Number of entries listed

I noticed that on many pages the 'number of entries listed' is not equal to the number of minerals (valid and not) listed, although it is displayed right below the list of mineral species?

Example: please see the Bad Bleiberg page [www.mindat.org]: there are 48 mineral species in total (valid and not), but 55 entries listed.
Apparently, the number of entries listed includes both the mineral species entries (48) and the sub-locality entries (7).

What is the rationale behind lumping together mineral species and sub-localities?
Wouldn't it be more clear to present those two totals separately?

Ok, so while I am still waiting for any response from those who know how the counting algorithm works, here is another example:
For the Allchar mine [www.mindat.org] the number of entries listed is 73, while there are in fact only 59 mineral species listed, including 53 valid species. The number of sub-locality entries is just 1, so it cannot account for the difference of 14.
How could a simple count be so far off? It's got to be picking up some garbage in the DB...

The sublocality has 16 minerals listed. Alchar has 57 minerals listed as occurring there. 57+16=73.

Thanks for your response, David. It raises two questions though:
1. Sorry, but there are 59 mineral species listed for Allchar -- I just re-counted them. Which 2 species did you exclude, to arrive to the count of 57?
2. What is the point of combining the two lists into one total count? It was my understanding all along that the sub-locality minerals list is supposed to be included in the next higher-level locality list. Is it not always the case?