Indeed it is the proper thing to do, to read out the numerical vote results for the members to hear -- see p. 417, line 18 ff. - and to include them in the minutes
Consider some possibilities:
1) The winner got nearly all the votes and the loser has had a long history of fruitlessly running for office. Reading the vote count might send him a message, that it is time to quit making a fool of himself.
2) The vote is "reasonably" close. This way the loser will be encouraged to try again, as it seems, by the vote, that he has a good deal of potential, and many friends, but just went up against a better person this time. This may help to keep a good candidate in the game.
3) The vote is "extremely" close - one or two votes different. The assembly may very well want to order a recount (RONR p. 419, line 1, see index also) just to be sure of the result. This way there are no (or fewer) hard feelings.
4) The president, when declaring who won, makes a simple mistake and names the wrong person, or he does not understand the vote required to adopt the motion (majority, 2/3, &c.) and states the "wrong" outcome.
5) The tellers make an error. Reading the results out loud may not help to catch this but studying the printed documentation in the minutes at leisure probably would. The documentation would also serve as evidence if there were serious questions about the outcome.
Without the teller having read the numbers, how will anybody (except the teller, if he is paying attention) know to correct this?
6) The winner of the election (or partisans of the winning side of a critical issue) could weigh the numerical results in terms of whether they have a "mandate" to proceed at full bore, or whether there might be some fence mending to look after first.
If the vote results were not made immediately available to the membership, none of the above good things could happen.
And this listing doesn't even mention the myriad possibilities for knavery or outright fraud that are available when vote counts are kept secret.