The standard deviation measures variation from the average. So for example, if you have cells of width 3, 4, 4, 4, and 5, the standard deviation would be 0.707107, which is not going to be wide enough for anything.

Do you mean that you want the column width to be one standard deviation above the average? Citing the same example, the average width is 4, so one standard deviation above the average would be 4.707107.

If so, accomplishing this would be a fairly extensive operation. When you say that you want to measure the width of each cell in the column, what I assume you mean is that you want to measure the value of the column width if the column were autofitted to that cell. To do that, it would require the following

a) Create a new worksheet.

b) Copy-transpose the first 65,536 non-blank values in your source column into the first row of the worksheet.

c) Autofit those columns

d) Capture the column widths

e) Repeat as needed until all values have been captured.

f) Delete the new worksheet

g) Perform the average and std. dev. calculations.

h) Fit the source column.

Depending on the amount of data you have, this will be a time-intensive operation.