The adaptive dataflow accesspath precision will scale down the precision at which certain accesspath are tracked based on an over-approximation of the tuple-count cost (and fan-out) of a more precise tracking. This cost estimation is essentially done by multiplying the accesspath fan-out with the number of relevant dataflow nodes, but for certain ubiquitous Contents this approximation can be too pessimistic - if the fan-out of a particular accesspath head (i.e. Content) involves a number of tails, which are individually constrained to isolated scc's of the flow graph then the cost approximation becomes much larger than the actual cost. In addition, it has been observed that correctly tracking the contents of, say, Object[] in Java is quite important for removing FP flow. In particular, this change reduces the number of results of certain queries on jdk11 with more than 90% (all of the removed results were FPs due to lost precision in Object[]).