So on MeWe there has been discussion of a putative association between the Covid-19 vaccines and cardiac deaths in athletes. It is quite difficult to show such a thing with a reasonable statistical certainty and we can compute why this is the case. We know from past studies that these happen rarely, about 1-2 per 100,000 athlete-years. So what sort of sample size would one need to detect a change of 1 per 100,000 athlete-years (a 50 or 100% increase)?
This can be determined if we specify the chance we want to have of detecting such such a change if it is present and the chance that we are wrong when we detect a change.
For a chance of incorrectly detecting a change of 20% and an 80% chance of detecting a change if it is there would require a study with 714,645 athletes. For each you would have to know if they have had a major cardiac event, whether they were vaccinated, and whether they had been infected by Covid-19.
The sample sizes only get larger if one wants to be more accurate. What if we only want to detect a 20% increase in the rate? That requires a smaller number of about 249,525 athletes.
So from this I think it is fair to say that we are very unlikely to ever be able to show with any reasonable statistical certainty that there has been a change. Such assertions based on anecdotes about athletes dying are basically speculation.
This may seem strange to people, but this is what happens when one is talking about rare events. Human beings just aren't very good reasoning about them without the use of math and statistics.
See this article for prior rates of this type of event. And the G-Power program for a way to compute these numbers.