I've been a fan of Stefan Feld since my first game of his; Castles of Burgundy. And while over the years his designs have come and gone (and in some cases come back), Castles of Burgundy remains my favorite to this day.
I can't say why that is....most of his games tend to focus on point salad to the nth degree; and even a game like Oracle of Delphi, which technically is goal oriented...manages to feel exactly like a point salad game. He also loves dice in most of his games, and yet not a single one of those games feels luck based, rather they feel like it's very much up to the player to pick at the thread they've been given, and to unravel towards the endgame.
So why is Castles my favorite? I can't say....I know Bruges is a close second for me, but Castles still manages to most delicately balance risk, reward, strategy, luck, and variance for me. In Bruges I can argue that the cards you draw can sometimes make or break your strategy, especially in the end game. In Oracles of Delphi it can sometimes feel a little bit too disconnected, as if the other players (almost) don't matter. In Notre Dame those darn rats feel just a tad bit too annoying. I can keep going but you get the idea. In nearly every Feld game, both those I love and own and those I don't...I can point to something I don't like, or that I wish were marginally different/better. In The Castles of Burgundy the only thing I can point at is the art.
Speaking of the art, as the picture above illustrates...I currently have the new edition with the new artwork. And while it's not as bad in person as it initially looked, I'll still say that Castles is a game I like despite the artwork not because of it. And that stands true for both editions.
The Castles of Burgundy is one of my favorite games of all time. I can't wait for The Castles of Tuscany, one of the next Feld games...I'm very intrigued.