When ordering by Relevance, the count of the keyword, meaning how many times the keyword used for the Search is present in the same field (Title, Summary, Problem, etc..), is not taken in consideration; what is considered is the presence of the keyword in the specific fields (Title, Summary,..).
When there are more documents having the keyword present in all the relevant fields, NOT depending on how many times it is present in the single field, the documents are ordered with the most recent at the top of the list.
If the keyword is not present in all the specified fields but only in some of them, the documents are ordered on the base of the weights they have (keyword present in Title is more important than keyword present in Summary and this is more important than keyword present in Problem etc.) and, as before, if more than one document has the same situation, meaning the keyword is present in the same fields (not depending on how many times it is present in the same field), the document displayed at the top of the list is the most recent.
For details and info about the weight of the relevant fields, see TEC502619.
To give an example, consider a scenario with several documents (KDn) having the keyword present in the following way (included in parenthesis the relevant fields where the keyword is present):
KD1 (Title + Problem)
KD3 (Summary + Problem)
KD5 (Title + Summary)
In this situation the order will be:
KD5 Title + Summary
KD1 Title + Problem
KD3 Summary + Problem
The document containing the keyword in Title and Summary is displayed before the document containing the keyword in Title and in Problem and this last document is displayed before the document containing the keyword only in Title, etc..
The above result, as said, doesn't depend on how many times the keyword is present in Title, Summary and Problem and it depends only on the keyword present in the fields; when more documents have the keyword present in the same fields, the most recent is displayed at the top of the list.