In the GA, non-transport cat context I think Trim Stab's approach is rational and sensible. Once you have a line of best fit you can easily see how it measures up to the data points and offset it by adjusting some constant so the output is never less conservative than the original data. To my mind it would be perfectly acceptable for the OP to do this. Should the question arise as to whether the performance of the aircraft was checked prior to departure this utility would provide evidence that it was (after all, in most GA operations there is no record that the pilot actually looked at the charts). In the end what matters is that the aircraft is de facto operated in accordance with the certified performance and limits, not what software, blunt pencil or crude table you used to actually check.