I played Terra Mystica for the first time this weekend. It was fun, a very dense game. It made me think of Settlers of Catan crossed with Seven Wonders, except that there is no uncontrolled randomness: no dice, no hidden cards, just player choices. The headlining feature is that it has 14 asymmetrical factions. Having read a bit online, most people agree that the factions are imbalanced. They just disagree on which factions are weak or strong.
There are some consistencies. Many people argue the factions are balanced enough, so it does not matter below competitive play (and there are balancing methods there, since you can start with more or fewer victory points). Most ratings have some overall similarities: Darklings and Mermaids are strong, Giants and Fakirs are weak. Then you start seeing radical differences in assessments, some of which are backed by data, and some of the data seems conflicting (which should diminish at larger sample sizes). Grabbing the first few examples I found on Google: are Cultists “worse than terrible” or the winningest faction? Or darklings were mid-tier those statistics but #1 in this larger pool of statistics. I see Alchemists ranked both in the top 4 and the bottom 4. Or this thread is a good example of divergent opinions being professed with or without support.
The answer I favor after a lot of reading is “it depends,” which has more nuance than you might think. Like Smash Up, some factions do better or worse with more or fewer opponents. Some do better or worse against particular other factions or combinations of factions. Some do better or worse on particular maps, or with particular bonuses in the game. Some are more straightforward while others require unusual strategies or expert play. Some factions will be better than others given your usual play environment. This is especially true if your usual play environment includes people who consistently pick the same faction or strategy and are particularly good/bad. I am convinced that most cries of imbalance in games can be traced back to who is the best player in your gaming group and their play preferences. That differs, so different players see radically different results on whether a game favors or penalizes combat, quick expansion, or a particular branch of the tech tree.
Which is not to say that there cannot be imbalances. Maybe one degenerate approach is particularly strong, or it is easier to excel with a certain strategy, or a configuration that favors X is more common. One of the fun rankings is those links is about how resilient the faction is: some are situationally powerful but often weak, while another might be a consistent B across most options. Of course, given that Terra Mystica is a game with all information available at the start, reading the board and knowing which faction to pick (and factoring in others’ picks) is an expert-level skill. See also players who refuse to think about team composition in League of Legends or switching to counters in Overwatch.
This can also poison data-driven rankings. The strongest faction could easily come out ranked mid-tier because everyone knows it is the strongest faction, so the weaker players flock to it as a handicap, and they still lose. Meanwhile, expert players both know their situational strategies and recognize those situations, so some of the weaker factions can rack up wins punching above their weight. When someone makes a weak pick, you rarely know if they are too bad to know better or too good for you to see their reasoning.