They both deserve it.
That's the first thing I want to stress as we begin a deep and thorough conversation about the 2019 NBA MVP race. There is a threshold that an MVP has to meet, and after that it's about parsing through the worthy candidates to figure out the most worthy one. The nightmare scenario is when there aren't really worthy candidates, but someone has to win. That's not the case this year.
Both Giannis Antetokounmpo and James Harden are deserving of the MVP award. Both guys pass the test for a worthy winner — Paul George is right on that threshold. Their respective fanbases will gnash their teeth and throw things if their guy doesn't win, but, honestly, it won't be any sort of sports tragedy. Both guys deserve it; one guy's going to win.
So which one? How do you choose? In 2017 — the toughest MVP race I've ever covered — I broke down the relevant questions then. I decided to bring back the approach for this year's showdown between the Beard and the Greek Freak.
How do you define "valuable"?
This is, of course, the heart of the thing. Don't think for one second there's a single definition to this. The league keeps it vague, precisely because it prompts so much debate on a daily basis. These endless debates about how to define "most valuable" fuels the controversy and discussion with zero cost. Think of all the ways you can align this question:
Is it how much your team needs you?
Is it best overall player?
Is it most impactful?
Is it best player on best team?
There are all sorts of paradigms you can use to define it, but you have to decide on one first. Everything else you need to figure out about which candidate is the right one goes from that point.
You don't have to keep the same standards year over year, either. Each season is a microcosm. If a team wins 65-plus games, that might be enough to lean towards the best player on that best team. If a player crosses a statistical threshold never before matched, that can swing things in his favor. The important thing is to start there.
How much does the legacy of the award matter?
What I mean by this is how much stock do you put into what happened when previous winners were judged by certain standards.
For example, Russell Westbrook won in 2017 with the Thunder having won just 47 games. While players have gotten MVP with fewer wins — Kareem Abdul-Jabbar won it with only 40 wins for the Lakers during the 1975-76 season — that's well below the standard that's been set. Does it matter to you that historically the player has had to win more games? Does it matter more to you that Westbrook broke that trend in recent history?
Personally, I don't regard past voting trends at all. For starters, the voting bloc has evolved substantially over the past 10 years. More analysis, better metrics, more context and more shared knowledge change how the voters make their decision. Holding on to standards that predate these evolutions just seems flawed.
That's not to say there aren't still narrative voters, because there are. A lot of times they're local media that don't have the time or access to have a fully formed opinion, and in that case the "story" of a season outweighs other factors.
There's also an argument to be made on the merits of this, that too much analysis comes from a place lacking in full understanding gained from day-to-day coverage of the team, and that a 10,000-foot view is the only one with actual clarity. However, it's going to lack in evidence and sustained reason.
This is separate from how much wins matter, which we'll get to later.
How much does offense-defense balance matter?
Let's be honest: Defense has not mattered in the case of this award historically. Defense is usually just a sweetener in the deal.
But given what we know now about how important defense is on both an individual and team level, how much does it matter today?
Do you need to have a good defensive profile, or just not a bad one?
Let's start here:
Defensive Rating is a team metric; it's not based on any individual player's performance. However, at the very least, it reveals how a team defends when that particular player is on the floor.
Harden fans will let you know that he leads the league in deflections (just ahead of George) and that he is 88th percentile in defending the post. But the Rockets defense has been an absolute mess this season and is absolutely part of why they've slipped from being the top team in the West last season. Harden's played a part in that.
Now, to really figure out Harden's defensive impact you have to get into the weeds, and we'll do that later this month when we look at his candidacy in detail. But you have to watch the games and evaluate Harden's defense for yourself while considering all these metrics and more.
Meanwhile, George is considered to be a leader for Defensive Player of the Year. Antetokounmpo is right there with him. They're both leading high-level defenses. Does that matter? Should it matter?
What's your wins threshold?
Pretty self-explanatory. There have been considerations in the past that it doesn't matter how many wins your team has — it's how you play. (This is currently known as the Anthony Davis Postulate.) I've always appreciated this as an academic exercise, but I think it's a little unrealistic. The whole point of professional sports is … well, OK, it's money. But the second-most important goal of professional sports is winning.
You can't win Most Valuable Player if your team doesn't win. You didn't impact your team in a meaningful enough way, or you didn't make your teammates better, or your contributions weren't what your team needed, etc. There's no serious paradigm for the definition of "most valuable" that allows for wins to not matter.
However, you can definitely argue about how many wins you need. Kareem won with 40. Westbrook won with 47. Oftentimes, you need the most wins. Typically you need at least 50. This is also contextual, though. For example, no team is going to win 65 games this season.
You can argue that's a sign of team strength — and that Giannis' case as the candidate with the most wins is mitigated by that — but it's also a sign of league strength. Fewer teams are tanking this season and the parity is better, so the wins are more evenly distributed. The league may still be top-heavy, but not nearly as much as in previous years.
I wind up going with 50 wins as a prerequisite. I'm willing to consider outliers, but that's the base level for consideration. Of note, the Rockets are on pace for just over 50, the Thunder are on pace for just under 50 and the Bucks are on pace for just over 60.
Is it more important how much better a player makes his teammates, or how good he makes his team?
This gets to my favorite question of all, because it directly addresses the two ways to evaluate impact.
Again, Net Ratings and on/off numbers don't tell the whole story, but they provide a good torch to light the path.
We'll just move past that figure for Harden entirely and focus on the latter two. Without George, the Thunder get smacked. They are outright garbage when he isn't on the floor, even with Russell Westbrook.
However, Antetokounmpo's on-court figure is significantly higher than George's. It's actually higher than George's by more than Harden's on-court Net Rating.
The key here is do you care more about how bad a team is without a player or how good they are with him?
I lean towards the latter.
I don't want to punish a player for having good teammates or reward a player for having bad ones. It shouldn't be part of the discussion. If you raise your team's play, you'll be good enough. Teammates also look better when you play better. For example, there was an idea in 2017 that the team around Westbrook was poor and it was ridiculous. Steven Adams, Domantas Sabonis and Victor Oladipo look wildly different today.
You need to be able to define what "making his team better" means. It can mean improvement or top level, but you have to be able to discern and define between the two.
How much does historical statistical significance matter?
How about statistical thresholds? If Westbrook won at least in part because of his historic triple-double average (on top of leading the league in scoring, a fact that often gets omitted), then should other "never before done" marks carry as much weight?
Points per 100 possessions still isn't in the mainstream for sports fans, but it is the best way to evaluate production because it isn't weighted by pace or minutes. In 1987, Michael Jordan averaged 37.1 points per game playing 40 minutes per game.
Harden is averaging 36.6 points per game this season. When you control for minutes load and look at per 100 possessions? Harden is having one the most productive scoring seasons ever.
I should note that possession estimates are not available for the league's early history, so all of Wilt Chamberlain's numbers can't be evaluated. But Harden is making history this season.
You need per-game figures, though? How many players have ever averaged 35 points per game with 7 assists per game? You guessed it: just Harden.
But hold on. How many players have ever averaged 27 points, 12 rebounds and 6 assists per game with a steal and a block per game? No one but Giannis Antetokounmpo. (Note that steals and blocks were not tracked in Oscar Robertson's day.) So you can pull the "No one's ever done this!" card with Harden, but Giannis backers get to plug the same.
The fact that you can distill Harden's candidacy down to "Harden is averaging the second-most points in the past 50 years while averaging 7 assists per game, and no one has ever done that" is a pretty strong argument.
However, Giannis can bring his own "No one's ever put up these numbers" argument to the table. George is the one that's lacking … and yet, only one other player (you guessed it: Jordan) has ever averaged 28 points, 8 rebounds, 4 assists and 2 steals per game.
There is a feeling that this season, offenses are "juiced" by the lack of defense, pace and space that has become standard. Does that influence your evaluation of the numbers?
How do you feel about usage?
This one gets complicated, but it's worth asking.
Basically, the critics would say "Of course Harden is scoring all those points; look at how often he has the ball." But there's also a natural drag on efficiency as usage increases; we've seen that proven before, and yet Harden is producing at unheard-of levels. The other side of that coin is "Of course Giannis has all these great efficiency stats; he plays less than 35 minutes per game!"
But Giannis being able to produce at the level he is with lower usage matters as well. Antetokounmpo's usage rate is so much lower than Harden's, and yet he's all over the box score. Giannis is the Bucks' best player and obviously their engine, but he's also much more integrated into the offense in a variety of ways compared to Harden, who quite literally is the Rockets.
That can be a good thing or a bad thing. You just have to sort out how much that matters to you.
How do you define "making teammates better"?
I suppose first you should see whether or not you care about this concept. Answer that first, then either skip ahead or come back to this.
There are all sorts of quantifiable metrics for this. The above-referenced Net Rating is a good starting place. Assist points created is a good one to throw in. Looking at how teams perform in whatever key areas that player contributes is another.
For example, the Thunder are dependent on forcing turnovers to create fast-break opportunities. OKC generates three fewer points off turnovers per 100 possessions with George on the floor.
It's more than assists or specific scoring, though. There's a leadership quality to it and a style of play that enables teammates to be their best. All three candidates qualify on that this season, but to various degrees.
Does it matter how "complete" a player is?
This is slightly different from what we're talking about with defense. Giannis Antetokounmpo has shot 35% or better on 3s since Feb. 1, but he's still just 24% on the season from deep. He's also a dominant defensive player in overall impact but can get beat on the perimeter by quicker guards.
Harden is a complete offensive player but at least partially absent on defense.
George, on the other hand, has been elite in every regard. Shooting, driving, passing (though not as good as the other two there), off screens, off the dribble, in transition, etc. If you know a guy can be schemed a certain way (which is highly debatable with all three candidates), does that matter?
Do you care about stat padding?
A player can pursue wins and personal accomplishments at the same time. They shouldn't have to decide between the two. However, there's definitely a strain of voter who feels you shouldn't be gaming the system.
This obviously was most on display with Westbrook back in 2017. It was considered distasteful that Westbrook would chase stats in pursuit of a triple-double. Of course, this ignores that Westbrook grabbing rebounds means the Thunder can push the ball faster and has a statistically significant impact on their offense.
It also ignores this: Since 2016-17, the Thunder are 70-22 when Westbrook records a triple-double. When Westbrook records at least 10 points, nine rebounds and nine assists, their win percentage drops from 76% to 58%. The point is, sometimes this stat padding has a strong correlation with team success.
The big debate this year is about Harden chasing that 30-points-or-more streak. But in games where Harden scores 30 or more, the Rockets are 31-16 (66%). In games in which he scores fewer than 30 points, the Rockets are 7-7.
So there's a definable argument that Harden's streak directly helps the Rockets win. But sometimes these are performances in comfortable blowouts, or clear losses, and it seems to be pursuing the numbers for the numbers' sake. (I also don't know when "30-point games" became a relevant streak. But whatever.) Does that matter to you?
What about aesthetics?
Yet another thing I don't care about, but it's worth asking. Folks really hate the way Harden plays, particularly the foul drawing. PG has the most aesthetically-pleasing game with his jumper; Giannis the most exciting.
They're rare but "I just hate the way he plays" is definitely within the realm of how some voters think. You have to decide for yourself whether it should matter. I would implore you, if you're searching for the "best" answer to who MVP should be, to ignore this element.
Does it matter what happens if you take that player off their team?
This is similar to the question above about how much better you make their team. But it's slightly different in that it suggests the idea, "if this player were not on the team at all, how often would they win?"
This, also, is a flawed premise. If the Rockets didn't have Harden, they'd have his $30 million in cap space available to sign other players. It wouldn't just be a zero. You can't just negate them entirely. Rotations, style of play, everything shifts if you remove that player. Still, people wind up asking this question.
How is this vote going to look in six months? What about in five years?
Something odd about this: I worried when I voted for Harden in 2017 that I would regret not voting for the first player to lead the league in scoring while averaging a triple-double. But instead, time has mostly shown Harden to be more impactful. The Rockets went further in the playoffs, and in subsequent seasons Harden has brought his team to contender status while Westbrook's Thunder have stagnated (before this season).
It's something to consider, though. If Harden winds up averaging the second-most points since Jordan, do you want to be on the other side? This question shouldn't reflect playoff performance whatsoever, but it creeps in anyway. If Harden flames out in the playoffs in a fit of exhaustion again, the vote looks bad. It absolutely should not matter for a regular-season award, though.
How memorable is his season?
The Kobe '06 corollary. Steve Nash made his teammates better, had more wins and had better team efficiency, but Kobe's 2006 season is what everyone remembers.
If everyone's going to remember Harden's scoring season, isn't that most impactful? This goes beyond the stats and analysis and cuts directly to the idea of intangible eye test impact. Harden's been the talk of the league for his scoring exploits. Fewer have talked about Antetokounmpo's season, despite the numbers and success.
Some of this is built on signature games. Harden has the game-winner vs. the Warriors. A 50-point and 48-point game vs. the Lakers. Two 40-point performances vs. the Celtics. All of them wins. Giannis doesn't have as many of those games (in part because Milwaukee plays on national TV far less often and the East has far fewer impact games).
George is in the middle. He has some incredible games (vs. Milwaukee in January, vs. Utah in an overtime classic in February), but not as many since he really entered the MVP conversation in December.
If you go to 10,000 feet, doesn't Harden's season feel "bigger"?
Who’s most impactful?
As I said, this is the one where I decide my pick. Whose performance contributes most to his team's success while he's on the floor? I don't especially care how his team performs when he's not on the court; I care about how they perform with him.
An interesting side bar to this is what George told me about Harden's case, oddly enough. George said that Harden has done "whatever it takes" for the Rockets to win. That they needed his offense, scoring 50 a night, to win games, and that if the cost of that was defense, then that's the cost.
Who does the most to help their team win? That's the measure of impact, and at the end of the day that should be the best and most telling signature trademark for an MVP.
