I can answer this question only based on guesses because I am not really sure about your claim that activators are higher in number than repressors. So consider this as an extended comment.
While activators can interact directly or indirectly with the core
machinery of transcription through enhancer binding, repressors
predominantly recruit co-repressor complexes leading to
transcriptional repression by chromatin condensation of enhancer
Not really true. A repressor can simply sit on the the promoter and prevent RNAP from binding the to the latter.
Transcriptional repressors are usually viewed as proteins that bind to
promoters in a way that impedes subsequent binding of RNA polymerase.
Although this repression mechanism is found at several promoters,
there is a growing list of repressors that inhibit transcription
initiation in other ways. For example, several repressors allow the
simultaneous binding of RNA polymerase to the promoter, but interfere
with subsequent events of the initiation process, eventually
inhibiting transcription initiation. The recent increase in the number
of repressors for which the repression mechanism has been
characterized in detail has shown an amazing variety of strategies to
repress transcription initiation. It is not surprising to find that
the repression mechanism used is usually exquisitely adapted to the
characteristics of the promoter and of the repressor involved.
Some justification on why the number activators would be more:
A network perspective:
Lets assume an that gene-A somehow causes the repression of the gene-B. This repression can be either direct (A being a repressor of B) or indirect via some other genes. In the case of indirect regulation you just need one repressor in the network path A → B to result in repression of the latter.
So all repressive paths would minimally require only one repressor, all other steps in the path can be activation; whereas in an activating path you would need 'n' activators for 'n' steps.
Now if you include post-transcriptional regulation then you can find more number of repressors: miRNAs (2588 reported in humans: miRbase-21) and many proteins as well (For an e.g. see here). Most cases of events leading to translational activation are actually derepression. It is logical in a way because an mRNA is a temporary product; it should not by default require an additional signal to start producing proteins.