We went to machines as a way of counting votes because the machines were faster and more accurate than humans doing the same job. Machines have much less margin of error than do humans.

Thank God for this difference. The imaginative and inventive human mind is what makes us different than the rest of the animal kingdom. Even at rest, the human mind jumps all over the place, dreaming. Awake, the human mind is constantly taking in tons of stimuli, correlating it, associating it and selecting from it. Machines can’t do that, not yet. Invented by man, the machines just count. And yes, machines can be manipulated and corrupted by their inventors, especially the older machines.

Unlike machines, the human mind cannot do one simple thing for any great length of time. It cannot even hold the image of a loved one for more than a minute or so before it starts recalling experiences, other things involving that person or jumps to something completely different. Try it. Lock yourself in a closet, dark bedroom, or stare at a blank wall. It’ll be good for you. Then realize that, if programmed to do so, a machine could hold that same image forever.

But even the best machines still have a margin of error, especially the old ones like the first IBM punch cards and sorters. This margin may be less than one percent, but when you’re dealing with six million votes in the state of Florida and trying to find differences of one or two thousand, it means you are looking for the proverbial needle in a haystack and this metaphor is real. When you are looking for .00025 of the six million, 1,500 intended votes, you are truly looking for a real needle in a real haystack.

If you ran all the ballots through the sorters again and again, you would get a different result every time. About the best you might do is to make ten or twenty runs and then average them out.

Look at it this way. If the margin of error was one percent and you counted 1,000 votes on the old machines, then you would very likely get an answer differing between one and ten each time you ran the data. You would have a very good chance of getting the same answer once in awhile. If you had 10,000 votes to count, you would get answers somewhere between 1 and 100 every time you ran the data, and much less of a chance of getting the same answer twice. If you had one million votes to count you would get some different number between 1 and 10,000 every time you ran the data. And so on and on. And this is using the more accurate machine method. It’s what happens when what you’re looking for is within the machine’s margin of error.

If you are dealing with human hand counts the margin of error is phenomenally higher. Adding more and more people, or trying to do the job faster only compounds the margin of error.

Maybe, just maybe, if you had unlimited funds and resources, if you did it very slowly and very thoroughly, taking several minutes to evaluate each ballot, spent six months or a year on each total count, ran about ten recounts—then maybe you would have as accurate a count as the old sorter machines gave you the first time. There’s that much difference.

So, what are you supposed to do with the results of a presidential election that turn out this close? You do what any normal data processing firm using equipment like this does. You reconcile. You report your margin of error and then you report your results. You see this all the time. Such and such research has a margin of error of three or four percent. There isn’t any other rational thing to do. Better luck next time.

Next time, you might make certain you’ve got better equipment. You might even make certain that you have more differences in candidates. Only then do you try to establish standards on what you might do in future recounts or if you should ever get into this mess again.

You do not start screaming about everyone’s right to vote if you didn’t provide the equipment to do it in the first place. And you especially do not complain when putting better equipment in every polling place in the country is significantly less than the retirement money you have been stealing from us every week. That’s hypocrisy.

You do not start complaining about people being disenfranchised because ballots and instructions were not printed in Creole. Especially since becoming a citizen of this country involves rudimentary abilities to read English and only citizens are allowed to vote.

You do not start complaining about people being confused if you dragged, bribed and bused them to the polls to do something they hadn’t even thought about until you or your cohorts came on the scene.

You do not come back to us screaming about finding different results without being able to tell us exactly how you did it, who observed you doing it and how many times you did it. We are going to want to know specifically and in great detail exactly how you overcame the margin of error, human or machine. Are you prepared to do this?

You might just as well put a bunch of red and blue beans in a large jar or milk can and then reach in a pull out a handful. We’ll believe you if you tell us that you did this ten times in a row and it always came out the same. We’ll believe you in the hope that you will just go away and stop making such a fuss.