Think of it this way. Let's say that to get 1 Bitcoin I could just install an app on my computer, and within 1 second get a Bitcoin. Then everyone in the world can make as many bitcoins as they want, there are no limits on the number of bitcoins, and the process to get more doesn't increase in difficulty.

So now we have bascially infinite bitcoins so the value of them is zero, because why would I buy one, when I can just make one at any time.

The opposite of this is where Bitcoin derived it's value. First there is a fixed number of Bitcoin, so you can't simply make more past a point. Secondly, the effort required to mine Bitcoin increases over time.

This effort requires real world currency exchange. I need to buy a powerful computer and then pay a lot of electricity, cooling, space.

So effectively my mining operation costs $x. Now as time goes on to get the same number of Bitcoins I need to increase my spend because the computational needs increase as well.

Effectively, I'm exchanging my USD to equipment and energy, which are giving me Bitcoins.

A reasonable value for Bitcoin would be to calculate the total amount of hardware and energy invested in mining and there you have a decent floor value. You can then extrapolate this for the future, since the total amount of Bitcoin to be created is known, you could say that multiplying that out you get a total amount of value.

To say that they are overvalued would mean that the total current dollars invested in mining bitcoin is far less than the current total market cap of said coins. If you look at the price of Bitcoin 2014 to the start of 2017, you see a gradual increase in price, which would be explained by the amount of money going into mining and with some markup.

In 2017 you see a 20x increase, but that isn't matched with a 20x increase in spend for minting Bitcoins, so the increase is irrational.