Over the past 70 years, researchers have proposed two distinct psychological mechanisms (fast vs. slow) to explain how expert chess players differ from non-experts. Although most researchers agree that both fast and slow mechanisms play at least some roles, there is disagreement about their relative importance. I have proposed a “key-position theory” in this thesis, stating that slow processes such as searching deeply and planning, are required in only a few key positions. Further, key-position theory posits that stronger players are able to search more deeply than weaker players even in speed chess and do the time-consuming search in key positions. I tested an implication of key-position theory in two studies. In Study 1, I focused on the distribution of individual move times in speed chess. The stronger players spent more time on some positions than the weaker players, which indicate that the players were able to do calculate and search even when time is severely limited in speed chess. In Study 2, an indirect measure of depth of search on poor moves has been used as a function of skill difference. The stronger player made proportionally few blunders than the weaker players, and even if they make poor moves, they tend to search deeper on these poor moves than the weaker players. In sum, both studies provide explicit and implicit evidence consistent with key-position theory. These findings suggest that building up numerous patterns is necessary, but not sufficient for the development of chess expertise. Other factors including the ability to plan, search deeply, and apply principles likely play important roles.