This is a great article on the same topic that somewhat agrees with me:
https://hackernoon.com/popping-the-bubble-blockchain-and-cryptocurrency-7130156f91b2

Here's what this article says we should look for before we are actually concerned with a bubble:

But this is to simply acknowledge that predicting the top of the bubble is a fool’s game. Instead, we should focus on the identifiers that scream “warning.” Here are the indicators:

1. Mass media begins focusing not only on Bitcoin (which they already are), but on DLT, on its potential, and on the associated projects. This will signify that a large portion of the population is becoming aware of the underlying technology — far different than today’s situation. CNBC is quickly making this a reality…

2. A flood of institutional money: hedge funds, retirement accounts, personal savings. This will result in a dramatic market cap increase.

3. Working blockchain products that are actually supporting large networks of users. As we progress into this development period, it will become clear that many of today’s projects are failing to live up to expectations. The first failed blockchain project will create a snowball effect.

4. A large influx of private, centralized blockchains produced by existing companies. This will demonstrate that many previously predicted, decentralized DLT solutions will in fact be overtaken by existing, traditional companies, developing their own, private blockchain solutions instead of adopting decentralized, token based platforms. This will likely result in a large reevaluation of what the reality of blockchain actually is.

5. A market cap between $5 and $10 trillion. Anything of that magnitude any time soon would be very worrying. However, it is important to understand that velocity is what creates a bubble, not total marketcap. A quick run up will almost always be followed by a dramatic crash.

Another thing he doesn't mention is leveraged investments. A bubble can't happen until ma and pa bet the farm on crypto. Actually I guess that's covered by #2. I think this article is very well thought out.