Tab Completion

I'm Tab Atkins Jr, and I wear many hats. I work for Google on the Chrome browser as a Web Standards Hacker. I'm also a member of the CSS Working Group, and am either a member or contributor to several other working groups in the W3C. You can contact me here.
Listing of All Posts

Boardgame Review: Strain

Last updated:

I play a lot of board games, so when my good friend Jeremy Apthorp (@nornagon) bought me Strain (BGG), I was pretty excited to play it.

Unfortunately, the game has some pretty serious flaws, so we'll have to spend some time patching it up ourselves before we seriously put it into the game rotation.

The Good Parts

One nice thing about Strain is that the good parts are very good.

Everything about the game looks awesome before you play it. The box is pretty and very solid, the concept is interesting, and the mechanics seem simple and intuitive.

Most impressively, the art direction is amazing. This is one of the best-looking board games I've ever seen. The illustrations are amusing and adorable, and the backgrounds are simply beautiful.

The Mechanics

Unfortunately, once you start playing the game, the mechanics are a letdown. The overall impression I got from it can be summed up as "newbish". The mistakes and mis-steps in the mechanics are not subtle or interesting, they're basic problems that are well-recognized in board game theory and worked around in all good games.

To really understand the problems, you need a basic grasp of the mechanics. They're very simple, and the full rules are up on the game's website (linked above), so I feel it's kosher to give a quick run-down.

You have three types of cards - organism cores, cytoplasm (free organelles), and "petri dish" (everything else - advanced organelles, viruses, and actions). Each turn, you draw three cards of whatever type you want, and at the end of the turn discard until you have four in your hand.

During the main phase, you can play up to one organism core, up to one free cytoplasm, and then as many petri dish cards as you can pay for. Some of your organelles produce ATP (currency) and/or toxin (attack). The goal is to evolve organelles onto your organism, then score it (discarding it) once you've reached a set number of organelles indicated on the organism core, while preventing your opponents from doing the same by attacking their organisms with toxin and viruses.

If you use an organelle for ATP or toxin, you flip it over for the turn. While flipped over, the organelle doesn't do anything, and has reduced defense.

Attacking is simple - flip as many organelles for their toxin as you want, then choose an opponent's organism. The opponent distributes the toxin among their organelles however they want. An organelle dies if it receives toxin equal to its defense, but survives unscathed otherwise (there are no lingering effects - the toxin damage disappears at the end of the battle).

Some Nitpicks

There were a few small things I disliked about the game, but which aren't bad enough to really be called "errors". None of these are enough to actually affect my mental review score, but it would be nice to have them fixed.

The game uses cards, but its basic flipping mechanism begs for thicker tiles. It's very frustrating trying to flip over a smooth, slick card on a table which is, itself, smooth and slick, particularly when they're arranged in a particular way so you can't just slide them toward the edge of the table easily. Plus, it's difficult to shuffle small square cards - you can't rifle-shuffle them like normal cards, but neither can you do the "throw them on the table and swirl them around" method like you can with cardboard tiles (it tears up the edges).

The way the defense mechanic is done enrages me. A basic rule-of-thumb in game design is to minimize the amount of math players need to do. You should try to limit math to just size comparison (is A greater than B?) and simple sums with small numbers. The defense mechanic, though, means that to figure out how much toxin an organism can absorb without damage (its "soak") you have to sum the defense minus 1 of every card. Every attack consists of distributing the organism's soak worth of toxin onto its organelles, and then killing off organelles equal to the remaining toxin (theoretically you can do something different, but anything else is strictly and obviously worse). As written, we had to constantly ask the other players what their organisms' soaks were, since it's not trivial to figure out at a glance. It would be much more player-friendly to reduce all the defense scores by 1, then rephrase the mechanic so that an organism explicitly "soaks up" toxin equal to its total defense, then loses organelles equal to the leftover toxin. This is exactly equivalent mathematically, but much friendlier and easier to use for everyone involved.

The Bad Parts

Now we get to the part that I don't like writing, but need to so that others can learn from these mistakes.

In short, it appears that Strain's developers weren't familiar with basic board game theory. It commits a lot of very basic errors that should have been obvious in the very first playtest.

There's a small group of core failings, each of which are only lightly damning on their own, but which sink the game in concert.

First, the costs are completely out of whack. Building up ATP generation is slow at the beginning of the game, where you're only building your total by 0-2 per round. Most organelles and viruses, though, cost 4 or more ATP. This means that, on average, it'll take four full rounds of drawing and playing before you get to play your first interesting card. In the meantime, you've possibly already built up a cell that looks like it should be scorable (if it was an organism core that requires 2 or 4 organelles to score), but you must avoid that urge! It's a terrible game-losing mistake to ever score your first organism, unless you've built up a second, better cell already.

Second, the attacking mechanic is extremely all-or-nothing. A well-developed cell can easily have a 8 or 9 soak, which means it requires a significant toxin expenditure to crack. Once you get past that soak, though, you only need a few additional points of toxin to cripple or kill the organism - knocking out three organelles (usually half the cell) only requires 3 toxin above the soak. This is bad for two reasons - if you produce a decent amount of toxin, but not quite enough to pierce someone's soak, you might as well not produce any toxin at all. However, if you do produce enough to get past their soak, you probably produce enough to completely cripple them. This makes the mechanic very swingy, which is not a good trait for a core mechanic to have.

Third, the boringness of the early game often extends out several turns further than it should need to, because tapped organelles drop to 1 defense (0 soak). You can't buy a 4 atp organelle as soon as you produce 4 atp, unless you're ready to have your organism completely wrecked. The opponent won't even have to commit much toxin to the attack, so they won't be opening themselves up to attack the same way you did.

Fourth, there are basically zero rubber-band mechanics in the game (mechanics that have little benefit for leaders, but are really good for underdogs), so once you get wrecked by the above, it's impossible to recover except by everyone feeling pity for you. If you lose your good organism, an opponent can trivially toss a few toxin at you every turn to keep you crippled for the rest of the game.

These four problems combine to just totally wreck the game.
It's boring at first, good play means it stays boring for a while after that, and one mistake can make you a lame duck for the rest of the game.

Unfortunately, that's not all that's wrong. Here's a few more major problems, in no particular order:

There's no late game. Organisms are worth 2, 4, 6, or 8, and you need 12 points to win the game. Because good play means that you never score your first organism, everyone ends up with a mature organism that's scorable, but is required to be left on the board. This pushes it out of your mind, so someone can sneak a win in by having 4 points (a third of the winning total, not nearly enough to be considered "nearly winning"), and a mature 6-point main organism, then playing a 2-point organism, slapping two organelles on it (cost is no object, since you're about to win), and then scoring both of them. A good game should have some long-term mechanic that builds slowly and obviously throughout the game and which prevents the game from being won too early, so that all the players understand when the game is about to be won and don't get caught by surprise by unsatisfying sneak-wins.

The core mechanic of viruses is that they prevent an infected organism from being scored - the player has to spend toxin to kill the virus first. However, this is basically meaningless. If you're about to score your organism, there's no downside to flipping all your organelles for toxin, so good play involves just ignoring the viruses until you're ready to score them. The only worthwhile viruses are those with ongoing effects (as you want to kill them early, before you're ready to score, when building up toxin is risky), and those with effects that trigger when they're removed, as they can actually hamper your ability to score the organism if you kill them right before scoring. Unfortunately, only the expensive viruses (which are too expensive to play unless you're already nearly winning) have these effects, so all the cheap playable ones are worthless.

The game is too simple and predictable. The only randomness is in the order of the draw piles, there's almost no hidden information (everything on the board is public and simple, and your hand doesn't contribute much), and there's very little interaction between cards, so it's extremely easy to grok the entire board state. A good game invokes at least one of these three things to increase the difficulty of predicting and planning - when well done, it's easy to make "good" moves, but practically impossible to make "perfect" ones because of the uncertainty and intractability of understanding all the interactions in the board state.

Related to that, there are no global or multi-player effects, or any cards that affect other cards. Nearly every card has a single, simple effect on a single piece of the core mechanics, and that's it. Again, good games lean heavily on this kind of thing - meta-cards that affects other cards don't harm your ability to do short-term planning, but make long-term planning much harder due to the second-order effects. Multi-player effects similarly are easy to understand the local effects of, but hard to predict the overall effect because of the other players' hidden information. Both of these make for much more interesting games than just "I make the number bigger and then do the same core mechanic I've used every turn.".

Conclusion

Despite the obvious love and skill that went into some parts of Strain, the horrible mistakes made in the basic mechanics mean that I have to recommend that people not buy this game, at least until the developers produce a second edition with a significant rules overhaul.

I'm going to see how much I can tweak the rules without changing the actual card text, so the game is at least playable. I expect to spend some time on this next time Jeremy's in town, as he loves game design. ^^;

I might also see about a more extensive revamp, maintaining the general rules shape but not worrying about rewriting cards. I already have a complete re-flavor sitting in my drafts, which changes the game from evolving organisms to building a nation, so that's a good excuse to try my hand at designing a real board game without the difficulty of coming up with the base mechanics. That'll be for future posts, though.

(a limited set of Markdown is supported)