Take, for example, an electronics or hardware project. The majority is a bunch of experienced engineers who will tell you that your project will take years based on their own experience with an earlier project. What they neglect to account for is that their earlier project took years because of the state of the art at the time. They're taking experience that applies to the era of NAND flash and board bring-up and applying it to the era of eMMC and main-line Linux drivers. So while their estimate is very conservative, it is also very wrong and you should take it with a grain of salt.

Any judgement that relies on some external context, like the industry state of the art, can no longer be trusted in a different context.