No. It will not be a better platform if the flagging feature is removed. Flagging is the only mechanism by which an over-rewarded post can be corrected. If you agree that the ultimate goal is to have a good distribution of the reward pool as to encourage adoption of the token, then you must also recognize that in the absence of flagging, the system boils down to an N-person prisoner's dilemma instance when it comes to voting. The main properties of this are:

1. A stakeholder will earn more if they use their voting power to solely vote for themselves.
2. If every stakeholder voted selfishly for themselves, the distribution of the rewards pool becomes flat and it is harmful for the platform, as there is no chance that new players will enter the system.

Some people dispute (1), claiming that actually there are ways to earn much more than self voting, however I believe this to actually be an illusion, involving community groups that give higher than self-voting returns, but this ultimately comes from various community members not maximizing their portion of the community stake.

But the main point of this setup is that in the absence of flagging, stakeholders are incentivized to engage in selfish behavior. In a system in which all participants behave rationally, the ultimate configuration will be that all participants will self vote.

The presence of flagging, or the threat of flagging, introduces an extra element where cooperators can correct the behavior of defectors (self-voters), and indeed there's quite a pressure on large stake-holders not to self vote. The white paper argues that this flag "solves" this problem, and prevents the voting problem from being an instance of Prisoner's Dilemma.

A more realistic look

Where this simplistic view falls short, I suppose, is that even in the absence of flagging, it's possible that a prisoner's dilemma does not exist in other ways. For example, there is a time-dependent element to the evaluation function concerning the output of the game. Many people believe that restricting themselves from self voting, and creating communities that encourage the spreading of votes and rewarding valuable contributions will lead to long term growth of the token, and subsequently act in a way that maximizes the long term value.

We have an interesting set of factions on this platform. There are certainly those trying to milk the platform selfishly as much as possible. Delegating to bid bots is a form of this, though well-hidden. And they turn around and cash it out.

On the other hand, there are those that recognize that there's not enough power to correct for much of this behavior, and the delegations even make it trickier to effectively target the benefactors of this. In certain factions, they are using stake to spread to many people, but they also self-vote to boost the future effectiveness of their stake. In other words, they are rather in the middle: self-vote a little to not lose too much ground to pure self-voters and still spread some of the stake around to the wider community.

And then there'a subset that uses flagging to correct for various abusive behaviors on the platform. There are cases where a whale with a lot of power can get a little too much say here, but in theory as the stake gets distributed further and further, there will be a better ability to counter whale votes. Since stake weighting is linear, multiple smaller parties can combine their stake to counter the actions of a larger party. But in the early stages, with an imbalance of stake the effect of this is rather limited.

In any case, this is all a very interesting experiment to see. Whatever the situation is, it is an instance of a complex multi-person game, and it will need a lot more nuanced analysis to determine how best to make changes to it to encourage a better outcome.