When it comes to science, "wrong" is a relative term.
I was hoping a mathematician would pipe in.
So a mathematician is going to tell us how dating craters doesn't work?
I'm all ears.
I'm mainly concerned with why one set of metamorphic rocks are considered to have "reset to zero", and another set are considered to have only partially reset.
What determines "partial" or "total"?
I'll say this again:
In my opinion, for what it's worth, if you have a meteor crater with 10 million year old tektites, coupled with 10 million year old metamorphic rock; then the metamorphic rock is considered to have reset to zero.
But if you have a meteor crater with 10 million year old tektites, coupled with 15 million year old metamorphic rock, then the metamorphic rock is considered to have partially reset.
In ... my ... opinion.
Partial resetting has to do with how hot the rock gets. If it gets somewhere close to the closure temperature for the mineral you are testing, then some of the Argon, for example, can escape, but not necessarily all of it. If it gets way hotter than the closure temperature, then all the Argon will escape and the clock is totally reset. Also, different minerals have different closing temperatures, so scientists have to be aware of this when selecting which ones to date.
Partial resetting of the K/Ar clock is not very common, because Argon is an inert gas that escapes easily, so if there is a heating event, it usually loses most or all of it. The fact that Argon is lost easily with heating, though, can be a problem. So scientists have to be careful with which minerals they use and be aware of possible metamorphic events. That's why it is mostly used on igneous rock.
However, it is often used on metamorphic rock in conjunction with other methods, like Rb/Sr, in order to get an idea of when there may have been some kind of heating event, since the Rb/Sr clock is not reset so easily.
But the example that you gave of the tektites dating to 10 million and the metamorphic rock dating to 15 generally doesn't happen. There are plenty of examples where the tektites and metamorphic rock come up with the same date. Some of them were given in the links that Rick provided on the first page. By and large, rocks dated with multiple methods come up with congruent results; approximately 90% of the time according to some of the books I've read.
If they do come up different, they don't just say "oh well, better luck on the next test." They try to figure out WHY they don't agree, because they are the exception, not the rule.
Also, if they use the K/Ar on both the tektites and the metamorphic rock, it is easy to see if there was only a partial reset of the metamorphic rock, because they can determine the initial amounts of Argon in each, and if the metamorphic was only partially reset, there would be more initial Argon in the metamorphic rock than there was in the tektites.