reduce the Home Team PF by 3.5 each game, then calculates the Pyth W% based on these updated figures?
Seems like a complicated black box, not sure I want to try to decipher.