In general, how to find the smallest number which corresponds to best score in a noisy curve? Below is an example result form a recursive feature elimination process. The x axis is the number of features selected and the y-axis is the well-known R2 metric.
It seems either 65 or 80 is a good choice?