Understanding NBA MVP Voting Parameters

[HotRocks34]

So, recently I took the time to research the threshold for how many games an NBA player is historically expected to play in a non-lockout season and still be able to win a regular-season MVP award. I put that information here on RealGM:
http://bit.ly/1rphxeM

About 15 hours after I made that post, a similar post popped up on Pro Basketball Talk. This piece:
http://bit.ly/1sHfTuS

Which I pointed out on RealGM:
http://bit.ly/1x2OGmx

After I made that last post, an apparently credulous -- some might prefer the phrase "apparently gullible" here -- Br0ken_Sp0ke informed me that I was incorrect, after having done his own research:
http://bit.ly/1x1l61N


All of this is neither here nor there, however. Whether or not NBA bloggers and media are scanning RealGM for ideas (and I 100% feel that they are, as they should), information will flow on the internet and the good minds at RealGM are, at the very least, thinking along the same lines as those bloggers and NBA media, which is a good sign for how high the quality of thinking is at RealGM, IMO.

With all of this in mind, I did some more research on the history of the NBA MVP voting. Though I have not done any Google searches to check if this information is out there on other sites, blogs or media, my guess is that it is not. At least not in this form or packaging. So, as I alluded to with the "Durant no shot at MVP" post cited above, my guess is that you saw the forthcoming information in this post here at RealGM first.

*****

Everyone knows that an NBA MVP award winner has to have a good year, almost certainly has to play on a good team, and has to play a significant amount of games to win the award. But what, exactly, does the quantification of "play on a good team" mean? That's what I wanted to find out in my latest research, after having tracked down the "how many games does an MVP candidate need to play?" data. Or to put it another way, as mentioned in the title of this thread:

What are the quantified team parameters for a serious MVP candidate as shown by NBA history?

If we know this information, and we combine it with the "minimum games played for a serious NBA MVP candidate," then we probably can have a much better predictive mechanism for determining who might actually win future NBA MVP awards.

Here's what I found in the new research.
"NLY" = Non-Lockout Year


Times that a person on team that not make playoffs won MVP -------> 01 [1975-76]
------------------------> Never happened in the NBA's Modern Era (1979-80 onwards)
Times person on losing team won MVP -----------------------------------> 02 [1975-76 is last time]
------------------------> Never happened in the NBA's Modern Era
Times person on 3rd or lower Conference seed won MVP --------------> 07 [2005-06 is last time]
------------------------> 4th seed in Conference is lowest-ever for MVP winner
------------------------> Last time player from 4th-seeded Conference team won MVP was in 1981-82
Times person on non-50-win team won MVP in NLY --------------------> 09 [1981-82 is last time]
------------------------> Twice in the Modern Era
Times person who plyd fewer than 70 games won MVP in NLY --------> 03 [1977-78 is last time]
------------------------> Never happened in the Modern Era


Here is what all of this information taken together means for a player who wants to win an NBA regular-season MVP in the NBA's Modern Era, given the history of NBA MVP voting:

* You will need to play at least 70 games in the regular season
* Your team must have a winning record
* Your team must make the playoffs
* Your team almost certainly will need to win 50 games in a Non-Lockout Year
* Your team almost certainly will need to be seeded 3rd or better in your Conference


Taking this historical information, we can apply it to the current season and examine the potential MVP candidates and those who likely can be checked off the list at this time. Also, we can keep an eye on the key historical team metrics that potential candidates are likely to have to meet.


PLAYERS LIKELY OUT OF THE MVP DISCUSSION
=======================================

Kevin Durant ----------> Will not play enough games to qualify
Russell Westbrook ----> Will almost certainly not play enough games to qualify


QUESTIONS TO WATCH FOR OTHER POTENTIAL MVP CANDIDATES
=======================================================

Derrick Rose -----------> Will he be able to stay healthy enough to play enough games to qualify?
Anthony Davis ----------> Can NOP make the playoffs? Win 50 games? Get 3rd or better West seed?
Carmelo Anthony ------> Can NYK make the playoffs? Win 50 games? Get 3rd or better East seed?
James Harden ----------> Can the Rockets win 50 games? Get 3rd or better West seed?


In addition to helping predict future MVPs, knowing these historical NBA MVP parameters helps to explain or settle past controversies or disputes about awards that were won or not won. Since this post is already more than long enough, I will give an example of what I mean in the next post I put in this thread.

[HotRocks34]

Now I want to take a look at how quantifying historical NBA MVP voting can help end disputes from the past as regards MVP awards won and not won. Once again, here is the general criteria that the MVP voting tends to follow in the NBA's Modern Era (1979-80 onwards):


* You will need to play at least 70 games in the regular season
* Your team must have a winning record
* Your team must make the playoffs
* Your team almost certainly will need to win 50 games in a Non-Lockout Year
* Your team almost certainly will need to be seeded 3rd or better in your Conference


One point of discussion that has been had in the past has been about the MVP awards Kobe Bryant has not won, specifically those around the years of 2004/05 to 2006/07. Looking at the team MVP criteria above, it is relatively easy to find "reasons" (or at least explanations given the historical data) why Kobe was not in a position to win the award during those seasons. I found one data point that applied to all three seasons (2004-05; 2005-06; 2006-07):

* Your team almost certainly will need to be seeded 3rd or better in your Conference

In none of the three mentioned seasons were the Lakers seeded even 4th in the Western Conference. Thus, using the unearthed historical data, it would not be surprising to see that Kobe did not "qualify" for the MVP award in those seasons.


As I have said here before on RealGM, I am an Old School guy who grew up on The Eye Test and Simple Boxscore Stats. However, I have found that the more I dig into NBA research -- both analytical research and just simple historical data mining as I did for this NBA MVP project -- the more it rewards me and the more certain patterns emerge. And the more things make sense in terms of what has, and has not, happened in the league.

[MixedUp]

I would go with the following top-10 right now!
[youtube]https://www.youtube.com/watch?v=J7SmGxYgcJQ[/youtube]

[dautjazz]

People are overreacting on the Cavs start.

Knicks game - Return of Lebron to Cleveland, matchup with Melo, first game for the big three. Basically lots of pressure, and lack of team chemistry, there was a big chance of something like this happening.

Blazers - Very good team, anybody can lose to them on any given night.

Jazz - 2nd night of back to back, and for whatever reason, Utah has Lebron's card. Sudiata Gaines, Millsap, now Hayward.

On the other hand they beat the Bulls, Nuggets and Pelicans, so solid wins. They'll be fine, already are 8th, and the 2011 Heat which made the finals were 11-11 after 22 themselves, so this is no worse a start. I think Lebron has to be the favorite unless New Orleans gets 50+ wins, then I can see Davis getting it.

[Texas Chuck]

Is this post about MVP Voting Parameters or is it about you being first?

Because while maybe nobody has ever taken the time to specifically lay out this information in this exact format, I think a ton of people already know that the MVP needs to play the majority of the games and their team needs to be a contender.

Really nothing particularly noteworthy or proprietary here(or in your previous post).

[DayofMourning]

Awesome posts HR!

[The Realist]

I agree with your premise but if there is one scenario which can break the rule is the one that might be occurring with the Thunder.

1) With Westbrook and Durant out, Thunder drops to the 2nd worst record in the West with 6-14/5-15.
2) Westbrook and Durant both return at the same point (haven't been following when they should come back) after 20 games played
3) Westbrook plays like a monster throughout the season but all of his positive impact is credited to the Durant MVP campaign
4) Durant loses very little from the last RS but that is compensated by him being babied even more by the refs, produced a ridiculous stat line of 32/7/4 with 62 TS%
5) They lead the Thunder to 50-52 wins in the last 62 games to get the 3rd-5th seed in the West, after being down and out quarter into the season.

That kind of turnaround could get Durant the MVP, especially if Lebron isn't playing at quite his top level. There's even room to adjust the Thunder's record to be better before they return and Durant/Westbrook to return earlier (I think).

I'm not saying it's likely (far from it), but the table is "perfectly" set for the anomaly to happen.

[HotRocks34]

Because while maybe nobody has ever taken the time to specifically lay out this information in this exact format, I think a ton of people already know that the MVP needs to play the majority of the games and their team needs to be a contender.

Someone who is waiting to have General Board rights (I'm not sure if he wants me to give his real name so I won't) wrote me after I made this thread and told me that the outline I made had already pretty much been done, which I did not know:

http://www.82games.com/wilq.htm

But Wilczynski's post had even more information than my own. Great work by him.

The point of my saying "this hasn't been written before" or whatever I said in the first post was not to grab credit for myself (which I obviously can't claim after looking at Wilczynski's work); the point was to illustrate that the RealGM forums have, and are, producing good work. Work that people in the NBA media are almost certainly looking at, IMO. I gain nothing from saying that something that came up on RealGM first is then seen elsewhere on the internet. If anything, the point is to promote the site.

My pointing that out, about RealGM potentially influencing others, was because of the discussion that Broken Spoke and I had previously which I referenced in the first post of this thread.

[HotRocks34]

Awesome posts HR!

Thanks bro!

[HotRocks34]

I agree with your premise but if there is one scenario which can break the rule is the one that might be occurring with the Thunder.

1) With Westbrook and Durant out, Thunder drops to the 2nd worst record in the West with 6-14/5-15.
2) Westbrook and Durant both return at the same point (haven't been following when they should come back) after 20 games played
3) Westbrook plays like a monster throughout the season but all of his positive impact is credited to the Durant MVP campaign
4) Durant loses very little from the last RS but that is compensated by him being babied even more by the refs, produced a ridiculous stat line of 32/7/4 with 62 TS%
5) They lead the Thunder to 50-52 wins in the last 62 games to get the 3rd-5th seed in the West, after being down and out quarter into the season.

That kind of turnaround could get Durant the MVP, especially if Lebron isn't playing at quite his top level. There's even room to adjust the Thunder's record to be better before they return and Durant/Westbrook to return earlier (I think).

I'm not saying it's likely (far from it), but the table is "perfectly" set for the anomaly to happen.

The person who made the PBT article on this subject said that the best scenario for Durant to win the award this year would be if LeBron (or whomever is at the top of the MVP race) also missed some time, as Durant is doing now. That was the apparent scenario for Walton winning the award in the 1970's when he played just 58 games.

I think that is probably correct. But you never know, perhaps your scenario will come true! History can't fully predict the future, particularly when the MVP voters are subjective and can choose whatever criteria they want.

[YogurtProducer]

Due to the conferences being completely unbalanced, I'm not sure if top 3 in conference is the best way to look at things. Where did those teams rank overall in the league? Top 5 im gonna guess?

[HotRocks34]

Due to the conferences being completely unbalanced, I'm not sure if top 3 in conference is the best way to look at things. Where did those teams rank overall in the league? Top 5 im gonna guess?

In the modern NBA (1979-80+), the lowest a team that produced an MVP has finished in the overall NBA standings was 7th. This happened twice.

1981-82 Houston Rockets ------ 46-36 ------- Moses Malone, MVP
1987-88 Chicago Bulls ----------- 50-32 ------- Michael Jordan, MVP

[KD35Brah]

I agree with your premise but if there is one scenario which can break the rule is the one that might be occurring with the Thunder.

1) With Westbrook and Durant out, Thunder drops to the 2nd worst record in the West with 6-14/5-15.
2) Westbrook and Durant both return at the same point (haven't been following when they should come back) after 20 games played
3) Westbrook plays like a monster throughout the season but all of his positive impact is credited to the Durant MVP campaign
4) Durant loses very little from the last RS but that is compensated by him being babied even more by the refs, produced a ridiculous stat line of 32/7/4 with 62 TS%
5) They lead the Thunder to 50-52 wins in the last 62 games to get the 3rd-5th seed in the West, after being down and out quarter into the season.

That kind of turnaround could get Durant the MVP, especially if Lebron isn't playing at quite his top level. There's even room to adjust the Thunder's record to be better before they return and Durant/Westbrook to return earlier (I think).

I'm not saying it's likely (far from it), but the table is "perfectly" set for the anomaly to happen.

What a pathetic post.

Wouldn't expect anything less from a Lebron stan.

[The Realist]

I agree with your premise but if there is one scenario which can break the rule is the one that might be occurring with the Thunder.

1) With Westbrook and Durant out, Thunder drops to the 2nd worst record in the West with 6-14/5-15.
2) Westbrook and Durant both return at the same point (haven't been following when they should come back) after 20 games played
3) Westbrook plays like a monster throughout the season but all of his positive impact is credited to the Durant MVP campaign
4) Durant loses very little from the last RS but that is compensated by him being babied even more by the refs, produced a ridiculous stat line of 32/7/4 with 62 TS%
5) They lead the Thunder to 50-52 wins in the last 62 games to get the 3rd-5th seed in the West, after being down and out quarter into the season.

That kind of turnaround could get Durant the MVP, especially if Lebron isn't playing at quite his top level. There's even room to adjust the Thunder's record to be better before they return and Durant/Westbrook to return earlier (I think).

I'm not saying it's likely (far from it), but the table is "perfectly" set for the anomaly to happen.

What a pathetic post.

Wouldn't expect anything less from a Lebron stan.

How so? Both of those points are likely. If Westbrook came back and played like a borderline top-5 player, it would only boost Durant's narrative if he played better than Russell, because the turnaround in the teams results would be even more significant and because of the widespread narrative that Westbrook is holding Durant back.

Certainly you can't be arguing against Durant getting ridiculous amounts of ref assistance?