While what srk says is true to a degree the usage will be obviously skewed, as the Malaconda play test is not the OU metagame. In order to address the success of the CAP we have to do much more than just looking at the usage statistics.
For example, people are inherently using sun because they want to use Malaconda for the purpose that was intended. So if sun (and therefore Fire, Grass, etc.) usage is up X% from typical OU statistics that's great, but if sun is losing the majority of the time we haven't really achieved anything as we've done nothing to compel users to run sun in the OU metagame. By the same token it is difficult to assess the success of Malaconda given the over-preparation that inevitably occurs.
As long as these inherent flaws in the play test system exist it is ill-advised to use usage statistics alone as an assessment of our success, so I regret that a degree of speculation and interpretation will be required to assess the performance of this CAP.
For example, people are inherently using sun because they want to use Malaconda for the purpose that was intended. So if sun (and therefore Fire, Grass, etc.) usage is up X% from typical OU statistics that's great, but if sun is losing the majority of the time we haven't really achieved anything as we've done nothing to compel users to run sun in the OU metagame. By the same token it is difficult to assess the success of Malaconda given the over-preparation that inevitably occurs.
As long as these inherent flaws in the play test system exist it is ill-advised to use usage statistics alone as an assessment of our success, so I regret that a degree of speculation and interpretation will be required to assess the performance of this CAP.