MVP: What Does Value Even Mean In NBA’s Most Prestigious Award


The NBA’s MVP is one of the most oft-debated and misunderstood awards in the online basketball world. It’s hard to see a reference to Derrick Rose, Steve Nash or Allen Iverson without some sort of swipe at the recipient.

But the fault isn’t with the winner, it’s with the notion that they didn’t deserve the award.

More from NBA

Kobe Bryant

never got robbed.

LeBron James

wasn’t punished for going to Miami. And voters did not suffer from fatigue in deciding to stop giving awards to

Michael Jordan


All of those storylines are re-written history by stats geeks intent on re-debating the awards by erecting strawmen.

Full disclosure, I am a stat geek. I love stats.

I’m the guy who is always defending that the “eye test” is more likely to lie than the stats. That said, stats really don’t mean everything, and the MVP doesn’t have to be the best statistical performer. That’s the flaw in re-arguing the debate after the fact. The things that the stats don’t show rarely end up in post-awards analysis.

Let’s begin with the most blatantly obvious and sadly necessary thing to say. MVP is an acronym for Most Valuable Player. What does value mean, and what goes into it? Google says it’s defined as, “The regard that something is held to deserve; the importance, worth, or usefulness of something.”

The important thing to focus on in that definition are words “held to deserve.” Value isn’t based on the quality of the thing, it’s the perceived worth of a thing. By definition, value is not objective; it’s subjective.

The idea that the “best” player is the most “valuable” player makes the assumption that quality is equal to value, and any good capitalist will tell you that’s folly. The best product doesn’t always win out. Just ask Betamax.

In the economics world, value is determined by supply and demand, i.e. what people are willing to sell or buy a thing for. It’s called the market, and it is god in capitalism. While there’s nothing to be bought or sold in terms of the MVP award, there is a kind of marketplace.

It’s similar in the sense that there is a collection of people (the media) through a type of exchange (discussing of ideas through writing, both online and in print, radio, TV podcasts et al) reaching a consensus on value through a transaction (the vote).

Consider the two houses below (h/t McKay Coppins). Guess which one costs more?

The house on the left is $459,000. The one on the right is $485,000.

We don’t know anything else about the condition or interior of the houses, but the eye test is telling you pretty quickly and easily the one on the left is just a lot-of-bit nicer than the one on the right. So why?

There are four things that go into the value of a home, and each has a parallel to the MVP. By considering each, we can understand how the “Roses” of the world sometimes win out over the “LeBrons.”

The four things are: quality, location, neighborhood and narrative.


The first measure of value is quality. And we can measure that statistically. There are two widely-used metrics which evaluate overall performance: win shares (WS) and player efficiency rating (PER). While each of those metrics can be argued to have flaws, they have their place. Furthermore some of the weaknesses each has is offset by the other.

I looked back since the NBA/ABA merger in 1977 and tracked where the WS and PER champs finished in the MVP race. I also tracked where the MVP ranked in both categories.

First, here is how the MVP compared in WS ranking with how the WS champ did in MVP voting.

With a few exceptions there are two clear trends. The WS champ almost always finishes in the top three in voting, and the MVP is almost always in the top three in win shares.

Now, let’s take a look at the PER champ.

While PER doesn’t work quite as well as WS, it still holds that the MVP tends to be near the top in PER. In 38 years the MVP winner has finished top four in PER 33 times. Surprisingly, all five times have come since the millennium flipped. (I would have thought with the increased focus on efficiency, it would have been the other way around).

There are a few exceptions on both fronts, but they’re generally explainable. In 1977-78 Bill Walton won the MVP in spite of only finishing 19th in win shares. That’s because he missed 24 games. He was second in both PER and WS per 48 minutes.

Steve Nash was the other severe outlier, but assists aren’t sufficiently rewarded in either advanced metric, and he was the league leader in that category both years he won, so he did have a statistical argument.

The first criteria is quality. While a player doesn’t have to be the best in the league, history shows that he needs to be arguably a top-five player.

Location, Location, Location

You know the three rules of real estate, right? Location! Location! Location! In the above analogy, the biggest reason for the difference in price is the house on the left is Plano, Texas. The one on the right is in Montclair, New Jersey.

And there is a very similar rule in the MVP voting. Standings! Standings! Standings. If you want to know why Kevin Garnett—the only player to win the WS and PER championship in the same year and not finish in the top-10 in voting—got so little respect in 2006, it’s because his Minnesota Timberwolves missed the playoffs.

Here is a look at how the teams of MVP winners have fared in the standings over the last 38 years:

Once again, we see a very common theme. Almost universally, the MVP winner comes from a top-tier team. In fact, 23 times the MVP winner has come from the team with the best record in the league, and nine times he’s been on the second-best. Only three winners have not boasted one of the league’s four best records.

The Neighborhood

Another aspect that impacts home values is the neighborhood. A negative and shameful example of this is the “white flight” which occurred during the 70s. A more positive example is when a homeowner adds an extension to his house, raising his property value. Because his value goes up, the surrounding homes do.

There is an inverse aspect of this effect in MVP voting. This is probably the most unrecognized impacts of MVP voting. “Help” matters to voters. In the last 38 years, 23 times the MVP did not have a single vote cast for a teammate.

There’s a logic to this. Consider these two scenarios. I had a roommate once who owned a beef farm. His only car was a pickup truck he got for around $25,000. Every day of the winter, he’d fill it up with silage (fermented corn which gave the cattle nutrition) drive it out to the herd, and feed them.

He also went to college and drove the truck to and from school. As we lived on the farm, and there was no public transportation, that was the only way he had to get to the grocery store, too. Or go out to the movies or anything else for that matter. In other words, that truck was wrapped up in his whole life. He needed it for everything.

By contrast, my grandmother kept a very nice luxury sedan as a second car that was probably worth around $40,000. She kept it exclusively for something us grandkids could drive around when we flew in for a visit. I guess she wanted to save us the trouble of renting a car. That’s the only way it ever got used. When she drove, she had her Caddy.

Anyway, which vehicle had the most value? The extra one which sat in the garage doing nothing most of the time, but cost more money or the one which was vital to a man’s way of life but cost less?

The chart below shows what kind of help MVP winners had over the years. Here’s the best teammate finish for every winner. I assigned a value of 25 (one greater than any teammate who received votes) for those who were their team’s lone representative.

Once again, history is pretty consistent. Of the 38 winners, 23 didn’t have a recognized teammate and another 10 failed to have a teammate rank in single digits. Only two had teammates finish in the top five.

The first pair was Michael Jordan with Scottie Pippen in their 72-win season of 1995-96 for the Chicago Bulls.

The other was Moses Malone and Julius Erving—the two previous winners combined. They led the 1982-83 Philadelphia 76ers to the “fo fo fo” championship.

Both teams were named among the 10 greatest teams in NBA history in 1997 and would probably still be today. You could argue both are top five.

There’s a bit of contradiction between this and the previous criteria. You need to have enough help to win, but not so much that your team can win without you.


The last aspect to consider is the dreaded “narrative.” In real estate the story of the house can make it go up or down in value. For example, if the previous owner was axed to death by Lizzie Bordon, you might get a discount. If Michael Jordan were conceived in it, you might drop a few extra dollars to won it.

Narrative became a dirty word after 2011, when Rose beat out Dwight Howard for the award (remember that James actually finished third). The general feeling by those who didn’t vote for Rose was that he, “just won on narrative.”

That’s not quite fair or accurate. When considering the criteria we’ve already outlined, Rose meets the normal standards. He was fifth in win shares, played on the best team and didn’t have a teammate who received a vote for MVP.

Howard’s Orlando Magic was the eighth-best team in the NBA, a full 10 games behind the Chicago Bulls. Remember that Jordan and Malone were the only ones who won that far behind, and they had historically good seasons. Howard’s wasn’t on that level.

James’ Miami Heat had the third best record, but Dwyane Wade finished seventh in MVP voting and the Bulls were one of the two teams who finished ahead of the Heat in the standings.

So, on both counts, Rose had an argument based on the conventional arguments but narrative was in his favor, too. And no, I don’t mean that he was a humble kid from Chicago making good. Nor was it voters punishing James for leaving Cleveland. That’s the disingenuous spin put on the narrative slant.

Nov 21, 2013; Denver, CO, USA; Chicago Bulls guard Derrick Rose (1) drives to the basket during the second half against the Denver Nuggets at Pepsi Center. Mandatory Credit: Chris Humphreys-USA TODAY Sports
Nov 21, 2013; Denver, CO, USA; Chicago Bulls guard Derrick Rose (1) drives to the basket during the second half against the Denver Nuggets at Pepsi Center. Mandatory Credit: Chris Humphreys-USA TODAY Sports /

Rose won because he had Keith Bogans as his starting shooting guard, was missing the Bulls’ second-leading scorer Carlos Boozer for 23 games and Joakim Noah for another 34, and still led his team to the NBA’s best record. Narrative did work in his favor—he did more than James with less around him—but he didn’t need narrative to win. He won based on the same sort of things as previous MVPs.

Other winners who are unfairly labeled as winning based on narrative include Karl Malone (first in standings and PER, second in WS, and Stockton was just 15th voting) and Charles Barkley (fourth in WS, first in standings and no teammate’s received votes).

The point is that just because narrative is given as a reason by people who don’t agree, it doesn’t mean that it is the reason for people who voted. Rose, Malone and Barkley all met the conventional standards.

Kobe Bryant’s “career achievement award” in 2008 is another one that’s often floated as a win for narrative. But Bryant finished fourth in win shares, his team was third in the standings and didn’t have a teammate get a vote for MVP.

The two teams that finished ahead of the Lakers, the Boston Celtics and Detroit Pistons, did not boast a viable candidate. Chris Paul, who led the league in win shares finished a close second in the voting. It’s probable that there was some “career achievement” factored into the voting but not as dramatically as some would poise.

Allen Iverson won in 2000 having met just two of the criteria. He was seventh in PER and 10th in WS, but he was also on the league’s second-best team and didn’t see any teammates get votes. He was the league’s leading scorer and the way he got those points—recklessly bouncing through the lane like a pinball crashing off bumpers and ringing up points while doing so—left an indelible impression. Narrative was a factor, but not the only factor.

If you’re looking for someone who won almost uniquely on narrative it would be Steve Nash’s two awards. He was not a top-five player either year. He had a teammate finish in the top-10 in voting in 2004-05 and he failed to lead his team to a top-three finish the following year. That makes him the only player to win without meeting at least two of the criteria, and he did so twice.

So what explains his awards? First, in both years he led the league in assists, and that counts for something.

In 2004-05, he was the principal difference in the Phoenix Suns going from 29 to 62 wins, the best record in the league. In 2005-06, he led the Suns to a division title in spite of losing three of his top scorers (Joe Johnson, Quentin Richards and Amar’e Stoudemire) to free agency or injury.

Further complicating the picture is that in 2006 there was no clear frontrunner. James finished second, but his Cavaliers were only sixth and didn’t win their own division. Dirk Nowitzki led the league in win shares, and his Mavericks had the third-best record, but they were also second in their division. Bryant’s Lakers were only 10th. Chauncey Billups with the first-place Pistons also received 15 first-place votes, but his numbers didn’t scream MVP, either.

Nash won, but with a share of just .739, per Basketball-Reference, an unusually low percentage (though Karl Malone’s .701 was worse in 1999).

So what’s the point of all this? Narrative never uniquely decides the MVP. There are times when it factors in, though, particularly when there is no obvious winner.

What is the Deciding Factor?

So which of these factors matters most? That’s where it gets really interesting. Let’s look at each of the criteria from a different angle.

Since WS had a better predictive value. Here’s how the MVPs finished in Win Shares by placement.

Over half the winners were first in win shares, but 42 percent weren’t. Being the “best” player helps but it by no means guarantees winning the MVP. However, 90 percent of winners were in the top five.

Now here’s the pie chart for the standings.

Sixty-one percent of the time the winner is on a first place team, and 87 percent of the time, he’s s on a top three team (with only one of those occurrences being third).

Finally, let’s look at the teammate factor.

MVPs didn’t have a teammate who received a vote 61 percent of the time and one who was no better than 10th 26 percent of the time.

Only seven winners led the league in WS, were on the best team, and didn’t have a teammate who received a single vote for MVP, meeting the most-strict standards. However, 28 of the winners were top five-players on top-three teams without a top-10 teammate, and as previously stated, only Nash failed to meet at least two of the criteria.

Value is determined evenly between each of the three standards, and when it’s too close to call, narrative comes in as the tie-breaker. It goes neither to “best player” or “best player on the best team.” If you’re looking to reduce the meaning to a phrase, “Top-five player on a top-three team, who had enough help to win, but not from a star, and it doesn’t hurt if he has a good story” seems to fit the bill, but it doesn’t quite roll off the tongue.