A Professor Looked At 15 Years' Worth Of Information. Then A Designer Packed It Into 1 Punchy GIF.

[Donald Ian Rankin]

28. Your conclusion certainly follows from your assumption.

Conversely, if one person in seven voted fraudulently, for a total of 18,000,000 fraudulent votes, that would work out to 1 in 600,000 being caught. That's the conclusion that follows from that assumption.

For what it's worth, it wouldn't surprise me if your one in a thousand was an underestimate rather than an overestimate - I see no reason to suppose that fraudulent voting in this fashion is widespread. But I don't like seeing data being overtly abused in the way the OP was, and I'm nervous about the way that you're using it (although you're not doing anything nearly as egregious as the OP - it was a direct lie; yours is perfectly correct and honest, but potentially misleading because it might potentially lead people to assume that there's some reason to suppose that 1 in a 1,000 rather than 1 in 100 or 1 in 10,000 fraudulent would-be voters got caught).