In this thesis, we study 2-player combinatorial games on graphs. We devote a lot of attention to strong positional games, where both players have the same goal. First, we consider the so-called fixed graph strong Avoider-Avoider game in which two players called Red and Blue alternately claim edges of the complete graph Kn, and the player who first completes a copy of a fixed graph F loses the game. If neither of the players claimed a copy of F in his graph and all the elements of the board are claimed, the game is declared a draw. Even though these games have been studied for decades, there are very few known results. We make a step forward by proving that Blue has a winning strategy it two different games of this kind. Furthermore, we introduce strong CAvoiderCAvoider F games where the claimed edges of each player must form a connected graph throughout the game. This is a natural extension of the strong Avoider-Avoider games, with a connectedness constraint. We prove that Blue can win in three standard CAvoider-CAvoider F games. Next, we study strong Maker-Maker F games, where now, the player who first occupies a copy of F is the winner. It is well-known that the outcome of these games when both players play optimally can be either the first player’s win or a draw. We are interested in finding the achievement number a(F) of a strong Maker-Maker F game, that is, the smallest n for which Red has a winning strategy. We can find the exact value a(F) for several graphs F, including paths, cycles, perfect matchings, and a subclass of trees on n vertices. We also give the upper and lower bounds for the achievement number of stars and trees. Finally, we introduce generalized saturation games as a natural extension of two different types of combinatorial games, saturation games and Constructor-Blocker games. In the generalized saturation game, two graphs H and F are given in advance. Two players called Max and Mini alternately claim unclaimed edges of the complete graph Kn and together gradually building the game graph G, the graph that consists of all edges claimed by both players. The graph G must never contain a copy of F, and the game ends when there are no more moves, i.e. when G is a saturated F-free graph. We are interested in the score of this game, that is, the number of copies of the graph H in G at the end of the game. Max wants to maximize this score, whereas Mini tries to minimize it. The game is played under the assumption that both players play optimally. We study several generalized saturation games for natural choices of F and H, in an effort to locate the score of the game as precisely as possible.