Scheme sucks

Most Lisps specify an order of evaluation for procedure arguments. Scheme does not. Order of evaluation—including the order in which the expression in the operator position is evaluated—may be chosen by an implementation
that is why.

This is the same reason why lazy evaluation is used in Haskell, it is able to make speed and memory optimisations, because it only evaluates when it has to, it does not do unnecessary evaluation, and it can make faster running speed by choosing the best order of evaluation.