Uh, no.
The reason analysis is performed is to achieve overall understanding of a situation, to drive an action. The reason behind wanting to understand a situation could be anything - problem solving, finding cost savings, driving employee engagement, identifying areas of risk or opportunity, etc. A good analyst will include ALL data in their models, so that they can use different approaches to do their work.
F'rinstance: Your productivity metric, to begin with, is worthless in problem solving, except to perhaps highlight that further data is required, in the event that it slips below targeted levels. As soon as it slips below target, you will need to bring back all the variables to get to root cause. Furthermore, it can easily hide problems, if no other variables are checked - your productivity numbers might be fine, but if your quality control levels are achieved are bad, your customer satisfaction will drop, and your level of sales will plummet. Sure, ya got good productivity, but your plant closes because you're sending out garbage. This is the danger of what is known as "Watermelon metrics" - green on the outside, red on the inside - meaning that while the target is being met, it's either directly or indirectly leading to a negative result for the company. I'm sure you can imagine why Florida might be tempted by Watermelon metrics right now.
No successful company in the world limits data analysis to views that tell them they're always right, and no competent analyst carves metrics in stone.
I'd expect someone with 40 years of analysis experience to know this.
Also, your notion that one should only do their job is ridiculous. Going above and beyond is at the core of anyone's journey to promotion, no matter what their occupation is. Challenging and scrutinizing models is an analyst's job - not only was it allowed in every private sector job I've had as an analyst working in Fortune 500 companies, but it was encouraged and rewarded, especially if the challenging and scrutinization leads to greater visibility of organizational workings and results in a way that improves overall process or profitability. I have been promoted for doing EXACTLY what this woman got fired for, because simply changing a dashboard caused people to look at a problem differently, leading to the realization of millions of dollars in cost savings. At worst, even if nothing new was discovered, it would have lead to greater confidence in the metrics being used, having stood up to greater scrutiny.
Sorry, there is nothing in this scenario that speaks to a prioritization of analysis over messaging...and that's the most generous appraisal. As an analyst, I'd say this looks more like a bunch of hacks trying to bury their responsibility behind a curtain of numbers. If you truly were an analyst for 40 years, you know how often and easily that happens as well. Being fired for providing greater clarity through better data is HIGHLY suspect. Incompetent leadership in this situation is, sadly, the best case scenario.