APBRmetrics Forum Index APBRmetrics
The statistical revolution will not be televised.
 
 FAQFAQ   SearchSearch   MemberlistMemberlist   UsergroupsUsergroups   RegisterRegister 
 ProfileProfile   Log in to check your private messagesLog in to check your private messages   Log inLog in 

Evolving strength of the league

 
Post new topic   Reply to topic    APBRmetrics Forum Index -> General discussion
View previous topic :: View next topic  

Which column represents the evolving strength of the NBA?
.990 (NBA inevitably gets weaker)
0%
 0%  [ 0 ]
.995
0%
 0%  [ 0 ]
.998
0%
 0%  [ 0 ]
1.00
14%
 14%  [ 2 ]
1.002
21%
 21%  [ 3 ]
1.005
21%
 21%  [ 3 ]
1.010 (NBA inevitably gets stronger)
0%
 0%  [ 0 ]
Don't understand the data, but would like to
14%
 14%  [ 2 ]
Understand data but don't care
28%
 28%  [ 4 ]
Don't understand and don't care
0%
 0%  [ 0 ]
Total Votes : 14

Author Message
Mike G



Joined: 14 Jan 2005
Posts: 1871
Location: Delphi, Indiana

PostPosted: Tue Apr 22, 2008 4:53 pm    Post subject: Evolving strength of the league Reply with quote

As outlined in this thread:
http://sonicscentral.com/apbrmetrics/viewtopic.php?t=1680&start=15

... the strength of the average NBA player can be measured relative to the previous season, by his minutes and his rebounding rate, over those 2 seasons.

league-strength relative to 1977: varying cumulative index
Code:

year   .990   .995   .998   1.00  1.002  1.005  1.010    year
1953   1.05    .93    .87    .82    .79    .73    .65    1953
1965   1.18   1.11   1.07   1.05   1.02    .99    .93    1965
1971    .88    .85    .84    .83    .82    .80    .78    1971
1977   1.00   1.00   1.00   1.00   1.00   1.00   1.00    1977
1987   1.11   1.17   1.21   1.23   1.25   1.29   1.36    1987
1993    .95   1.03   1.08   1.12   1.15   1.21   1.31    1993
1998    .99   1.10   1.17   1.22   1.27   1.36   1.51    1998
2002    .89   1.01   1.09   1.15   1.20   1.30   1.47    2002
2007    .87   1.01   1.11   1.17   1.25   1.36   1.58    2007


These representative seasons are relative highs and lows (inflection points, in graphing terms), across the columns. The relative values vary, depending on the parameter at top.
Under the column headers, larger number = greater strength of competition.
_________________
`
There's no I in analysys.
Back to top
View user's profile Send private message Send e-mail
94by50



Joined: 01 Jan 2006
Posts: 458
Location: Phoenix

PostPosted: Wed Apr 23, 2008 12:12 am    Post subject: Reply with quote

I would say it's somewhere between 1.004 and 1.008. I think 1.005 might be a little low.

Now that I'm thinking about it more, maybe not. I'm sure a lot of people would choose .998 (referring to the idea that the game peaked around the mid-80s). I voted 1.005. But you know, it's probably lower than that. I'm not sure that the league is any stronger than it was in the mid-80s, but I don't think it's weaker.
Back to top
View user's profile Send private message
Mike G



Joined: 14 Jan 2005
Posts: 1871
Location: Delphi, Indiana

PostPosted: Thu Apr 24, 2008 6:14 am    Post subject: Reply with quote

As I assembled these data, I noticed a couple of things about the '50s-thru-'00s:

- As is pretty well known, later decades have seen players going strong well into their 30's, even to age 40. In the '50s-'60s, everything after age 30 was 'borrowed time'.

- Also easily seen, lots more younger players in the league lately.

So basically, the core/plateau group of 24-28 year-old players is a smaller % of the total (minutes) than it once was.

- Yet through all this, the 24-28 contingent is and has always been a 'plateau' period: Minutes and RebRt do not change much or at all during this time. RebRt drops very slightly, though: Correcting for this, I've introduced the factors (.99, .995, ... 1.01) shown as column headers.

So when you 'agree with' a factor like 1.002 (say) because you feel the league is just about as strong as it was in 1987 (1.27), you also should check your feeling for the index in 1971 (.82), and 1953 (.79). You should find yourself agreeing equally all along the timeline.

Does anyone suggest why there should be a peak at 1998? It's a peak in all columns, but it's an alltime high according to a 1.002-1.004 factor.
_________________
`
There's no I in analysys.
Back to top
View user's profile Send private message Send e-mail
94by50



Joined: 01 Jan 2006
Posts: 458
Location: Phoenix

PostPosted: Fri Apr 25, 2008 10:10 pm    Post subject: Reply with quote

I wish I had an idea for the last question. My first thought was that the prep-to-pro exodus hadn't yet begun, but Garnett was two years in by then. Certainly it hadn't peaked yet.
Back to top
View user's profile Send private message
Mike G



Joined: 14 Jan 2005
Posts: 1871
Location: Delphi, Indiana

PostPosted: Sat Apr 26, 2008 7:57 am    Post subject: Reply with quote

Using the 'most favored' 1.002 factor, the last 32 years look like this:
Code:
1976     .920
1977    1.000 - The merger year; players from 5 ABA teams assimilated.
1978    1.011
1979    1.041 > Steady rise in talent concentration
1980    1.093
1981    1.089 - Dallas franchise added, 1/23 of the total league.
1982    1.125
1983    1.141
1984    1.165 > More steady rising
1985    1.190
1986    1.248
1987    1.255
1988    1.255 - Rise stops; talent has peaked.
1989    1.224 - Charlotte + Miami. Biggest dropoff since '69.
1990    1.188 - Minny + Orlando. Even bigger talent drop.
1991    1.197 - One year of modest rise
1992    1.172
1993    1.152 - League bottoms out.
1994    1.187 > On the rise again
1995    1.206
1996    1.188 - Vancouver + Toronto added
1997    1.244 > Rise continues
1998    1.274 - All-time high for talent?
1999    1.251 - Lockout season interrupts rise
2000    1.260
2001    1.248 > 2 years of decline
2002    1.205
2003    1.237 > 2 years of rise
2004    1.261
2005    1.238 - Charlotte added (again)
2006    1.214 - Another year of decline
2007    1.247 - And a rise.

It's gratifying to see the expected downturn in every expansion year. The recent Bobcats addition isn't particularly noticeable in the 2004-06 interval.

The anomalies, to me, are: the '91-93 dropoffs, the '98 high, the '00-02 recession, and the '05-'06 sag.
A one-year drop could be the result of an abnormally high injury rate among better players. There would be an average rate, and relatively injury-free seasons. Could 1997-98 be one of the latter?
The biggest one-year rise in the competitive index is from '96 to '97; the biggest dropoff (other than expansion seasons) was from '01 to '02.

Comparing these 2 seasons, 1997 and 2002, for overall player health, I can sort (from players with just 200 minutes), by my own T Rate, and compare their minutes/games:
Code:
by 'T':  top 50        top 100      top 200
year    gms   min     gms   min    gms   min
1997   73.2  2700    74.1  2547   69.5  2085
2002   70.5  2568    71.6  2430   70.8  2063

So the top 200 players were about equally viable in these sample years. But the top 50 and 100 were noticeably more impaired in 2002 than their equivalents in 1997.
Missing at least 20 games in '02: Webber 28, Iverson 22, Brandon 50, Francis 25, Mashburn 42, Marshall 24, Hill 68, McDyess 72, Coleman 24, Jordan 22, Carter. That's 11 from the top 50, and doesn't detect players <200 minutes.
In '97, 9 of the top 50 'T's missed 20+ games; Shaq most with 31.

2002, relative to 1997, missed some 6600 minutes of play, from various allstar-level players. And 11,700 minutes from top-100's -- equivalent to losing 400 minutes from each team's top 3-4 players.
_________________
`
There's no I in analysys.
Back to top
View user's profile Send private message Send e-mail
gabefarkas



Joined: 31 Dec 2004
Posts: 980
Location: Durham, NC

PostPosted: Tue May 06, 2008 7:57 am    Post subject: Reply with quote

First shalt thou take out the Holy Calculator. Then, shalt thou use three. No more. No less. Three shalt be the number thou shalt use, and the number used shall be three. Four shalt thou not use, nor either use thou two, excepting that thou then proceed to three. Five is right out. Once the number three, being the third number, is used, then, calculatest thou thy Holy Microsoft Excel Spreadsheet with thy merger year, who, being pivotal in Mike's mind, shall snuff it.





But seriously, I chose 1. Why? Because I think it's silly to compare 2007 to 1977. It's impossible to take everything into account.
Back to top
View user's profile Send private message Send e-mail AIM Address
Mike G



Joined: 14 Jan 2005
Posts: 1871
Location: Delphi, Indiana

PostPosted: Tue May 06, 2008 10:26 am    Post subject: Reply with quote

Yes, it's probably silly to play, watch, analyze, or talk about basketball, since it's just a bunch of guys running around in shorts, bouncing a ball, etc.

Some say the old-timers would mop up today, and some say the opposite. Maybe one side is entirely right, and the other is quite wrong. Maybe curiosity, in and of itself, has no survival value, and thus is a 'silly' impulse.

It's possible to study voting trends from the '50s to the present. You don't have to take everything into account, or anything specifically. You may just track the evolution of voter psychology: How many voted, whether they voted conservatively, etc.

How you interpret a trend is up to you. You may even doubt that a trend has been detected, and so have no interpretation. Or you may just not care. If you don't care, but you go to much trouble to say so, that may seem silly, too.
_________________
`
There's no I in analysys.
Back to top
View user's profile Send private message Send e-mail
gabefarkas



Joined: 31 Dec 2004
Posts: 980
Location: Durham, NC

PostPosted: Tue May 06, 2008 9:26 pm    Post subject: Reply with quote

You began by posing a question to the group:
Quote:
Which column represents the evolving strength of the NBA?


Among the answers were three options I would say could reasonably be called "silly" by most readers.

So, I gave an answer. To me, trying to draw these comparisons is a very daunting task, and one I'd be afraid of attempting. Nu?
Back to top
View user's profile Send private message Send e-mail AIM Address
Mike G



Joined: 14 Jan 2005
Posts: 1871
Location: Delphi, Indiana

PostPosted: Tue May 06, 2008 9:44 pm    Post subject: Reply with quote

I guess if you have a strong feeling toward one part of the spectrum, the other parts may seem 'silly'. But each person will have his idea of which are plausible and which aren't.

But no, I don't know which end(s) of the spectrum you'd find implausible. I figure it's good to have some extreme options, if only so everyone can feel comfortably moderate.
_________________
`
There's no I in analysys.
Back to top
View user's profile Send private message Send e-mail
Display posts from previous:   
Post new topic   Reply to topic    APBRmetrics Forum Index -> General discussion All times are GMT - 5 Hours
Page 1 of 1

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum


Powered by phpBB © 2001, 2005 phpBB Group