The reason you cannot split test a smart campaign is actually quite simple. In some ways the smart triggering works like a split test in that it uses different triggers to find out what gives the best conversion rates for that specific campaign.
This means that there is actually a range of different campaign (with different triggers) within a single SleekBox.
Apart from the technical difficulties this presents, it also means that a A/B split test of a campaign with smart triggers enabled will not yield any meaningful results.
A quick example:
Campaign A has smart triggers enabled
Campaign B doesn't have smart triggers enabled
Campaign A with trigger 1, has a conversion rate of 2,1%
Campaign A with trigger 2, has a conversion rate of 3,6%
Campaign A with trigger 3, has a conversion rate of 2,9%
Campaign A with trigger 4, has a conversion rate of 1,8%
Campaign B, has a conversion rate of 2,7%
As you can see this means that campaign B is performing better than campaign A with triggers 2 and 3 but worse than campaign A with triggers 1 and 4.