A typical factor of number of additional bases on target is around 2-4x, depending on the size of the region of interest and the properties of the sample.

Please keep in mind that the theoretical maximum gain is the inverse of the proportion of the number of reads which match your target.

For example, targeting 50% of the genome can never give you more than 2x enrichment, while targeting 0.1% of the genome would theoretically mean a maximum of 1000x improvement, but in practice there is a constant cost for a rejection of a read, meaning that the gain is much less than this.