Retro Player of the Year Project

[JordansBulls]

I think the most interesting thing here is that Russell, Kareem and MJ each had something over the other.

Russell had the most total Shares (beat out MJ and Kareem) and he won 7 POY
Kareem had the 2nd most total shares (beat out MJ) and he won 8 POY (beating out Russell)
MJ had the 3rd most total shares and he won 9 POY (beating out Kareem and Russell)

So if you ranked them on that it would be:

Code: Select all

         Total Shares (ranked)   Player of the Year (won-ranked)
Russell         1                                      3
Kareem          2                                      2
MJ              3                                      1


They each average out to the same value

[bastillon]

I think the most interesting thing here is that Russell, Kareem and MJ each had something over the other.

Russell had the most total Shares (beat out MJ and Kareem) and he won 7 POY
Kareem had the 2nd most total shares (beat out MJ) and he won 8 POY (beating out Russell)
MJ had the 3rd most total shares and he won 9 POY (beating out Kareem and Russell)

So if you ranked them on that it would be:

Code: Select all

         Total Shares (ranked)   Player of the Year (won-ranked)
Russell         1                                      3
Kareem          2                                      2
MJ              3                                      1


They each average out to the same value

let's call that JBShares

[Doctor MJ]

I think the most interesting thing here is that Russell, Kareem and MJ each had something over the other.

Russell had the most total Shares (beat out MJ and Kareem) and he won 7 POY
Kareem had the 2nd most total shares (beat out MJ) and he won 8 POY (beating out Russell)
MJ had the 3rd most total shares and he won 9 POY (beating out Kareem and Russell)

So if you ranked them on that it would be:

Code: Select all

         Total Shares (ranked)   Player of the Year (won-ranked)
Russell         1                                      3
Kareem          2                                      2
MJ              3                                      1


They each average out to the same value

I'm right there with you until you get to that last line. Such meta-analysis doesn't lend itself to simplistic weighting.

[ronnymac2]

Ahhhh....saying our goodbyes to the project. All right, all right. It's been fun. I learned a lot about over 50 years of NBA history. Thanks to DocMJ and Semi for their work. Thanks to everybody else for their statistical findings, anecdotal pieces, historical research, analysis of teams, players, eras, and evolution of the game, and input regarding how to judge all aspects of the game. The project definitely forced me to be critical of what I value in basketball.

[penbeast0]

Some players I have reevaluated:

Russell -- I had him as GOAT before the project but it was more a 1A/1B thing with Jordan based mainly on his winning. Have gotten more evidence from some great posts by other posters about just HOW dominant his defense was and how that led to winning.

Cousy -- Downgraded him a bit for poor clutch play as well as inefficiency

Havlicek -- Really unimpressive statistically in the 60s much to my surprise and Cowens clearly the main man in the 70s

Frank Ramsey -- really had some nice postseasons where they played him more minutes and he responded with big games. Always thought his HOF award was silly but at least some justification for it compared to his peers.

Pettit -- Always thought of him as clutch from lots of stories about that game 7 takeover against the Celtics but generally weakened in the clutch (where Hagan stepped up)

Hagan -- see Pettit

Wilt -- Has slid below Kareem in my ratings

West -- Mr. Clutch is right, has bypassed Oscar in my thinking both for the playoffs and just how bad the Royals were defensively every year

As I have said . . . wish I had gotten involved earlier though these are the players from before I really watched basketball so they are the ones that I am most open to information about.

[mopper8]

Well, I'm flattered that some people singled me out for thank appreciation. I was happy to contribute even as a non-voter. Thank you all for teaching me so much about the history of the game. I have a much fuller understanding of the game thanks to this project.

[raptorforlife88]

So I've started following what you guys were doing recently, and it's really impressive. Awesome job. I've got a question though, since I can't get through all 77 pages.

Are you guys weighting all the POY shares equally? Since there are far more players now and teams, there's obviously a far greater pool of players to earn POY votes. Whereas in a 9 team league there's much more limited selection. So is the weighting equal all the way through, or is there some sort of compensation?

[Manuel Calavera]

If we were voting on the top 50 players each season then you'd probably need compensation since the 50th best player in 1960 was likely on the bench while the 50th best player today is a starter and likely the second best player on his team but because we only ever voted on 5 there really isn't any problem.

[drza]

If we were voting on the top 50 players each season then you'd probably need compensation since the 50th best player in 1960 was likely on the bench while the 50th best player today is a starter and likely the second best player on his team but because we only ever voted on 5 there really isn't any problem.

Well, he has a point. It's not even, as we've discussed a few times now in the last few weeks. For example, in 2005 alone we had 13 players receive votes. There haven't been 13 players get votes total in the whole 1950s so far. The concentrated player pool/few teams definitely played a big part in why the old school guys that do get votes tend to get more votes than their recent counterparts. Nevertheless, there's not an obvious way to correct for that so, if it really bothers you, I'd suggest setting some type of threshold year and separating players into before and after. For example, ElGee used 1980 as his separating point and counted up his player shares before and after 1980 to have 2 different votes.

[raptorforlife88]

Well, he has a point. It's not even, as we've discussed a few times now in the last few weeks. For example, in 2005 alone we had 13 players receive votes. There haven't been 13 players get votes total in the whole 1950s so far. The concentrated player pool/few teams definitely played a big part in why the old school guys that do get votes tend to get more votes than their recent counterparts. Nevertheless, there's not an obvious way to correct for that so, if it really bothers you, I'd suggest setting some type of threshold year and separating players into before and after. For example, ElGee used 1980 as his separating point and counted up his player shares before and after 1980 to have 2 different votes.

Yeah that's probably the best idea, to seperate the era's. Otherwise like you said, the guys in the 1950's wind up looking a whole lot better as they're the limited set of guys to get continual votes.

[Doctor MJ]

Yeah that's probably the best idea, to seperate the era's. Otherwise like you said, the guys in the 1950's wind up looking a whole lot better as they're the limited set of guys to get continual votes.

I don't want to cloud the raw data with subjectivity when coming up with the initial summarization. Understand that by no means do I take this number to be any kind of definitive statement on player ranking, it's simply a baseline tool to use.

[mopper8]

I think nice baseline that the project doesn't necessarily have to include but for individual score-keeping would be to separate eras by a tripartite system - pre-ABA/NBA + ABA/post-merger

I don't know if you'd want to do it by player shares within each era, or by when a player was drafted, or when a player played in his prime, or what not. But that seems pretty clean to me overall - you have a specific event (the ABA) that played a role in both the dilution of talent and then the concentration of talent to help figure things out a bit. I dunno, just a thought.

[Doctor MJ]

I think nice baseline that the project doesn't necessarily have to include but for individual score-keeping would be to separate eras by a tripartite system - pre-ABA/NBA + ABA/post-merger

I don't know if you'd want to do it by player shares within each era, or by when a player was drafted, or when a player played in his prime, or what not. But that seems pretty clean to me overall - you have a specific event (the ABA) that played a role in both the dilution of talent and then the concentration of talent to help figure things out a bit. I dunno, just a thought.

When we do do a recap of the project, I do intend to show more data than simply total POY shares

[Manuel Calavera]

If we were voting on the top 50 players each season then you'd probably need compensation since the 50th best player in 1960 was likely on the bench while the 50th best player today is a starter and likely the second best player on his team but because we only ever voted on 5 there really isn't any problem.

Well, he has a point. It's not even, as we've discussed a few times now in the last few weeks. For example, in 2005 alone we had 13 players receive votes. There haven't been 13 players get votes total in the whole 1950s so far. The concentrated player pool/few teams definitely played a big part in why the old school guys that do get votes tend to get more votes than their recent counterparts. Nevertheless, there's not an obvious way to correct for that so, if it really bothers you, I'd suggest setting some type of threshold year and separating players into before and after. For example, ElGee used 1980 as his separating point and counted up his player shares before and after 1980 to have 2 different votes.

That's more a product of people just not knowing much about the 1950's and 1960's compared to the present.

[Doctor MJ]

That's more a product of people just not knowing much about the 1950's and 1960's compared to the present.

I think that's one important thing to realize.

I think another point to understand, when thinking about why so many less players got points in our project than in MVP voting, is frankly, how bizarre early MVP voting looks.

Number of guys receiving #1 votes:
'56: 8
'57: 12
'58: 7
'59: 9

'07: 3
'08: 4
'09: 5
'10: 3

So crazy, in '56-57, there were 8 teams, and 12 different guys getting #1 votes.

[ThaRegul8r]

If we were voting on the top 50 players each season then you'd probably need compensation since the 50th best player in 1960 was likely on the bench while the 50th best player today is a starter and likely the second best player on his team but because we only ever voted on 5 there really isn't any problem.

Well, he has a point. It's not even, as we've discussed a few times now in the last few weeks. For example, in 2005 alone we had 13 players receive votes. There haven't been 13 players get votes total in the whole 1950s so far. The concentrated player pool/few teams definitely played a big part in why the old school guys that do get votes tend to get more votes than their recent counterparts. Nevertheless, there's not an obvious way to correct for that so, if it really bothers you, I'd suggest setting some type of threshold year and separating players into before and after. For example, ElGee used 1980 as his separating point and counted up his player shares before and after 1980 to have 2 different votes.

That's more a product of people just not knowing much about the 1950's and 1960's compared to the present.

This. Common sense says that people are going to know more about the players who played during the time they watched basketball than the players that played before they started watching basketball, and before they were even born in most cases.

[mopper8]

While that does seem to make sense, is it really true? Was there really a whole host of guys who were out-playing Russell, Wilt, West, and oscar on a year-to-year basis, and we just don't see it because we weren't born then?

I just don't see it. "Common sense" also says that pro basketball is both more popular and more lucrative today, and as such has a wider pool of talent from which its drawn and see more parity in terms of top performers as a result.

Maybe Shaq and Duncan would have played no matter what, because height is always prized and sought out, but Dirk? Nash? Kobe? Wade? Gasol? Foreigners and small guys still playing in the league are not nearly as given IMO if the league was in the same place now as it was in the 60s

[Manuel Calavera]

While that does seem to make sense, is it really true? Was there really a whole host of guys who were out-playing Russell, Wilt, West, and oscar on a year-to-year basis, and we just don't see it because we weren't born then?

I just don't see it. "Common sense" also says that pro basketball is both more popular and more lucrative today, and as such has a wider pool of talent from which its drawn and see more parity in terms of top performers as a result.

Maybe Shaq and Duncan would have played no matter what, because height is always prized and sought out, but Dirk? Nash? Kobe? Wade? Gasol? Foreigners and small guys still playing in the league are not nearly as given IMO if the league was in the same place now as it was in the 60s

This is a very different argument than what was said earlier.

Earlier it was said that because there were fewer teams that there were less stars and therefore votes were much more concentrated. That's at least sort of rooted in truth, I don't agree with it though. Again if this was top 50 every year sure, players today need compensation because the 50th best player today is likely the second best player on a below average team, while the 50th best player in 1960 was deep on the bench. But with only 5 players we are talking about only the exceptions to the rule, and those aren't governed by things like the number of teams. It's far more likely that people simply don't know a whole lot about the 1950s-1960s and therefore the votes are a lot more concentrated because people are relying much more on others opinions (and black and white stats). There's also ALOT less bias because none of us were fans back then and therefore didn't have favorite players.

But that's just reiterating what I already said. Your new point is just as flawed though. Basketball is more lucrative today and it's very possible that many of the things involved in basketball has improved as the sport became more lucrative/popular like scouting, training regiments, advancements in statistics, medical advancements, etc. But I'm not convinced that players are being driven to the NBA because they can make more money now, like Kobe would have become an insurance salesman if he didn't get paid as much. The best of the best have always gone to "their" sport. Guys like Havlicek and Danny Ainge went to the NBA even though they would have done excellent in more popular leagues (not sure if the NFL was more popular than the NBA in the 60's....). As for the role players and the like, that MIGHT be true but there are 30 teams now as opposed to 8 back then and I'm not convinced the popularity in the sport has made up the difference. International players probably have, but it could go either way and the only fair thing to do is treat them as equally talented leagues.

[mopper8]

But I'm not convinced that players are being driven to the NBA because they can make more money now,

Really? That just strikes me as obvious. We had this discussion earlier, but I remain convinced that more people pursue basketball as an actual career both locally and world-wide than ever before. I'm shocked that some people disagree with this statement. Would anybody disagree with at least the fact that basketball's prevalence in Europe and South America has exploded in the last 20-25 years? Does that not count at all? I have a hard time buying that any of the foreign-born NBA players in today's league would've been in the NBA in the 60s.

And I'm curious as to how people would explain the obvious change in racial make-up of the NBA without a significant shift in the racial makeup of the US (ITOs African-Americans vs whites). That might not be the most pleasant thing to talk about, but its there.

[mopper8]

Just to hit on the racial integration a little, John Taylor:

While the NBA in general integrated more quickly than any other team sport in the country, the Celtics set the pace within the league. They were the first team to field a majority of black players, the first to field an all-black team, and the first team in any major-league sport to hire a black coach. These developments were greeted with incredulity, skepticism, and outright resistance from some fans and sportswriters in the racially torn city of Boston, but they contributed to the Celtics' dominance of the NBA during the 1960s.

There's more with more specific dates, but I need to look it up.