If people buy it anyway at the full price, then the game publisher will correctly deduce that it indeed worth at least that much money for enough people (otherwise those people would not part ways with that much money to get it) to get that game as soon as it comes out.

In Economics, perfect pricing (which is not yet possible but, damn, they're really trying hard) from the point of view of a seller (i.e. for maximum profits) is when they get exactly as much money from each individual as that person is willing to pay for it, so the "ideal" world for them would be individually-tailored prices going as high as it could possibly go for each person whilst still managing to sell to that person.

As they can't as of yet sell at different prices to each and every individual, they've gone as far as they can (regional pricing, different prices in different stores with different audiences and, maybe more importantly, time-from-publishing pricing) and then push prices up and up slowly whilst checking if in total the price increase has yielded more money or not (they have no issue with loosing customers due to higher prices if in total they still make more money at the price point than at a lower price point).

IMHO, in the face of this, the easist and best reaction for somebody who wants the game but does not think it's worth $70, is to wait until the price falls down to how much they're willing to pay for it (even better, let it fall some more and buy a couple more games with the savings). In fact if enough people do it the price will fall much faster as the publisher's sales data analysis will signal to them that they've put the game at too high a price point and they'll lower it trying to pick up the "money left on the table" from those who are interested but not at that price point before those people lose interest.