As there is every likeliness that a poor research paper may receive a higher number of citations, while as on the contrary, a good quality research paper may be poorly cited.
The article 'The impact of article titles on citation hits: an analysis of general and specialist medical journals' (doi: 10.1258/shorts.2009.100020) has interesting findings on this quesstion.
1. Out of fashion topics are out of public fervor. So Quality does not beget quantity. Whereas the currently hot areas of research sure get citations.
2. Simple papers, mediocre papers are read more widely. So they get quoted. More conceptual papers are possibly not recommended (read not cited), because they might demean your own work.
3.extending work of a premier (& active) authors work should get support from erstwhile author, because of their need to find support. It is akin to a multilevel marketing case. Supporting each improves the total citation scenario.
The number of citations received by a research paper depends to a great extent on the journal in which it is published. Some journals have a very strict policy for acceptance of the research paper while others do not. So also some journals are more popular than others and have a very high impact factor. All these determine the various reasons why a particular research paper receives more citations than the other. Metrics is only an indicator of the number of times the work has been cited. It does not necessarily throw light on the quality of the work that has been cited. Nevertheless, one would expect that if a particular work is criticized by any researcher then it is more likely to lose popularity and gradually drop off in future citations. So in a way, metrics would give a measure of the quality of work, though not exact, it would at least give some approximate yardstick to go by.
Over a period of time, perhaps we may not be having hesitations in admitting when citing a source will become more a matter of chance rather a matter of choice, as citing a source will simply depend on the document over which a researcher stumbles. which in fact is what is going on. Especially given the unprecedented publishing, where quantity is openly compromising with quality.
Metrics are important. But in some very specific areas, paper metrics will be low, which is not related to the quality of work. So I think the best metric is in the journal where the work is, and not the work itself.
Therefore, impact factor, time of existence of the journal, amount of database, among others, are some important metrics.