A theoretical expansion of the Sprout game

Abstract: Sprout is a two-player pen and paper game which starts with $n$ vertices, and the players take turns to join two pre-existing dots by a subdivided edge while keeping the graph sub-cubic planar at all times. The first player not being able to move loses. A major conjecture claims that Player 1 has a winning strategy if and only if $n \equiv 3,4,5$ ($\bmod~6$). The conjecture is verified until $44$, and a few isolated values of $n$, usually with the help of a computer. However, to the best of our understanding, not too much progress could be made towards finding a theoretical proof of the conjecture till now. In this article, we try to take a bottom-up approach and start building a theory around the problem. We start by expanding a related game called Brussels Sprout (where dots are replaced by crosses) introduced by Conway, possibly to help the understanding of Sprout. In particular, we introduce and study a generalized version of Brussels Sprout where crosses are replaced by a dot having an arbitrary number of ``partial edges'' (say, general cross) coming out, and planar graphs are replaced by any (pre-decided) hereditary class of graphs. We study the game for forests, graphs on surfaces, and sparse planar graphs. We also do a nimber characterization of the game when the hereditary class is taken to be triangle-free planar graphs, and we have started the game with two arbitrary generalized crosses. Moreover, while studying this particular case, we naturally stumble upon a circular version of the same game and solve a difficult nimber characterization using the method of structural induction. The above mentioned proof may potentially be one approach to solving the Sprout conjecture.