In this paper, we present two algorithms for approximating a period given a discrete data set. These algorithms superimpose two consecutive sections of the data for several candidate periods. The first algorithm counts the number of shuffling points per candidate period, whereas the second algorithm computes a distance between points when sorted by time. The best candidate period maximizes the number of shuffling points in the first algorithm, whereas the second algorithm minimizes the distance between points. The experimental validation with noiseless data demonstrates that the relative error for the estimations is less than half of the sampling period and shows that this error does not depend on the harmonic content, as normally occurs with algorithms that estimate a period. The application of the algorithms demonstrates that they properly track the frequency of a power grid and accurately estimate the period of a Van der Pol oscillator, which serves to confirm their applicability to real-time problems.