"Thou shalt make no effort to ride the hot hand, for the hot hand is but a shape in the wind." -- Commandment No. 8 of Bill James' "10 Commandments of Sabermetrics"
There have been rigorous studies that have shown that when you examine the data, even the best hot streaks have little predictive value of what a player is going to do in the future. In other words, just because a player has had two good weeks, it doesn't make him more or less likely to have a third week that is just as hot.
But we all know streaks do happen, and they can last days, weeks or even months. Sometimes these streaks are just normal statistical fluctuations that happen over the course of a season. Sometimes a closer examination reveals a fundamental change that allows the player to sustain improved performance. A mechanical adjustment in a swing, a pitcher that has learned to keep his front side closed in his delivery, a speedster whose hamstring is finally 100 percent healthy ... all these things can contribute to sustained streaks. Sometimes these "streaks" wind up being just a return to a previous, better level of performance, or they could be the beginning of a new level.