Almost everyone has their own definition of "most valuable", mine is the player whose individual efforts are most responsible for his team's success.
A great basketball player can have a much larger impact on the game than in a sport like baseball, where an MVP can come from a last-place team. That's why I'd have a hard time giving the MVP to any player not on a 50-win team with home-court advantage in the first round. One great player can lift a team full of replacement-level players to that level; see LeBron and last year's Cavaliers.
So if we look at the top eight or so teams in the NBA, we want to find the player most responsible for his team's success—the guy whose team would suffer the most if the was replaced with an average player for his position.
This season, the Cavs kept the majority of their core together (Varejao, Mo Williams, Anthony Parker, JJ Hickson and Antawn Jamison) but replaced LeBron with Jamario Moon. While there were a few other personnel changes, the LeBron/Moon switch is the primary reason why they won 43 fewer games this season. Conversely, if you had replaced Kobe with CJ Miles in the off-season, the Lakers would still be a playoff team.
A fair criticism of this logic is that it rewards the player with the most incompetent teammates, but it's no more or less fair than giving the MVP to the the top player on the league's best team, in effect rewarding the guy with the best teammates.
The strengths and weaknesses of your best player are going to end up being reflected in the players put around him. If he's a dominant defender, then you're going to look for offensive-minded players; if he's a great offensive player, then you're going to look for guys who can get his back defensively.
And if he can dominate both sides of the ball, than you don't really need that much talent around him at all. Tim Duncan won a championship on a team whose next three best players were Steven Jackson, Speedy Claxton and Malik Rose. Dirk Nowitzki couldn't win one playing with Steve Nash, Mike Finley and Nick Van Exel.
So which player on a top team makes his teammates better on both sides of the ball, therefore making his contributions the most irreplaceable? To me, the answer is Dwight Howard.
At 6'11" 265 lbs. with a 7'4 wingspan, a 9'4 standing reach and an over 30' vertical, he dominates the interior of the lane. He's won the last two Defensive Player of the Year Awards and there's no reason he can't win the next five.
He leads the league in blocks (2.4) and he's second in rebounds (14.3), which is a pretty rare combination. Because going after a block means you can't box out your man defensively, most of the league's best shot-blockers aren't great rebounders. Only two other players—Utah's Al Jefferson and Milwaukee's Andrew Bogut—are in the top ten in both categories.
** While rate statistics (like block percentage and rebound percentage) more accurately reflect a player's underlying skill level, the sheer number of rebounds and blocks reflect how much he contributed in one particular season. Howard blocks shots at the same rate as the Bulls' Taj Gibson (4.9% of the opposing team's field goals when they are on the floor); the difference is his superior all-around game gets him more minutes than Gibson, meaning Orlando benefits more from his shot-blocking than Chicago does from Gibson's. **
For all intents and purposes, he's a one-man defense. You can put almost any combination of players on the floor with Howard and you're going to have a great defense, which the Magic found out this season. Orlando had the 3rd-highest defensive rating in the NBA, despite starting four guys who are average at best defenders—Jameer Nelson, Jason Richardson, Hedo Turkoglu and Brandon Bass.
Howard's defensive ability gave the Magic the luxury of specifically targeting one-dimensional offensive players, knowing that he would make up for their defensive short-comings. Since coming over in a series of mid-season trades, Jason Richardson's defensive rating went from 111 to 104, Gilbert Arenas' from 111 to 102 and Hedo Turkoglu's from 110 to 103.
A dominating shot-blocker at the center position makes his teammates better defenders; they don't have to double down on opposing post players and leave their men open on the perimeter—and they can play much more aggressive individual defense, challenging jumpers and funneling their men into Howard when they drive.
What's different this year is Howard's contributions on the offensive side of the ball. His summer with Hakeem Olajuwon seems to have paid dividends. While he'll never have the most aesthetically pleasing or fundamentally sound post-up game, he's developed much better touch on his jump hooks in the lane.
As a result, Orlando runs a lot more of their offense through Howard than in the past; his usage rating made a massive jump from 23.9 last year (which put him in the mid 30's amongst the NBA's top scorers) to 27.2 this season (the top 10). Yet despite creating more shots for himself, his field goal percentage has barely dipped at all—dropping a percentage point from 61 to 60.
None of the Magic's other top seven players, in terms of minutes played, have a usage rating north of 22. Most cluster around 20, which is the average number of possessions a player would get run through him if the ball was distributed equally. Asides from one half-season from Jameer Nelson in 2009, none of Howard's teammates have put in All-Star level performances recently and none could have offense run through them efficiently for very long.
** Gilbert Arenas, 8th in minutes played, is second in the team in usage rating at 24.2. But he's so incredibly ineffective, getting an 8.4 PER and shooting a ghastly 34% from the field and 26% from three, that it reflects rather poorly on his teammates that the coaching staff doesn't trust any of them to take the ball out of Arenas' hands. Long-story short: if Orlando is going to make noise in the playoffs, than Arenas is going to have to play better. **
Orlando's offense is pretty simple: surround Howard with four three-point shooters and let the defense pick its poison. Either live with single-covering him inside (which the vast majority of NBA teams cannot) or send doubles and give their shooters open looks.
After an embarrassing 4-0 drubbing by the Magic in the second round of the playoffs last year, Atlanta decided to start Jason Collins, a 7'0" 255 lb. defensive stalwart, when they play Orlando, despite his unbelievably awful offensive statistics (a 0.9 PER last season, a 5.3 this season). Because they didn't have to send double-teams, the Hawks could stay on Orlando's three-point shooters. As a result, the Magic took less threes in those four games than they did for the season (12 per game as opposed to 26) and made them at a much less efficient rate (22% compared to 37%).
None of Howard's competitors for the MVP are as valuable to their teams on both sides of the floor. Replace Howard with Jeff Foster and the Magic look a lot like the Toronto Raptors, a 20-win team comprised of one-dimensional shooters, unable to create open looks for each other or defend anyone.
Derrick Rose has carried the Bulls' offense all-season; he has a monstrous 32.5 usage rating (second in the NBA only to Kobe) yet scores 24.0 points on 45% shooting and dishes out 7.5 assists despite receiving nearly all of the defense's attention. But the Bulls are actually better defensively when Rose is off the floor.
Which doesn't mean all that much, since Rose is usually replaced with CJ Watson, a defensive-minded role player who doesn't have to expend nearly as much energy on offense as Rose does. In reality, Rose has <Jazz-Derrick-Roses-defense-easy-to-see.html"> become a much better defensive player under new coach Tom Thibodeau, maximizing his incredible physical gifts at 6'3" 190 lbs. and his defensive rating has dropped from 109 to 102.
What that stat tells you is that Rose isn't the lynch-pin of the Bulls' defense in the same way that Howard is for the Magic. Which makes sense, it's much harder to affect your teammates' defense when you're assigned to a man far out on the perimeter than when you can station yourself next to the rim.
At this point, an observer might wonder what more could Rose do to get an MVP? The answer is "not much". Centers are a lot more valuable to winning than point guards. Since the NBA began giving out the MVP award in 1955, a point guard under 6'3" (i.e. not Magic Johnson) has won it only four times—Bob Cousy in 1957, Allen Iverson in 2001 and Steve Nash in 2005 and 2006.
** The only two offensive players to make the NFL All-Decade teams in the 1990's and 2000's were Willie Roaf and Larry Allen, both offensive-linemen. But because one individual lineman isn't nearly as important as a single QB or RB, neither ever won an MVP award. **
The case for Dirk Nowitzki is very similar to Rose's; Dirk's had to single-handedly carry a Dallas offense that doesn't have any other All-Stars you can run offense through consistently. After his 28.2 and with Caron Butler out for the year, the Mavs next highest-usage players are Jason Terry at 25.0 and JJ Barea at 23.7, but neither undersized guard is a particularly efficient scorer (Terry shoots 45% from the field, Barea 43%) and both are huge defensive liabilities.
Dallas has the 8th-highest offensive rating in the NBA and they average a little over 100 points a game. But in the nine games Dirk missed with a knee injury in the middle of the season, the Mavericks went 2-7 while their offense imploded, dropping 10 points off their season average.
However, unlike Rose, Dirk is a passable defender at best. One of the main reasons why the Mavericks have out-performed expectations this year is because they've paired Dirk with Tyson Chandler and Shawn Marion in the front-court. Chandler can guard 5's and 4's while Marion can guard 4's and 3's, so as a result, Dirk can always be hid on the other team's worst front-court player. Dirk even said that he'd give Chandler the team's MVP because of how much he improved their defense.
** While I think that statement is more a reflection of how modest and self-effacing the German is, it is nevertheless still interesting. **
Which brings us to South Beach. LeBron and Wade are still as great as ever, but their games replicate more than complement each other. They are #1 and #2 in terms of the ability to drive the ball to the basket and get shots for their teammates—and they finished 2nd and 3rd respectively in the Defensive Player of the Year voting in 2009.
Both players' usage ratings have dropped this season, a reflection of the better talent they are now playing with. It's just hard to imagine a situation where either could be the most valuable player in the league, when in their absence, the other, along with Chris Bosh, would just start dominating the ball more and the Heat would still be a very good team.
That's why winning the MVP is no guarantee of a championship, in fact it's almost the exact opposite. The last MVP to go on to win the NBA title was Tim Duncan in 2003. Creating a team that depends so greatly on one player is a big gamble; if there is no Plan B, what happens when Plan A doesn't work?
In the last two playoffs, the Celtics and the Magic were able to use athletic and mobile seven-foot defenders to wall off the paint, where LeBron does most of his damage. And when LeBron, the two-time MVP, wasn't dominating the game, the Cavs had no chance to win. A similar situation happened in '08 with Kobe. In the Finals, the Celtics tightened the screws on him defensively and held him to 40% shooting; and the Lakers fell apart offensively.
** Two years later, a more complete Laker team was able to offset his 6-24 shooting performance in Game 7 with 20 points from Ron Artest and 19 from Pau Gasol. Say what you will about Artest's antics this season, but LA would not have beaten Boston in the Finals last year without Artest's incredible series-long effort on Paul Pierce and his offensive explosion in Game 7. **
Of course, with Duncan, no effective counter-measure was ever found. But until LeBron learns some post-moves, you can't say that about any of today's stars.
This has to be somewhat discomforting news to the top 3 on my hypothetical MVP ballot—Howard, Rose and Dirk. If any of the three is contained in a seven-game series, then their team will lose.