In Luhansk, addition was dismissed as a Bandera method.

It can be considered proven that the "results" were determined arbitrarily.

Boris Ovchinnikov writes:
The results of voting are usually determined by an arithmetic operation called addition — first, data from precincts are summed up to get results for district election commissions, and then district results are further summed up into regional and national results.

In Luhansk, summation was rejected as a "Bandera method," and multiplication was used instead. Multiplication of numbers made up from thin air. This is evident from the fact that there are two percentage figures — the percentage of turnout relative to the number of registered voters and the percentage of votes in favour of independence relative to valid ballots — that match with a precision of up to three decimal places. 75.2000% and 96.2000% (for reference: if real numbers were summed, with a probability of more than 99.9%, their ratio would yield non-rounded percentages — like 75.2637%).

The probability that two rounded percentages coincidentally match this way is less than one in a million. Thus, it can be considered proven that the "results" of the referendum, at least in the Luhansk region, were determined arbitrarily by its organizers, without any connection to the actual voting results.