In regards to reading your opponent...
It's worth something, but not as much as some people would have you believe. If your opponent looks frustrated, that is a piece of information that you should take into account when making decisions. How much does it count for though? Without knowing their game plan at the moment, you don't know WHY they are frustrated. You might have an idea, but is that really reliable? Maybe they are frustrated because they were hoping to draw an Overwriter and kill you, and they missed the obvious win they had in hand. Meanwhile, you assume that they don't have the win in hand and are frustrated that they didn't draw a card that they actually do have. I find it better to use information like that as a general feel for the flow of the game. Maybe it's smarter to take a few small risks in order to put them on the back foot rather than play it too safe. Obviously they are not happy with their hand, and pressure will probably be a good bet. But it was probably the right move anyways, without that little bit of intel.
Look at Mushin No Shin itself. If your opponent plays Hedge Fund and Mushin No Shin turn 1, what do you do? Let's try to get into his mind. Assume we have no way of bringing him to less than 3 credits, so Overwriter is live. Is he the kind of player who would play Ronin first or Overwriter? If it's Overwriter, and we don't run it, does he Mushin a Ronin next, or an Overwriter, or an agenda? Does he assume we will run the second since we didn't run the first, or does he assume we won't run the second because we didn't run the first? If we refuse to run his remotes, does he continue to play out traps assuming eventually we will, or does he play agendas assuming we won't?
Or look at the Psi game. If the opponent always chooses randomly (rolling a die or some other random determination) and we know this, bidding 0 is always the correct play, as any number gets us the same chance of beating him, and 0 does not tax our credits. But what if the opponent is consciously choosing a number? Two is least favorable for either to choose, but it's probably the number each side chooses the least, and therefore a better bet if the results of the game are important. If the runner stealing TFP will win them the game, corp is incentivized to choose the safest bet, which is 2, but the runner also knows that the stakes are high and therefore 2 becomes more attractive as a choice, negating most of the reason for the corp to choose it. On the other hand, 0 and 1 are more common choices, and the runner may choose instead to just keep taxing the corp by accessing TFP and letting them waste money while he spends nothing, so choosing 0 is risky too. One is a reasonable middle ground for both sides monetarily, and therefore has risks for the Corp as well.
What information can we use to inform these decisions? If someone choose 0 three times in a row, how likely are they to choose 0 on the 4th decision? Are you willing to bet your entire hand on your ability to read his mind, or the entire game? Put 10 credits in front of you and a friend you know well, and play the psi game with them over and over. Do you think you could win more than 33% of the time?
My point is that games are won by sound tactical and strategic decision making for the most part. My last tournament playing Jinteki, I lost 4 TFP psi games in a row betting random numbers, one in which my opponent randomly rolled for his result as well. Did I fail to "mind game" my opponents properly? I think the more likely explanation is that I could have tightened up my technical play and I also got somewhat unlucky. I don't think my opponents had a "read" on me.
The other thing to consider is that if you lean on reading your opponent as a strategy, and then you fail to do so, you often give up more than you needed to, and open the door for them to win.
Playing against Jinteki can be a series of chances for you to misplay and lose the game, but if you don't misplay, the only way for Jinteki to win is to keep presenting you with 50% chances to lose. The chance that the runner will win more than a couple 50% chances to lose is low. This doesn't involved mind reading, its simply setting up two decisions, and leaving the rest up to statistics.
Example: Playing as Gabe, against Jinteki. He installs a card in a remote, then Mushins out another card. I play Quality Time, then run the non advanced. It's psychic field. I lose the psi and my hand. Now I have a 50% choice... the two cases that matter are: the advanced remote is Ronin, in which case I lose if I don't run it, or it is Cereberal Overwriter, and I lose if I do.
What's the right decision? Is there something I should base my decision on? I do believe that it's just a 50/50 shot. That's the strength of Jinteki... not "tricking" your opponent into doing something stupid (that happens, but it's not something you can rely on, unless you never play against strong players). When you take away misplays, there is only the 50/50 decision.