When it comes to actual quadratic voting, the idea is that each voter gets a set number of tokens, and then each vote costs the square of the number. So if you want to cast 3 votes, it costs 9 tokens, 4 votes costs 16 tokens (or +7), 5 votes costs 25 tokens (or +11).

To see why that's nice, consider 10 voters, 3 of which are die-hard for Candidate T, and 7 of which do not like Candidate T but are indifferent among 4 other candidates. With traditional voting, the results would look like:

| Candidate | Votes |

| --- | --- |

| T | 3 |

| A | 2 |

| B | 2 |

| C | 2 |

| D | 1 |

Here the voting doesn't allow the 7 voters to express that they'd prefer any other candidate aside from T. With quadratic voting though, they could.

With 100 tokens and quadratic voting, a die-hard would cast 10 votes for T at a cost of 100 tokens, so T would have 30 votes.

A non-die-hard would cast 5 tokens each for the other for candidates for ${tex`5^2 = 25`} tokens for each candidate (totaling to 100). That would mean all the other candidates would receive ${tex`7*5= 35`} votes, easily beating T.