Roger, et al, It is apparent that you missed some of the finer points of microbiology but the article is more than sufficient for comparing relative drinking water contaminants.
Getting back to sizes of microbes, this has a rather wider variance than you guys intimate. When a bacterium divides, it becomes difficult to determine the size of the daughters. All that is really required are the mechanical necessities for life. You can easily see that a 1/10th size reduction for one daughter is possible, as you mention, but often enough the difference can be greater. Even an absolute filter can let these small daughter cells through. While I have used filters, I do not trust them for this reason. Especially where I will be drinking water several hours after filtering, the cells get a chance to grow after dividing. Amoeba can often slip through pores far smaller than their actual size. I would point out that they have no structure in the cell, hence act almost like a thick fluid. Again, this can take time, but the high preponderance of gravity filters used for overnight filtering can allow a couple amoeba to flow through. On the theory side, this could show up as amoebic dysentery.
Filters are generally statistical in nature. Only the absolute number can be relied upon for filtration and sometimes, even this can fail. Filter fibers are generally laid down in a relatively disorganized state. Absolute filters have set pore sizes. The difference is the cost. Coffee filters, biodiesel filters, or even a cotton ball in a tube are all statistical in operation. Most backpacking filters are better, but still lack the filtration of a good absolute filter.
The overall assumption is that the person using any methode of cleaning water for drinking is healthy and in a "normal" state. Note that for any person, an immune compromised person especially, the active infection can be caused, in all cases, by a single active bacteria, protist or virus. It is statistical. A rather facetious example is the 99.9% safety. You can interpret this to mean 1 person in a thousand will get hammered. Just because one person still gets sick does NOT mean the cleaning method is at fault. This is a rather obscure point that most people miss. As you point out, a difficult thing to trace since it can also be caused by other problems: sanitary conditions, rain dripping off a tree and into your cup, etc.
I am glad you guys pointed out that chemical treatments all qualify as pesticides. Worse, they are generally toxic to people. Which is worse? The disease or the cure? Overdosing water with a large amount of tannic acid in it with chlorine dioxide can give you toxic disorders. Slight overdosing would work better by decreasing this risk, but, also increase the risk from some pathogens. A difficult call in the field, but overdosing with chemicals to improve the efficacy is not always a good solution. If you are out for a long time, alternating methods and chemical types will help reduce toxicity.
Anyway, you make the point that for a normal person, none of this is a problem. I certainly agree. But, every time I slurp up from a clear mountain stream, I cross my fingers… Yes, in most cases of drinking water from these sources, I know my body will handle it.