I’m sure there are plenty of exceptions, but my general observation is that coaches at top programs tend to do well immediately or very quickly, then stay a long time and are heralded as top coaches. And maybe they are. Coaches at good mid level programs (like CU) may either get on a roll that carries onto their successors (mac1-Neuheisel-Barnett) or drift into mediocrity or worse that also carries on (Hawkins-embers-mac2???).
So how do we know when it’s been too long? 5 years with no evidence of improvement (Hawk) is enough. For MM, skipping 2016 would place him perhaps a little ahead of Hawk - but 2016 did happen and bought him 3 more years IMO. Another decent year or two will buy him more time. 5-7 (if that happens) is not a decent year.
So I don’t have a massive database to confirm or reject your claim that coaches are often held for too long, but I suppose that fits any coach who never has a successful (winning) season, so almost every coach in a weak program is therefore kept too long by definition. And of course this is almost never an issue for the relatively few elite programs.
Given the benefits of stability for recruiting and retaining players, adjusting to the program culture, and creating a routine for directing a football team, every HC replacement is a setback / reset to some extent, and therefore confirming that a coach is unsuitable is a decision approached carefully. For this reason (as well as contract economics) I don’t see MM gone until after 2019, and then only if he has 3 straight losing records.