or put another way
deaths by bedsheet strangulation in 2000 = 327
population was 281,421,906 according to the census
or
odds of being strangled by bedsheet in year 2000 was 1 in 860,617
odds of winning $10,000 in powerball 1 in 698,976
just think, instead of sleeping those folks should have bought a powerball ticket
While what you have indicated may be correct in most circumstances, it is not technically correct. To avoid non-discrimination testing the formula in the plan has to actually be a safe harbor. Hence, if you allocate as if you had a safe harbor integrated formula you still have to do a non-discrimination test. Most of the time you will then perform the non-discrimination test on the basis of contributions, rather than benefits. So you won't typically be using cross-testing. But you still need to do a non-discrimination test.
All of that is academic if the allocation is done as if you had a safe harbor integrated formula AS LONG AS THE SAFE HARBOR FORMULA USES 100% OF THE SOCIAL SECURITY WAGEBASE. If you use one of the optional safe harbor formulas which uses a lower percentage of the Social Security Wagebase then you can easily run into a circumstance where the allocation will FAIL non-discrimination testing even though it faithfully allocates in accordance with a safe-harbor formula.