clock menu more-arrow no yes mobile

Filed under:

Why the MVP Award is Idiotic, Imbecilic, and Irrelevant

Well, it's that time of year again.

It's time for sportswriters to adamantly explain why there are only two candidates that deserve "real" consideration. It's time for bloggers to vehemently decry the big market bias of the media, and to lament the fact that their small market stars get no airtime. It's time for fans to chant those three letters every time their guy shoots a free throw.

It's time for the Race to the MVP.

But first, a little aside.

Question: What gives sports value?

Answer: The rules.

A sport is only as valuable, as fun, or as exciting as the rules that define it. Basketball would be ridiculous if there was no out of bounds, or if there were no foul calls, or if there were no salary cap rules. Beyond the rules that are written in ink, sports also have unwritten rules.

In recent times, we've seen a cardinal unwritten rule- "Thou shalt not cheat"- broken in the world of baseball, in the form of steroids. We've seen what a tumultuous era that whole sport has been tossed into, on the basis of one rule being broken. We've seen College Football ridiculed for its completely rule-less selection system process.

A sport without rules has no value. In fact, that statement applies not only to sports, but everything in life. When a "winner" can be judged based on five, six, seven, eight different criteria, the term "winner" ceases to hold much meaning.

Re-enter the NBA MVP award, stage left.

I talked to an NBA columnist from Charlotte (with an MVP vote) last year. My question to him: "What exactly are the guidelines for the vote? As in is there any technical language they use or phraseology you might pass along?"

His answer: "No guidelines. "Value" is in the eyes of the beholder."

But I'm not just going to sit here and say that the MVP is a meaningless award and be done with it. I mean, I could. I did that last year. No, this time I'm going to explain why it's so stupid. Specifically, I'm going to blast every MVP "criterion" to smithereens. An ambitious goal, you say? Given how fantastically dumb these criteria are, it's actually a pathetically easy thing to do.

#1: If you take that guy off the team, how good would the team be?

I think I've heard every TV guy say this one. JVG, Mark Jackson, Mike Breen, Reggie Miller, they're all in love with it. It sounds so deliciously simple too. I mean, if L.A. would win 50 games without Kobe, and Cleveland would win 30 without LeBron, LBJ is clearly better, right?

Wrong. Let's take this hypothetical example. In 2030, the Kansas City Dudes win 20 games. In 2030, the St. Louis Fellows win 30 games. That draft, the Dudes draft Chuck the Chap and the Fellows draft Larry the Lad. Through a few minor miracles, both the Dudes and the Fellows play the exact same schedule in 2031, with the exact same players, against the exact same opponents... exact same everything. Except now the Dudes have Chuck and the Fellows have Larry. The Dudes now win 40 games (20 more than last year) and the Fellows win 70 games (40 more than last year).

The imbecilic commentators du jour- Hedgie Giller and Jark Mackson- now have a foolproof method of determining "if you take that guy off the team, how good would the team be?" The Dudes were clearly worse (20 wins) than the Fellows (30 wins) if you took away those players.

But from an absolute sense, Larry added 40 wins to the Fellows, while Chuck added just 20 wins to the Dudes. As Hedgie and Jark pronounce Chuck the MVP, they unfairly penalize Larry for having better teammates.

That's the critical part here. The above is an idealized example, because you can never fully isolate a single player's performance. But whether the situation is idealized or not, the above flaw always exists. The player that has better teammates will always be somewhat unfairly penalized compared to the guy whose teammates suck.

Apply this to real life. If you took Kobe off the Lakers, they'd still be a decent team with All-Star Pau Gasol leading the way. If you took Wade off the Heat, they would suck, because they don't have Pau Gasol. Essentially, if you support Wade over Kobe by the "if you took him away..." logic, you're penalizing Kobe for something he has ZERO control over- having Gasol on his team.

A rebuttal to this would be the questioning of what "value" actually means. Is value meant in an absolute sense or should it be expressed as a percentage of total team value? For example, if Kobe was worth 5 WinnerPoints (a made up stat) and all other Lakers combined were worth 5 WinnerPoints total, Kobe would be 50% "valuable." If Chris Paul was worth 6 and all other Hornets were worth 2 total, Chris Paul would be 75% "valuable." Ultimately though, Chris Paul can only determine what he is worth. He can't force the rest of his team to be equal to 5 (more on this below). Defining value in absolute terms rather than relative ones confines the MVP conversation to a single player, rather than players.

At the end of the day, an MVP is the Most Valuable Player, not the Most Valuable Player with Good Teammates. To base it off his teammates is illogical, and it's especially silly to base it off the job his GM did or did not do in the summer.

#2: Best Player on the Best Team

Also known as the laziest possible way to pick an MVP. "Best player on the best team" is a perfectly reasonable criterion... when you're playing tennis. It's so funny to me how the same voters that slam "stat geek stats" as being too individualistic and non-team centric would ever resort to this. And yet, every year, they do.

The NBA is star-driven, no doubt about it. You've got the Iversons, the Bryants, the Shaqs, the LeBrons, and on and on. But in recent years, we've seen so, so, so many examples of good teams built on something other than star power. Boston's defense in 2008. Detroit's balanced offenses in the mid 2000's. As a lesser example, the extremely well coached Orlando teams of SVG that extract every last drop of defense from guys like Hedo Turkoglu or Rashard Lewis.

There's more than one way to skin a cat; there's more than one way to be the best team in the league. To automatically assume that the best team in the league must necessarily also possess the best player in the league is to completely discount the effects of teammates, coaching, schedule, league, fans, and a plethora of other factors.

It's equal parts ignorance and laziness, masquerading as intelligence.

Corollary to #2: Sub-50 game winners need not apply

This is another version of "best player on best team." It's basically saying, if you didn't win this many games, you're inherently not MVP worthy.

Again, basketball is a team sport. There are five players on the court at any given time. If LBJ is out there with the Beatles, his team will suck. If Kobe is out there with CP3, Wade, Duncan, and Superman, his team will be awesome.

I've seen a rebuttal to this logic, specifically from David Sparks ("The Arbitrarian"). He's claimed in the past that my Beatles vs. 4 All-Stars argument would be a stupid example, since there's a basic talent level required to play in the NBA. I can't remember his exact quote, but it went something like "a player isn't a true MVP candidate if he can't lead a team made of the World's 300 best players to a winning record."

My response: No! Absolutely no! There's a HUGE, HUGE discrepancy in talent level, within the NBA itself. Lumping the Hornets' bench with the Lakers' bench just because "they're both among the best ballplayers in the world" is ridiculous. Ignoring the enormous production efficiencies of NBA players- as measured by Hollinger, Berri, Oliver, or any person with functional eyes- is idiocy. And not only is the lumping together of players stupid, it totally ignores team defense, which is heavily dependent on coaching.

Choosing an MVP just because his team won more games is wildly illogical. While Criterion #1 penalized a player for having better teammates, Criteria #2 and its corollary penalize a player for having worse teammates.

#3: Makes his teammates better

This is the primary way that supporters of Criteria 1 and 2 can weasel their way out of criticism. It usually goes something like this: "Well, an MVP would make his teammates better and elevate their play to at least 50 wins."

Study after study has shown that, unless you're physically assisting your teammates and setting them up with dunks, there is no such thing as "making your teammates better." Doesn't exist. David Berri has shown that, if anything, the best players reduce the production of their teammates by using up additional possessions and opportunities. There is no such thing as inspiring your teammates into playing better, or motivating them to be the best that they can be, or whatever hackneyed cliche you want to throw out there. 

The aspects of "teammate improving" that are relevant also happen to be measurable and are reflected in various statistics. And, just for the sake of argument, if there was a player that miraculously made his teammates much better, there should be a disparity between that player's individual composite metrics (let's say PER for the sake of debate) and his adjusted +/-. The top 3 PER leaders? LBJ, Wade, CP3. The top 3 +/- leaders? CP3, LBJ, Wade.

Making your teammates better is one of those things that sounds good. It's comparable to "playing the game the right way." However, its relevant aspects are already measurable, and its irrelevant aspects are, well, irrelevant.

The Rest

In general, I think most other criteria are just derivations of the three above. "Leadership", another that is bandied about, is really a form of #3. Steve Nash's first MVP win was a specialized case of #1 where voters didn't take into account other factors that changed/improved simultaneous to his arrival, and his second MVP win was fueled by an undervaluing of the team with Amar'e gone (again #1) and elements of #3. Every once in a while you get writers arguing that Player X is "due"... I don't think I need to explain why that's stupid. Finally, I consider poor team records without Player X inconsequential, because they generally deal with miniscule sample sizes on the order of 5-10 games at most.

There is one special argument I should address though: "If you were starting a team, who would you take first overall?" It's often phrased as "who do I want taking the last shot" or something similar.

Overall, this judgement misses the point. The NBA MVP award is about who was most valuable, over the course of a given season. It's about who produced the most over an 82 game stretch. It's not about who the best (ie, most skilled) basketball player is. I've addressed this before, and it's a really subtle difference. The best basketball player may or may not produce the most value, because player performance is not static. Performance ebbs and flows over games and over seasons. The best player is the player that inherently has the most ability. The most valuable player is the player that translated his ability into the most production and therefore, wins. Again, it's a very subtle difference that depressingly few people understand.

In essence, the MVP voting process has freed voters to come up with whatever systems they find suitable. In turn, MVP voters have gone out and found some of the most indefensible ways of voting for players. It's tough to say who is more at blame, but the result is a disaster: questionable methodologies utilized to determine extremely vague "winners."

In a word, it's chaos. Left without boundaries, voters have devised their own sets of criteria, which they neither follow consistently nor have any logical explanation for following in the first place.

Some chap is going to win the MVP this year, and everyone will scramble to either defend or slam the selection with one of the above three ideas, and there will be a big commotion. "Oooh, Chris Paul not being considered is so unfair!" and "LeBron is just about the stats!" and "Miami would win 10 games without Flash!"

And none of it will matter.

It has no rules. It has no value.