There are too many variables and too few facts in the OP to be able to come to a reasonable conclusion. But...one thing we can count on is that amazon acts in its self interest. If they are in a wholesale review removal mode, then the reason for it has to be to help them maintain or increase book sales and thus profits. Ultimately that should help us too in a generalized way - or at least I would give it a 90% probability that it will.
As for gaming the reviews, there is one fairly credible tell that you can look for. There is a normal distribution of reviews that is called the "J" curve. Plot the reviews on a graph and generally (unless the book just sucks or is truly outstanding) you will see the J curve. Most of the reviews will be 5 star, the high end of the J. Then a tailing off of 4, 3and 2 stars, then an up tick in the 1 stars.
One theory for this distribution is that the people who are most likely to go to the trouble of leaving a review either love the book, or hate it. The people in between those extremes tend to fall off in numbers from apathy about the book. The folks who think the book is mediocre are the least likely to make the effort of leaving a review.
So look for a pattern similar to this - 20/5 star, 12/4 star, 8/3 star, 4/2 star, 12/1 star.
It is not dispositive of course, but it is a good early indicator that a distribution that isn't in a J pattern may be gamed, one way or another. Watch the distribution as you surf books and you will see the J curve more often than any other distribution.