I also think it is worth noting several other factors that go into what makes a market 'overpriced' 'underpriced' or 'fairly priced'.
First, let's think about cap rate compression. If a property, all things being equal, was trading at 8% cap two years ago, and is currently trading at 6% cap, then you have to ask yourself why? It's not magic, it's not some mysterious third party deciding what the proper cap rate is, it's a market. The market forces which move prices imply many things. (Not that everyone agrees with the implications, which is causes there to be both buyers and sellers) One of the things cap rate compression tells you is that there is implied rent appreciation. As an example, let's say that a property of a certain type has historically traded at a 8% cap in that market. If you believe in reversion to the mean, and it's currently trading at a 6% cap rate then the 'market' is telling you that rent appreciation is expected to be above the historical rental appreciation for that market, such that it will make this property have an NOI that reverts to a historical level close to 8% in the future.
It might seem counterintuitive at first to think that buying at a 6% cap today to 'grow into' an 8% cap later makes sense. If you think of the cash flow and the 'rate of change' of rents in the future, it makes sense.
Secondly, since rent appreciation isn't the only thing that matters and because real estate investing isn't in a vacuum, then it must imply other things as well. It may imply that the premium over the risk free rate of return is a constant (i.e. if 'normal' is 4.5% 10 year treasury rates and 'normal' is 8% cap rates, then the 'normal' premium over the risk free rate is 350bps) so that the expected 10 year treasury rate + 350 bps is 6%. This means that the expectation is that inflation is tame (the 10 year treasury rate is a historic inflation + gdp market expectation). It may imply that availability of credit for first time homebuyers is limited in the near future (which leads to rent appreciation). It may imply that buyer preferences are for rental properties instead of home ownership (i.e. geographically closer to work, a preference for community living instead of suburban living, home price unaffordability, etc) which again leads to rent appreciation.
You can debate these implications, but the information included in the price movements definitely means that more people / more money believes those implications than do not. It's also worth noting that changes in how the market views the implications do not happen over night so the odds of the 'bubble bursting' are very low. In fact, in my lifetime it's happened ONCE (2008 crisis) nationally. Locally, it happens more often, but then again, real estate has always been considered a local market up until the crisis.
Side note: Isn't it funny that people refer to the 2008 crisis as a ONCE IN A LIFETIME MAGNITUDE CRISIS, but then fear it happening again all the time.