DeletedUser8152
Cool, so over the whole span, you found 208 relics, when you would expect to find 234. The expected standard deviation is about 15, so you are about 1.7 standard deviations low. The a priori chance to be off by that much or more is about 10%. So that is unlucky, but I wouldn't say by enough to conclude anything.
Slicing up the data into smaller pieces is tricky, because you tend to make the slices based on what you see, and that can produce significant deviations out of random noise. But in the last 10 weeks you got 90 from an expected 108, which is again 1.8 standard deviations low. Again I wouldn't draw any confident conclusions.
Besides the fairly weak deviations from expected, you also have the problem that you decided to post here based on the results you saw. If other people have done this and gotten the expected results or better, they probably didn't feel compelled to write about it. If you'd gotten error probabilities like 0.1%, I would still think there's something fishy going on, but at 10% I feel like it can easily be explained by bad luck and reporting bias.
Slicing up the data into smaller pieces is tricky, because you tend to make the slices based on what you see, and that can produce significant deviations out of random noise. But in the last 10 weeks you got 90 from an expected 108, which is again 1.8 standard deviations low. Again I wouldn't draw any confident conclusions.
Besides the fairly weak deviations from expected, you also have the problem that you decided to post here based on the results you saw. If other people have done this and gotten the expected results or better, they probably didn't feel compelled to write about it. If you'd gotten error probabilities like 0.1%, I would still think there's something fishy going on, but at 10% I feel like it can easily be explained by bad luck and reporting bias.
