Fjords Tile Distribution & Drawing Odds

[kothmann]

After @Decar convinced me to get a copy of Fjords, it was inevitable that this post would happen.  But it isn't mere compulsion: I think the information is also interesting.

Here are the tiles, organized by topology of features (Plains / Mountains / Oceans):



The lovely artwork varies on repeats of a given tile type, but the feature connectivity is always the same on a given tile type.  Unlike normal Carcassonne and Land vs. Sea, the topology here is best described by focusing not on the edges, but rather on the corners of the tiles, like the 45-Deg Rotated Carcassonne variant of @wolnic.  So, the tile in the top right is PMMPOO.  Incidentally, in my limited experience, focusing on the corner compatibility makes it much faster to figure out where a tile can fit on the landscape.

With three feature types on a six-sided tile, someone who understands Burnside's Lemma (not me) can tell you that there are 92 possible corner configurations, excluding rotational symmetries (assuming I googled it correctly).  But the photo above shows only 23 types of tiles!  Pretty interesting that the gameplay allows such a limited set of tiles: mountains and oceans are essentially just dividers for the plains and all tiles have a single plains region (except the MMMMMM).  Rather than estimating quantities from the shadows in the image above, here is a table of the distribution (some tiles are marked with a small "3" or "4" indicating the minimum number of players for which that tile is to be used):



The rules require that all tile placements touch at least 2 other tile edges, which means they touch 3 corners.  So, here is a table of the percentage of tiles on which each of the 27 possible three-corner sequences occur:



One potentially useful fact: if a tile is to connect oceans and mountains, there must be plains between them.  Also plains alone or any non-alternating combination of plains and mountains (PPM, MPP, PMM, MMP) are roughly 50/50.  One limitation of the table is that you are not allowed to create a disconnected landmass, so even if a tile fits, it might not be playable (e.g., a POOOOO tile cannot be placed in a 2-tile OOO hole, because the land would be a new island).

I haven't played enough yet to glean much else from this.  The only other observation I'll share is that the revised (2022) rules call players to choose a tile from a "shared hand" of 4 tiles placed face-up on the table.  Having watched several YouTube videos of games being played, this really seems to slow things down.  I'm going to start learning and playing with the original Carcassonne-like rule of drawing one tile and discarding it if it cannot be placed.

Curious to know if anyone with more experience has other insights about the tile distribution.  Thanks.

Linkback: https://www.carcassonnecentral.com/community/index.php?topic=6013.0

[Decar]

It's a common misconception that you need a hand of tiles to MAKE THE GAME BETTER.
I'm not surprised it's been added to this edition with little testing for balance. It feels nice for players to have a few more options, so they don't have to play tactically and have more chance of succeeding with their overarching strategy.

The original Fjords, which also has nice art work was only a 2-player game too.  I've not played it anywhere near enough, and wonder how the addition of extra players affects the game.  I don't know if the tiles were adapted for this new version.  It would be interesting to compare them (I think there are some images on BGG).

But I'm digressing - I'm not sure I can hold all the information in my head given all the tile combinations.

+1 for referencing the POO tile legitimately