NBA over/under win totals

[Allah_In]

the problem with Houston is 53.5 is a really large number, especially considering the teams they play day in and day out

[Runner Runner]

Some words of caution before making these season long wagers:

There is a slight over-bias in these numbers as the average # of wins per team is 41.5 (should be 41). This should make you more likely to take an Under.

One thing you guys should also consider before wagering is regression to the mean from season to season. A typical team will move closer to .500 from one season to the next, this is a given.

So, in order to make a play on an over above 41 wins, hopefully you will have some real strong evidence of improvement for that team. Especially if they won 41 or less games last year.

Also, you are tying up money for a long time, it is often better used during the season.

[polkaface]

[ QUOTE ]

One thing you guys should also consider before wagering is regression to the mean from season to season. A typical team will move closer to .500 from one season to the next, this is a given.


[/ QUOTE ]

There is one major problem with your rationale. Each team is not created equal (or 'typical' as you said) so the "mean" for a certain team is not 41 wins. There is no way you can look at the Wolves and look at the Mavs and say that each teams expected win total is the same. If you played out 1000 seasons with these exact two teams as they are set up right now they would not go .500. They WOULD revert to a certain number, but that number would be different for the two teams and it would be that team's expected win total. Not the average win total for the league.


There are different players, different coaches and different front offices. Some are good and competent, others aren't. If all of these things were the same then yes you would make a very good point.

Your argument is more or less the same as saying that Mike Cameron and Ichiro Suzuki will end up with the same batting average over their careers because the mean BA for the league is .270 (or somewhere around there) and only random chance will keep that from happening.

There are reasons why the same teams, people, companies lead categories year in and year out and isn't attributed to luck (which is what you are suggesting if everyone and everything should average out to the same).

[Runner Runner]

[ QUOTE ]
[ QUOTE ]

One thing you guys should also consider before wagering is regression to the mean from season to season. A typical team will move closer to .500 from one season to the next, this is a given.


[/ QUOTE ]

There is one major problem with your rationale. Each team is not created equal (or 'typical' as you said) so the "mean" for a certain team is not 41 wins. There is no way you can look at the Wolves and look at the Mavs and say that each teams expected win total is the same. If you played out 1000 seasons with these exact two teams as they are set up right now they would not go .500. They WOULD revert to a certain number, but that number would be different for the two teams and it would be that team's expected win total. Not the average win total for the league.


There are different players, different coaches and different front offices. Some are good and competent, others aren't. If all of these things were the same then yes you would make a very good point.

Your argument is more or less the same as saying that Mike Cameron and Ichiro Suzuki will end up with the same batting average over their careers because the mean BA for the league is .270 (or somewhere around there) and only random chance will keep that from happening.

There are reasons why the same teams, people, companies lead categories year in and year out and isn't attributed to luck (which is what you are suggesting if everyone and everything should average out to the same).

[/ QUOTE ]

I think you misunderstood what I said, I said that a typical team will move towards .500. This is simple and inevitable. It has nothing to do with any teams in particular, just that in any given year, there will be more teams that move towards .500 then there are teams that move away from .500.

So, try and have a real strong data if you are picking a team to have more wins then last year and above 41 wins. Also, the same thing goes for unders where you expect a team to lose more then last year and below 41 wins.

A simple way of understanding what I am saying, is that there is a push towards mediocrity in sports that acts like gravity, bringing the good teams down and the bad teams up from year to year. There are teams that sometimes buck this trend, but the push does exist.

[polkaface]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]

One thing you guys should also consider before wagering is regression to the mean from season to season. A typical team will move closer to .500 from one season to the next, this is a given.


[/ QUOTE ]

There is one major problem with your rationale. Each team is not created equal (or 'typical' as you said) so the "mean" for a certain team is not 41 wins. There is no way you can look at the Wolves and look at the Mavs and say that each teams expected win total is the same. If you played out 1000 seasons with these exact two teams as they are set up right now they would not go .500. They WOULD revert to a certain number, but that number would be different for the two teams and it would be that team's expected win total. Not the average win total for the league.


There are different players, different coaches and different front offices. Some are good and competent, others aren't. If all of these things were the same then yes you would make a very good point.

Your argument is more or less the same as saying that Mike Cameron and Ichiro Suzuki will end up with the same batting average over their careers because the mean BA for the league is .270 (or somewhere around there) and only random chance will keep that from happening.

There are reasons why the same teams, people, companies lead categories year in and year out and isn't attributed to luck (which is what you are suggesting if everyone and everything should average out to the same).

[/ QUOTE ]

I think you misunderstood what I said, I said that a typical team will move towards .500. This is simple and inevitable. It has nothing to do with any teams in particular, just that in any given year, there will be more teams that move towards .500 then there are teams that move away from .500.

So, try and have a real strong data if you are picking a team to have more wins then last year and above 41 wins. Also, the same thing goes for unders where you expect a team to lose more then last year and below 41 wins.

A simple way of understanding what I am saying, is that there is a push towards mediocrity in sports that acts like gravity, bringing the good teams down and the bad teams up from year to year. There are teams that sometimes buck this trend, but the push does exist.

[/ QUOTE ]

I now understand what you meant and I can see your point.

It does turn out that about 1/3 of the teams from the past 5 years do not move IN THE DIRECTION OF (not necessarily closer to) .500 (I mean not necessarily closer in that a team could have been 40-42 and then the next year they are 45-37). I only went on win totals. It would be over 1/3 (by another 2 or 3 teams a year from 11 teams a year to about 13 teams a year) of teams if I went by actual distance from .500.

[New001]

The only line I've played so far is Toronto over 41.5. I'm eying Phoenix under as well.