Calculators used by investors to gauge their probability of retirement success may not actually be good gauges of that success. Comparing the outputs from several retirement planning programs shows a huge dispersion of results, underpinning how investors and advisers should use them as a guide rather than take them at face value.