The problem is you are illogical here.
Take the Luminol prints. You cannot deny that there is a possibility that a Luminol positive print may be due to blood or may be due to an alternative (false positive) mechanism. The question is how likely are the alternatives? This is the 'reliability' of the test. To deny the possibility of an alternative is to deny reality.
The next step is to assess the likelihood that the stains are due to blood or an alternative mechanism.
You (Machiavelli) are at least consistent with your view. You regard a Luminol positive print as most likely attributable to blood even if TMB negative; therefore you view the prints from Sollecito's car and flat as being attributable to blood.
I would argue there is a narrow spectrum that allows a positive Luminol result and a negative TMB. The lower limit for TMB (depending on the reference), is 1:10,000 dilution, for Luminol 1:100,000; the former equivalent to one drop of blood in a litre of water, not enough to be visibly blood stained. This is possible. However, the properties of blood suspensions make this unlikely. Particles like blood cells tend to concentrate at the edge of a drop due to surface tension. So you do not get even distribution of blood across a print as it dries. You get a concentration at the edge of a print. This is not seen here. It would seem unlikely that the same critical concentration that delivered a positive Luminol but negative TMB test was found not only in all the prints found at the second visit in the flat, but also the exact same concentration of blood in Sollecito's car and flat. It is excessively unlikely the same concentration was evenly distributed in all of these prints in multiple different locations.
As a second line of argument even if you believe the prints in the hall are made by Knox in dilute blood there is an entirely innocent explanation; that Knox contaminated here feet with highly dilute blood from the shower (note that this concentration would not be visibly bloody), she reported having the next day. So it is not only unlikely that the prints were attributable to blood, but there is an entirely innocent explanation. This evidence is equally evidence supporting Knox's description of her actions the morning after the murder, supporting her testimony. It is as much evidence for innocence as guilt. It has no informational value.
I'm sorry, but I reject both these argumentations as illogical or nonsensical.
On your first line of argumentation about an "alternative" substance I can see there are two aspects.
First, one problem is that there is always a logical fallacy on the part of those who argue about possibilityo of "alternative substance" and use negative TMB test as an argument to bolster that theory.
You just omit the implication from the fact that
no substance is known that has the property of
reacting to luminol but not to TMB, and that the two tests respond to the same compounds. But the logical consequence of this are of fundamental importance.
In relation to this, you also apparently fail to realize that "non-blood" is a non-alternative, "non-blood" is not a substanc, it's a non hypothesis, you fail to realize how it is necessary to formulize an alternative in positive terms, in order to build a logical possibility for doubt - and also to make any kind of assessment about how this alternative is, documented, plausible or realistic.
On a second level, you are still completely focused on the a chemical nature of the stain, while not considering the rest of its other properties, so underestimating the information that lies in features different from what can be found out through chemical analysis.
You underestimate the unusual nature of the stain, importance of the analogy of their features with the bathmat stain (and the strangeness of the bathmat stains itself, btw), their pattern, their distribution, the set of features. And you also do not cross this information with details such as the fact that one of putative Guede's shoeprints has been washed by half and another one has disappeared completely, and that there are both luminol stains and blood stains that yield mixed DNAs from both Knox and Kercher.
You try to build a reasoning that is analitic about the possibility of chemical analisys of a substance, you follow logical steps walking along in a line, and when you don't find confirmation you stop; but you walk only that line, you don't draw deductions from other cross-path, and you build an analysis from one detail, one property, but that's only analitical, it's only bottom-up, there is no top-down thinking the other way from different sets of information.
Now, on the second argument I radically disagree: I don't think it is realistic to beieve that the luminol prints could derive from "contaminating" one's feet with the bathmat carpet. I don't believe it's physically possible to produce that kind of luminol footprint shapes that way. You can produce that prints only if your feet were completely immersed in a uniform dilution of substance. That is, a dilution of the same kind of the dilution which
produced the bathmat prints at their origin.
