OHIO TURNS WHITE, FALLS INTO UNDECIDED COLUMN (Vote Tally Inflation; Vote switching) by Richard Hayes Phillips, Ph.D.
In 28 locations where returns ARE broken down into wards, the sum total of the votes cast in all the wards is less than the total reported for the location. For example, in Cleveland totalling the number of votes for each ward yields 165,578 votes, while the total number of votes reported is 214,902 (a discrepancy of 49,324 votes). Summing over all 28 locations the discrepancy is 214,193 votes.
In 11 locations where returns are not broken down into wards, the total votes cast are greater than the number of registered voters. In University Heights for example, there are 10,072 registered voters, and it was reported that 11,982 votes were cast - thus there were 1,910 more votes than registered voters. Summing over the 13 locations there appear to have been 26,416 more votes cast than there are registered voters.
In two cases, Solon and Strongsville, where returns are broken down into wards, the discrepancy is limited to a single ward each. Of the 7 wards in Solon, six have more registered voters than votes cast - only Ward 6 has more votes cast than voters. In Solon Ward 6 there were 4,300 votes cast while the ward only has 2,292 registered voters. Of the 4 wards in Strongsville, three wards have more registered voters than votes cast, only Ward 3 has more votes than voters). In Strongsville Ward 3 the total votes cast was 12,108 when the ward has only 7,806 registered voters.
In two cases, Mayfield Heights and Parma Heights, the total votes cast, while less than the number of registered voters, is so high 87.87% and 95.6%, respectively) as to cause a reasonable observer to question whether there was such a high turnout.
Link to Phillips' complete analysis: <
http://web.northnet.org/minstrel/ohio.htm >