SC-ST seats are not an indicator of SC-ST popularity
Some basic pointers on election analysis
In the aftermath of elections, a lot of questionable analysis - seemingly based on hard “data” - is doing the rounds. Some of it is rhetorical put out for partisan not analytical purposes but even the news media is putting out such obviously flawed reports that it hardly seems possible that the analysis was subjected to any editorial vetting at all. So in the interest of more substantive discourse, some basic pointers on election analysis.
Voteshares are more relevant than seats in assessing relative popularity: in the first-past-the-post system of voting, the candidate with the highest number of votes wins, even if the difference is only one vote. This underscores the importance of booth management and get-out-the-vote efforts on election day - but no one would argue that the candidate who lost with a difference of say 100 votes was the less “popular” candidate. Therefore, in ascertaining the relative strength and popularity of candidates/ political parties, one should look at the relative voteshares and not seats.
SC-ST seats are not an indicator of SC-ST endorsement: In every SC-ST seat, actual SC-ST votes are a minority of the electorate. Winning a SC reserved seat thus requires winning a plurality of votes from other caste groupings as well, not just SC votes. Moreover, SC-ST votes are distributed (with some variation) across all the seats in the state. It is thus not possible to extrapolate a party’s popularity with the SC-ST electorate simply by the number of reserved seats won by any party. In fact, reserved seats tally is especially unsuitable for extrapolation because the ST-ST vote inevitably gets divided between the candidates who are all mandatorily from the reserved category. This holds true of almost any grouping, including aspirational districts. The only way to do any such analysis is to look at seat-by-seat voteshare and voting surveys to determine how different social groups voted for any one party
Contests are seat by seat and state by state: 40% voteshare of one party may not be comparable to 40% voteshare of another party unless there is territorial relevance. Dispersion of votes, number of parties in contest and other such factors can make all the difference in how voteshares translate into actual seats because the contest is first seat by seat (the winning candidate has to get the highest number of votes) and then state by state (winning party has to get majority seats either alone or in coalition).