Time delay error is a significant error source in adaptive optics (AO) systems. It arises from the latency between sensing the wavefront and applying the correction. Predictive control algorithms reduce the time delay error, providing significant performance gains, especially for high-contrast imaging. However, the predictive controller’s performance depends on factors such as the wavefront sensor (WFS) type, the measurement noise level, the AO system’s geometry, and the atmospheric conditions. We study the limits of prediction under different imaging conditions through spatiotemporal Gaussian process models. The method provides a predictive reconstructor that is optimal in the least-squares sense, conditioned on the fixed times series of WFS data and our knowledge of the atmospheric conditions. We demonstrate that knowledge is power in predictive AO control. With a Shack–Hartmann sensor-based extreme AO instrument, perfect knowledge of the wind and atmospheric profile and exact frozen flow evolution lead to a reduction of the residual wavefront phase variance up to a factor of 3.5 compared with a non-predictive approach. If there is uncertainty in the profile or evolution models, the gain is more modest. Still, assuming that only effective wind speed is available (without direction) led to reductions in variance by a factor of ∼2.3. We also study the value of data for predictive filters by computing the experimental utility for different scenarios to answer questions such as how many past telemetry frames should the prediction filter consider and whether is it always most advantageous to use the most recent data. We show that within the scenarios considered, more data provide a consistent increase in prediction accuracy. Furthermore, we demonstrate that given a computational limitation on how many past frames, we can use an optimized selection of n past frames, which leads to a 10% to 15% additional improvement in root mean square over using the n latest consecutive frames of data.