Can induction be justified on practical grounds?

Recently, G. Schurz, D. Steel and F. Huber have argued for induction by showing it to be optimal or necessary and sufficient for a certain end. Yet, these arguments need not convince a skeptic who prioritizes the avoidance of error. A firmer argument for induction thus has to show that we should not prioritize the avoidance of error. This seems plausible, if only for practical reasons: Skeptics who do not base their decisions on induction will more often fail to get what they want. The aim of the paper is to discuss an argument for induction along these lines. As a formal tool, I use action games as considered by Schurz. But I cannot rely on his results on the optimality of metainduction, or so I argue. So I compare the inductivist and the skeptic from scratch. Although induction does pay off in certain worlds, there are possible worlds in which it does not. Consequently, the prospects of a practical justification of induction are dim.