As the statistics at the bottom of the previous page show, there is a great deal of variation from year to year with viruses that continually mutate like influenza or SARS-CoV-2 which causes COVID-19 disease. Over the ten years prior to the appearance of COVID-19, annual U.S. estimated flu deaths ranged from 12,447 to 51,646. The percentage of flu deaths comprised of those in the age range of 0-49 varied from 3.5% to 15.3%.
For the flu season 2010-2011, the 36,656 total flu season deaths were close to the average of 34,152; but for ages 0-49, this age group comprised 15.3% of all flu deaths, more than double the average of 7.6%. In 2014-2015, with a well above average number of total deaths at 51,376, ages 0-49 comprised just 3.5% of all flu deaths. This is why I stated that you can average flu statistics, but no flu year is typical. The variations from year-to-year are unpredictable and widespread.
While epidemiologists create models for infectious diseases to forecast trends, they do so with the knowledge and understanding that a model's forecast is limited by variables which are unpredictable and over which they have no control. Three major variables are the pathogen (how a virus may mutate and evolve in virulence or transmission), the host (level of immunity to existing or emerging virus variants thru acquired active immunity, induced by vaccination or previous exposure, or innate immune response), and environment (level of community immunity, availability and uptake of effective vaccines, public health interventions, public response, etc.).
It comes with the territory that statistical modeling is limited by variables which are unpredictable. Modeling by experts in their field; whether those are epidemiologists, meteorologists, or some other discipline are nonetheless valuable in advising planning and responses to events.
Predictions made by those without the knowledge, experience, or expertise to do so—such as those you find online or from commentators elsewhere—are of little to no informative or useful value to anyone but themselves. As always—buyer beware of the goods being offered.