2) I think that's valid, we would expect reports to go up. But Kirsch limited his search to ages 20-60, I think to alleviate one of your concerns. We're seeing e.g. 3000x normal, 1900x normal occurrences even when filtering like that. Does that not seem alarming?
The people who call in a bunch of unrelated leads during a murder investigation also aren't uneducated; it's perfectly reasonable, expected human behavior that when people are looking for trouble they find it more than if they aren't expecting any connection. Whether or not the connection is actually causal.
2) insufficient data to know if it should be alarming, because outside of a pandemic circumstance, the 20 to 60 demographic doesn't get very many vaccinations. So you're comparing on other years a population with possibly very different behavior to the general population (for example, one explanation could be that there's correlation between people being healthier and people getting vaccinated on years that aren't pandemic years... Possibly because the kind of person who goes out of their way to get vaccinated when it isn't required for school attendance is the kind of person that cares about their health more than average, is more prompt on upkeeping preventative medicine, has access to the kinds of doctors that catch early signs of heart disease before they evolve into full-blown conditions, etc.).
Edit: this has been unjustly flagged.
To give an absolutely terrible, but perhaps useful, analogy... You're an old sea captain on a sailing vessel without modern weather technology. VAERS is the crewman whose knee acts up about half the time before a bad storm. At best, if he tells you his knee's acting funny, you put your weather eye out... You don't turn around and go back to port.
... and that's the problem. The general populace doesn't know how rigorous medical testing works. And when they get their hands on something that sounds serious (but isn't actually at a scale that causation can be confirmed), or someone with an axe to grind decides to push a counter-narrative that doesn't stand up to scrutiny (but who cares whether it does if the general populace doesn't know how "scrutiny" works and is weighing the evidence of the medical community and the CDC vs. one pathologist in Germany as if their words have equal weight), it becomes a general problem for public health.
One that, apparently, Google has decided it doesn't want to be part of.
1) self-reporting systems are subject to memetic effects. Like people calling in every suspicious thing they see to the cops if they hear the cops are investigating a high-profile murder, adverse effect reporting increases if people hear a rumor a vaccine may cause problems, whether or not anything in their day-to-day has actually changed.
2) The push to vaccinate everyone means a lot more people are candidates for reporting than in an average year. Given no side-effects from the vaccine, we would still expect VAERS reports to go up significantly. And since heart disease is a leading killer in the US, and unlike most vaccines, this vaccine is being given to the older population almost exclusively relative to the median age of most vaccine recipients, we're hard-pressed to distinguish "vaccine-caused heart disease" from "greater amounts of reporting from older patients resulting in more instances of heart disease showing up in the VAERS reports."