I am partway done my Thurn and Taxis league games and all is well so far. Last season I won my way up to league 2, and I was concerned that this would prove an extremely difficult test. I am sure I have played less T&T than most of my opponents, as I have only played in the online leagues and the heats and semis at the world boardgaming championships.
But my worries were unfounded, or so the evidence suggests. I won 2 games so far, one of them an absolutely brutal smashing 31 - 18 - 12 - 6. I rewound through the game to see if I constantly just mized into the best cards for me and it didn't seem outrageous. I suspect that the reason for my domination was that I got a bunch of perfectly solid routes and did all the things, and some of the stuff my opponents had to do to not trash their routes were really bad. I definitely threw some plays in the trash, with a bit of meandering about the board, but it was at worst a solid play with a bit of waste. I never had a 7 route that only put 3 houses on or anything like that.
T&T is one of those games where I can't quite figure out how to play properly. I know that really good players win very consistently and yet it so often feels like I win or lose based on the exact card I need flipping up at the start of my turn. Clearly skill mitigates much of that randomness, but how precisely you do that isn't something I can articulate. I am doing well in a competitive league though so clearly something in my brain knows how to play T&T. I just can't tell you what it is I do!
One thing I have noticed this season is defensive plays that I would not have expected. In particular I have seen people scooping up a second copy of Pilsen that they have no intention of playing just to jam the other players from completing their rainbow. It also prevents red/orange completion to of course but that is less common and so far less important.
You can see this on the board above - if you take all the copies of Pilsen then nobody can get to Lodz and thus cannot get the brown component of the rainbow. That is good an all, but the question is does it actually make you win? Tossing away an action is worth roughly 1 point as games mostly give players about 16 turns, which is 24 actions, and the winner gains about 45 points. So the question becomes: Is losing 1 point worth it when you cost other people points?
Mostly what you will cost other people is 4-5 points for completing their rainbow, but the trick is that they can mitigate this. Instead of focusing on the rainbow they can aim to complete other colours instead. Swapping to a plan of completing beige when Lodz is blocked is a fine way to gain points, particularly since most of the beige towns are low priority. Considering this you will have to assume that the real effect of your action is far below the 4-5 points they technically lose.
The other problem is that you might not succeed in your block. If you have to place Pilsen as part of a route (and you bloody well need to, if you have two copies of it in your hand!) then the second time through the deck people will end up with another shot at it. The only way to crush people really thoroughly is to keep those copies of Pilsen in your hand until the deck reshuffles and that is going to destroy your game to a far greater extent than a single point. If you lose your point randomly stealing a card from other people but they manage to get their points later in the game anyway you are losing.
I think the best way to think about this is that a Pilsen block costs somebody about 3 points. You aren't even necessarily sure which somebody, and you also usually don't know for sure that it costs them 3 points. That is a pretty rubbish play, in my mind. You don't win 4 player games by losing 1 point to randomly punish another player for 3 points, particularly when the punish isn't certain. Somebody else is going to play greedy, dodge your punish, and beat you.
However, I do think that there are places where you can maximize your investment. For example, if somebody grabs Lodz early in the game on spec you can snatch the last Pilsen and put them in a bind. Either Lodz clogs their hand until the deck reshuffle, which is awful, or they chuck it and waste their own turn. This is a much bigger punish and you can target it at the person who you think is the biggest threat. Of course I don't think strong players will usually let themselves get stuck in a position where this sort of block is viable.
Generally I see the option to flush the up cards and generate a new pile to draw from as a much more powerful option if you want to hate people. It is often possible to ditch multiple key cards at once, and if you do it right you can hit several people at once. You don't keep the cards hidden safely in your hand of course but you can get more of them, and it might be just the move you want to do anyway.
Thinking about this I have decided that I shouldn't bother trying to punish people unless the punish is responding to specific board state and targetted. Randomly snagging copies of Pilsen (or Lodz, or Sigmaringen, or whatever) just isn't worth it unless you are going to get good use out of that second copy yourself.