SPIEL '17 Preview: Azul, or Tiling Up Your Hopes and Dreams

I've already posted an overview of Michael Kiesling's Azul, which Plan B Games will debut at SPIEL '17 in October, but now I've played the game many more times since then — ten total on a pre-production copy and a review copy, four times with four players and six times with two — so let's talk about it some more.

In a recent preview of a new edition of Alex Randolph's Venice Connection, I talked about that game's Nim-style gameplay. Nim is extremely basic: Start with three or more heaps of objects, then take turns with another player to remove any number of objects from precisely one heap; whoever removes the last object wins. Unfortunately, Nim is also solved, which makes it uninteresting to play — but the framework of Nim gives you a great structure upon which other better games can be built.

Azul is one of those games.

Now, the premise of Azul is ridiculous. You're supposed to be a tile-laying artist who has "been challenged to embellish the walls of the Royal Palace of Evora" in order to impress King Manuel I, but at most you'll complete a 5x5 grid of tiles and in most cases you won't have more than a couple of lines of tiles on that grid with numerous holes throughout, and I daresay that only the daftest of kings would be impressed by such a half-assed display of tilery. Perhaps that 5x5 grid is meant to suggest some larger tile-pattern that will be used throughout the palace, but even then you think the king would look at your unfinished work and suggest that you'd be better off as a sheep herder.

No matter. The premise of the game is mere window dressing on what's going to get you to the table — the wonderfully chunky colored bits — and keep you coming back to the table after that first playing, that being the Nim-style gameplay alluded to earlier.

At the start of each round, you fill up five, seven, or nine discs (depending on whether you have two, three, or four players) with four tiles drawn at random from a bag, then place a first player marker in the center of the table. You take turns choosing a disc or the center of the table, then taking all tiles of one color from this location and placing them in a single row on your personal player board; if you chose a disc, then you push any remaining tiles into the center of the table. If you're the first person to choose the center, you take the first player marker along with your chosen tiles.

The start of a three-player game; what do you want?

If you can't fit all of the titles into your row of choice, with the rows being 1-5 spaces in length, then excess tiles "fall" to the bottom of your player board, with you being penalized for such waste at round's end. You can add more tiles to a row you've already started as long as all the tiles are the same color.

You take turns choosing tiles until they've all been claimed, then for each complete row on your player board, you move one of those tiles into the matching-colored space on the same row in the 5x5 grid. You score points for each of these moved tiles, and if you can cluster those tile placements, you score more points. Tiles in incomplete rows stay where they are, while all excess tiles from completed rows are placed in a discard pile. Continue to play rounds this way until someone completes a horizontal row on their grid. At the end of this round, the game ends and you score bonus points for completed rows (not much), completed columns (worth more), and completed colors (the best of all).

That's it — Azul in three paragraphs, three dry paragraphs that don't get across the wonderful tenseness that develops during the game. In that first round, you're mostly trying to grab whatever seems best. Can you take three tiles of the same color? Then do that. Can you take only two? Then do that. But wait? Which tiles are being pushed into the middle, and how many of them lie on other discs? Which color did the player before you take, and are you giving them a gift of tiles on their next turn? You feel things out, take this or that, then the round ends, you score a few points, then the game picks up from there.

Now you all have investments, whether tiles in your grid or rows of tiles in waiting, and those claimed colors — both by you and everyone else — start affecting everything else that you do. Does the player behind you need two yellow to complete a row? Take them for yourself! Make them take actions to scratch out only a single tile each turn, thereby possibly dumping multiples of colors in the middle that you can then scoop up.

Not a great first round, having completed only two rows — that dark blue tile will be removed

The same number of tiles are put into play each round (except possibly in the four-player game when the bag runs low because so many tiles have been used), but those tiles needn't be distributed evenly. You want to grab great globs of them for yourself so that you can complete those four- and five-space rows quickly and repeatedly. If you let those rows drag along half-filled from one round to the next, then you have little hope of completing columns or grabbing five tiles of the same color, and that's where you land the big points, so be greedy at the expense of others, and the only ways to be greedy are to:

1. Pay attention to who's taking what, and
2. Take tiles from the right places at the right times.

This latter aspect of the game only starts emerging after your first couple of rounds. You realize that if you had taken that one red tile from the center instead of the final disc then someone else would have taken a tile from that disc, possibly allowing you to grab three blue tiles instead of two since one blue was dumped into the center. Or you see three blue already in the center and realize that everyone else either already has blue in their fourth and fifth rows or is at work on some other color and can't take blue, so you let it sit a round to build up to four or more tiles so that you can complete a row in one go. Anyone who's drafted in Magic: The Gathering or other games will recognize this tension: How long can you wait to take something? Will someone else snatch it first? What's more, can you wait longer so that the pot builds?

Round #2 begins with you going second; what are you aiming to collect?

I picture the ending of each round in Azul as the moment in The Matrix Revolutions when Neo and Trinity briefly rise above the clouds to spy sunlight they would have never expected to see — only to then plummet downward into the thick of battle once again where fresh tiles have been laid out and everyone is fighting and you're not sure what's going to happen. (I apologize for making you think about The Matrix Revolutions.) The rising tension mimics that feeling of when you're reaching for cards in a new round of a trick-taking game: What's possible this time? How can you score what you need and dodge the rest of the time?

Azul is even tighter at two players because you know that whichever tiles you don't get, the other player will, and this makes the Nim comparison even more evident: If I take this tile, then you'll likely take those two, which means I can take these and possibly set myself up for those the turn after. I might want to take these four tiles, but if I can leave them and force you to take them when you have no room, then you'll lose six points, which is pretty much the same as me gaining six points.

You can still do such things with four players, though, and this can be even more satisfying. You see that the next player likely wants a black tile and the person after that a yellow, so if you take this blue (when otherwise you might not have cared which color you took), then player #4 gets stuck with lots of tiles they can't use. Yay, collaborative kneecapping! We used similar tactics in my most recent game to ensure that a player couldn't grab two black tiles to complete a row and therefore trigger the end of the game. (He was the only player who had four tiles in a row.) We hadn't counted out the endgame bonuses, but that player seemed to be in the lead, so we wanted more time and took turns pushing him away from the door.

Possible trouble ahead

I haven't even mentioned the gray variant game board — not advanced, mind you, but "variant", although I'd advise playing on the colored side for your first game or two. When you play on the gray board, you place tiles under the same restrictions, with each row and column of your 5x5 grid allowed to have each color only once, but outside of that you have the freedom to place a tile anywhere in a row.

What has happened in my games is that scores are higher since you can cluster the tiles as soon as you place them, but you also tend to box yourself into corners that will be increasingly unpleasant as the game continues. In the image above, I'm on my way to completing a column — assuming that I can finish the second and fourth rows before the round ends (or the game, really) — but if I place black in the fourth spot in that column, then I'll need to place black in the fifth spot of the adjacent column, but I can't even start working on that black row until I clear out the yellow, and that yellow's not going to score my much anyway, but I've started it, so now I need to end it. If I instead place the black in the fourth spot of the middle column, then I can't place both the light blue and yellow in the second column or else I doom myself to never completing the column since black would have to go in the fourth spot and can't.

Whenever you take tiles, as with those yellow in the bottom row, you can choose to dump them on the floor instead of placing them in a row, but that's like purposefully stepping on a nail. Better to get small points than negative points, right? Maybe?

83 points, despite flubbing the top row

My only regret with Azul right now is that I have many more games to preview ahead of SPIEL '17, so I won't be able to play it as much as I want. I realize that might sound like bragging rather than an actual regret, but it's not. To do a good job in this space, I can't post over and over again about the same game, but thankfully Azul will wait for me until I call once again. That waiting time for the tiling shouldn't be a problem given the king's low expectations...