Two problems with your argument...
1) You're arguing apples and oranges with yourself, I noted that my CDC graph is cumulative but so is the flu data that was presented for the previous three years.
2) Your claim that flu is "orders of magnitude" worse is flat out false, you really need to show that you understand what those words mean.
....keep spinning your wheels
Math is hard. Reading is hard.
Time to go over everything in more detail. First off, I already linked to influenza hospitalizations for kids aged 0-17 (They broke it down into 0-4 and 5-17 age groups). Not difficult, straightforward data: total number of hospitalization cases and overall hospitalization rates per 100K. That's how its done properly. Straight out of the CDC database.
Now, lets look at the graph you posted, along with its associated links. You think its showing something similar to the above, straightforward method, but it is not. It is actually completely different, with different goals behind it. If you read the cdc's website, your graph is showing a cumulative total of the average rates, not overall rates based off of the raw, collected data. Its also looking at a very narrow time frame. Now, let's go over each axis to further illustrate.
Let's start with the x axis. First off, notice there are no dates, no months, weeks, whatever, that would show a definitive time frame. Why is that? Again, and I hate to keep harping on this, read what the CDC says about your graph, It is based off the number of weeks after the virus is initially detected and tracked. In other words, they don't have the same time frames to compare: January data from one year does not necessarily match up with January data from another year, February does not match up with February, etc. If covid's tracking starts in February, influenza for one year may start in September for one year, November for another, etc. Now, covid tracking in your graph arbitrarily starts at 40 weeks into the data tracking, What was going on then? Still no vaccine for kids. Influenza on the other hand, usually has vaccines developed and/or that strain has died out naturally. The graph is labelled surveillance week. It could more accurately be described as Weeks after Initial Tracking begins, but surveillance week isn't horrible. Just remember that time frames do not equate to similar times of the year, but even more importantly that the data being looked at is markedly different for each pathogen in their cycles.
Time for the y axis. First off, the label is Hospitalizations per 100K. That's an inaccurate label. It should be Cumulative total of average hospitalizations per 100K, not the truncated version they used. Of course, with the more accurate label its readily apparent the graph takes on a completely different meaning. In your link, they don't show the weekly raw data that they use, but they do show the monthly data. The problem is, they never show the total number of cases like they do with the flu that I linked to above. Instead, they rely on weekly averages for arbitrary snippets in time, and then total those averages. Not only is that different, manipulated data (they don't collect averages, they calculate them), but that is not how overall rates are calculated, which is what you and the people who posted the graph are claiming.
Again, all of this confusion could have been cleared up by reading the graph notes and cdc reasoning for the graph. It does not calculate total rate, it calculates a running total of averages for a partial and arbitrary time frame, They are completely different sets of data and calculations. If you read why the cdc does this, they openly state they are not calculating the totals you and the graph authors claim they are showing. They readily admit a small change in total numbers can cause a significant and disproportionate change in the graph. They openly admit they are not trying to show total cases, but instead use this method to look for changes for other purposes.
The CDC is very open in that the graph does not show total cases, and is not meant to be used in that manner. The publishers who tried to frame it in that context had the lazy and ignorant in mind who either couldn't be bothered to read (or even search for) the fine print, or didn't have the requisite skills to understand it. Congratulations on being part of their target audience.