I'll be honest, I got to thinking about this because of my Surge Mancrush--now brutally by the harsh realities of play (put me in the camp of "I'm not excited about either next year" with a little side dose of "Surge! DUDE! What happened to you this year?!"). We used to say that Surge and Roy couldn't play together because they both needed the ball.

So, here: we have a lot of statistical data about Roy and LMA. The question then: do they play better together or apart? I'm not sure what this all means but I'll lay it out and see what the BE Intelligentsia has to say about it.

In the 67 games the Blazers have played to date, NBA.COM shows Roy's average points as 19.6 and LMA's as 17.7. I ran through the game logs and did a quick tracking to see what happens to their averages:

1. If Roy's is above average is LMA's down? This might indicate that LMA gets bored when Roy starts taking over a game and can't contribute. (We might wonder about LMA's rebound totals and so forth but I'm not quite that ambitious yet.)
2. If LMA's is above average is Roy's down? The opposite condition of the above: it's the Zach Randolph "black hole" problem re-born.
3. Do both rise above average together? This might indicate that they feed off of each other.
4. Do both fall below average together? Now this is an interesting question because it casts doubt on the whole exercise. We might reasonably expect one player to pick up the slack when the other player is down and out and having a bad night. So we want a smattering of incongrouus averages from points 1 and 2 above, yes? Perhaps... but maybe not too many.
5. What happens when they go solo? This hasn't happened much but it's reasonable to take a look.

In the 67 games played, 9 of them have been "solo" games with Roy playing alone in 5 and scoring above average in each game and LMA going solo in 4 games and scoring above average in 3 of them. It's safe to say then that point 5 and the latter half of point 4 oare true: when one player is out, the other tends to pick up the scoring load. That's good I think we can all agree.

Now, in the remaining 58 games, an interesting trend occurs. Roy and LMA are congruent in beither above or below scoring averages in 44.8% of those games.

Both players score below average: 19.0%
Both players score above average: 25.8%

Roy and LMA are in opposition with each other in 55.2% of the games.

LMA is up and Roy is down: 24.1%
Roy is up and LMA is down: 31.0% (a bit of rounding error here means it doesn't quite add up)

By the way, the trend towards opposition is accelerating in this latter part of the season. During the winning streaks, they tended to be in congruence and later on, when the team is not winning, they tend to be in opposition.

October: congruence=5, opposition=4
November: congruence=9, opposition=5
December: congruence=5, opposition=5 (solo=5, all Roy)
January: congruence=3, opposition=11
February: congruence=3, opposition=6 (solo=4, all LMA)
March: congruence=5, opposition=4

Note: I didn't look at wins and losses because I wanted to screen that out. Maybe that was a bad idea... you can let me know in comments. Also, I didn't look specifically at how much above or below average the players were... just that they were above or below. There were a handful of occasions where one could have declared a tie but I ignored that so I could get on with my pontificating below.

These data seem to suggest a couple of things:

1. It's troubling that these two pillars of the team are in opposition so much more often than they're in congruence. To me, it suggests that they can't yet share the ball. They're both down or in opposition with each other almost 75% of the time. I think this is more often than "one picking up the slack for the other" than might be desirable for consistent play, consistent team roles, and consistent winning.
2. Roy seems to be able to contribute more often when Aldridge is playing well than vice-versa. This helps defend Roy from the criticisms he sometimes takes for not getting involved early in the game. Roy getting other players involves is important for their performance, as suggested by these figures.
3. LMA, in turn, needs to get better at garbage time. When Zach was first being born as a Blazer player, we loved him for his garbage play: dude would vaccuum up balls and put them back for 2nd chance points... he didn't need no stinkin' plays! When Roy is taking the load, it does not appear that LMA has learned to contribute in other ways.
4. Both players pick up their scoring when their counterpart is out. Again, we should probably expect this--who else, after all, is going to score--but it's nice to see.

I welcome sharp criticism in flaws on this approach. I like to think about stats, what they reveal, and how to get better at them.