By Ian Levy
With two weeks of the season under our belt, I thought it time to take a look at some NBA Power Rankings. My Power Rankings are different from many others in that I don’t use any subjective judgements, instead relying on a very specific statistical measurement. My Power Rankings are based on the Differential between each team’s Offensive and Defensive Ratings.
Offensive and Defensive Ratings show how many points a team scores or allows per 100 possessions rather than per game. This is a much more effective way of accounting for a team’s offensive or defensive performance in that it factors in the speed or pace a team plays at. Below is a table showing my rankings, current through today.
The first two columns on this table are the Offensive and Defensive Ratings for each team. The third column is the Differential between the two, and this is the category I used for my rankings; Miami as the best team in the league, Minnesota as the worst. Next you have each team’s record as of last night.
The next two columns represent each team’s Pythagorean Win Loss record. The Pythagorean record is a projection of how many games each team would win if they maintained the same point differential acorss 82 games. It’s important to note that this is not a projection in the traditional sense. This projection is based solely on the team’s play up to this point and doesn’t take into account any future factors such as injury or strength of schedule. The idea is if each team played at the exactly same level as they had in the first two weeks across the rest of the season what would their record be. A team’s points scored and allowed per game is needed to calculate the Pythagorean record and those are the last two columns in the table.
I am sure some people will be surprised to see both Miami and Orlando ranked ahead of undefeated Los Angeles and New Orleans. Orlando is especially surprising given the crushing blowout they suffered at the hands of Miami. This really speaks to how strongly Orlando has been playing other than that one game. Remember, I am not saying that I think Orlando or Miami is a better team, or will win the championship. I am just saying that up to this point they have statistically outperformed L.A. and New Orleans.
A lot of people had Phoenix pegged for a huge drop off this season without Amare Stoudemire. The fact that they are still on pace for a 54 win season depsite several close losses speaks to the quality and depth of their roster and the ability of the coaching staff to maximize their talent.
Washington and Minnesota are both really bad. I mean really really bad. Both teams will inevitably improve over the course of the season. Neither should challenge any historic levels of losing, but either of those teams winning 25 games would be a major accomplishment.
Oklahoma City is currently on pace for a 27 win season. This will certainly change, but they have some serious issues to figure out at both ends of the floor.
Stay tuned for future editions of these Power Rankings!
Ian’s basketball writing can also be found at Hickory-High.com and IndyCornrows.com
Carlos Boozer … WHERE ARE YOU???? Deng … Luol’s just not cuttin’ it. 4 RPG in 37 min/game? Might as well be 6’1″ instead of 6″9″.
I love it! We should call this NBAtrics. We’ll see if the rest of the folks on this site are ready to embrace this type of analysis.
One question: do the Pythagorean W-L projections take games already played into consideration, then project over the remainder of the season, or are they just a reflection of what a team’s record should be if they played at this level over the entire 82 games? In other words, Miami’s 5-2 with a Pythagorean W-L of 72-10. So, does this mean they project to go 67-8 the rest of the way?
I was surprised to read that Nash might want out of Phoenix. Never know what to believe any more. I thought he was pretty happy in the desert.
The Hornets are undefeated! Whoa! I gotta keep up with the standings.
I saw the end of the Warriors/ Raptors game last night. I like the way Golden State is using Curry.
Welcome to the Pine Ian! Great stuff!
Chas 🙂 I was wondering how long it would take you to respond.
Thanks everyone for the warm welcome.
Chas – Pythagorean W/L does not take into account games that are already played. It is a representation of the most likely record a team would have with that point differential across a blank slate of 82 games.
You could take a team’s differential, use the same formula, but just project it across the remaining games and then add in their record for games they had already played. This is likely a vastly oversimplified version of what people like John Hollinger at ESPN and Kevin Pelton at Basketball Prospectus do in creating their projected standings. In that simplified version there would still be a lot of variables unaddressed, variables which I am not totally equipped to deal with at this stage of my statistical analysis. For that reason I stayed away from trying to make projections and just used the Pythagorean W/L as a representation of what level each team has been playing at so far.
Thanks again everyone for reading and commenting.
I’m also curious why you use the actual point differentials to determine the Pythagorean W-L, rather than the Offensive and Defensive Efficiency ratings, which you refer to as a better measure of a team’s performance?
To be honest I don’t know the complete answer. Everytime I have come across the Pythagorean formula it has always used per game measurements as opposed to per 100 possesions measurements. My guess would be that since the formula is projecting games, which are not always 100 possessions, it is more reliable when you use their per game numbers. Pace is included in the per game numbers where it is removed in the Efficiency rating numbers. But pace is factor in winning or losing games so it seems like it would make sense to include that when making win projections. When looking for a flat comparison of Offensive Efficiency across two teams it makes more sense to level the playing field and look at per 100 possessions.
Guys,
Way to much unimportant information…Lakers 3-Peat !
Miami better get Pat Riley on the bench or they may go .500
Much overated Heat will not get past Boston anyway ! Sorry
Josh McRoberts still has not learned. Who recruited him to come to Durham anyway?
Ian –
Cool stuff. I was very interested to see OK City’s projection. They have had all close games however. Their interior defense is an atrocity, which begs the question how they’ll manage this pace a whole season. Good thing they are a very young and energetic team.
Miami’s blowouts have obviously helped them gain a PW projection of 72, but it’s the management of close games that counts in the long run. Do you have a way to measure the amount Chris Bosh is overrated? A post player with no back-to-the basket game playing with the two hardest rim attackers in the league? Might not work out quite as easily as planned. I’m loving the way Boston has handeld them twice WITHOUT Perkins. Paul Pierce gets tweet of the year: “It was a pleasure bringing my talents to South Beach.” I love how they permeate confidence. Can’t measure that with any statistic.