No, can you explain? What is wrong some citizens voting on one race, and other citizens voting on another one that they care about more?

Every citizen would still have the same exact voting power, so there would be no change in the power dynamic. This would just allow people to express their vote only on what they know about or care about.

Obviously people could use this system to vote exactly as they do now, equally spreading votes across all races for instance. But there would be no more "wasted votes" in this paradigm like there is now.

Yes, voters could be strategic about their votes, perhaps saving them for a certain election or candidate--but that's the whole point. If they did that, they would have no votes on anything else. In other words, the winner of the election is the candidate (or decision) that enough people deeply cared about that they sacrificed many other things they care about less.

It's also important to understand the downsides of today's elections, wherein our complex society gives voters an impossible number of options that no individual who had a daily job would ever keep track of. In today's society, probably 50% of people's votes are almost entirely uninformed--and probably 90% if you count party-line voting where people know nothing about the specific candidate. This gives enormous power to parties, and also makes money much more decisive.

For instance, take an election for the local water commissioner. Hardly any citizen would know this person's name, or know what they did. But a small constituency would know, and would know what is at stake for this position, e.g. the industrial use of the local water supply, pollution controls, etc.

In the current system, large monied interests could easy buy this election, because they could flood the airwaves with soundbite advertisements that skipped over any of the technical details of the decision (e.g. they could tar the opposing candidate with smears, or just insist that their candidate was for "better water" without any explanation).

In the tokenized election scenario, most citizens would not vote on this subject at all since they would have absolutely no idea what to vote on. However, a small number of citizens (and not industrial concerns or not outside money) would probably know a lot about the issues of this specific decision, and would vote on it accordingly. Yes, some citizens could be in line with the industrial concern, but there would likely be others who would be alive to the risks and vote against them. But the election in this case would be decided by a relatively small number of citizens which correctly mirrors the knowledge levels for once.

In this new system, the mantra is, "only vote on what you know, and don't waste your vote on something you don't".