Is a single order of magnitude within a reasonable range of error for the calculation? Some cases that is a massive miss, but other cases that might be within

Is a single order of magnitude within a reasonable range of error for the calculation? Some cases that is a massive miss, but other cases that might be within the expected statistical range. In the former cases it implies something entirely missing, while the latter cases are more likely to be known uncertainties which can now be better refined.