Originally Posted by torocan Actually, the problem isn't that AST% is not a perfect stat (not stat is), the problem with AST% is that it's a fundamentally flawed and misleading statistic.
Notice that within the formula it is a derivative of total Team assisted goals, in other words percentage of total assisted team buckets throughout a game. The problem with this is it inherently misrepresents assist efficiency as it's a gross volume stat.
What I mean by gross volume stat is this. Let's assume a hypothetical.
If Harden is the primary ball handler and handles the PG duties 3x as much as Lin, even if Lin is 2x more efficient, Harden will still have a 50% higher AST% as Lin. In other words, like PPG, it does NOT take into account efficiency in terms of the performance of those duties.
Now, when I talk about Lin as a facilitator, I'm specifically seperating that from his overall efficiency as a "playmaker", which you defined as being able to generate a shot for oneself AND team mates.
Harden is one of the most efficient "playmakers" in the game. This is because his ability to translate a possession into either a Shot OR an Assist that results in points is very high.
However... that does not necessarily mean that Harden should be the primary ball handler. There are multiple reasons for this, one of which is that his efficiency drops off dramatically when he is forced to play the PG AND Shooting guard roles simultaneously. It's just WAY too easy for teams to key in, double team/shadow and pack the paint (ala OKC or Miami).
When I talk about Facilitation, I actually try to look at more granular statistics (ones that attempt to isolate passing efficiency).
Now, there is no single stat that can paint an accurate picture, but when you take a look at a broad range of statistics that look at aspects of passing and they point in the same direction, I don't think you can fully discount their accuracy.
Notice that even after isolating Harden's playmaking (looking exclusively at passes attempted), every stat category favors Lin in terms of passing. Not just the number of successful passes leading to points scored, but the Quality of those passes, as well as the number of turnovers as a ratio compared to general assists.
Even if you completely exclude Harden's general ball handling TO's and adjust for passes attempted, the percentage of passes he attempts that result in TO's is higher than Lin's.
Now, whether that's due to Harden being double teamed more (as the primary offensive threat and PG simultaneously), or simply due to him being an inferior general passer (debatable), what is indisputable is the shots that result from his Facilitating is simply inferior.
Lin produces higher quality shots for his team mates, which statistically will always result in higher point production when viewed purely in terms of facilitation. In other words, Lin makes his team mates better MORE than Harden makes his team mates better.
That doesn't mean that Harden is NOT the less efficient playmaker. That actually can't be determined from passing stats exclusively. What it does mean is that Lin elevates those around him more, and that would include elevating Harden's shooting due to the quality of looks.
Does this mean Harden shouldn't be the primary "playmaker"? Not necessarily. However, I still maintain that it makes sense in terms of clock usage for Lin to *initially* generate shooting opportunities, but for those playmaking opportunities to work their way to Harden more often than to other players on the team in general.
In more broad strategic terms, it's also good strategy to start the ball AWAY from Harden more often. Not only does this give opposing defenses more different offensive looks to deal with, this gives Harden more room to operate off the ball, and more importantly to work with cutters and screens to give him open space opportunities to either isolate a defender, or to initiate a Pick and Roll without being under immediate double team pressure.