Blockers—Design History

Most of the games I’ve designed have ended up looking nothing like the games I started out to design. Blockers is no exception, but its evolution happened unusually quickly. The final version of the game is not much different than the game we were playing by the end of the first playtesting session.

The First Playtest

The very first game was played on an 8×8 grid, with the letters A-H across the top and I-P down the sides. There were four colored sets of sixteen letter-tiles. Instead of “owning” a color as you do in the current version of Blockers, you had a random mix of tile-colors in your hand. And instead of trying to build contiguous groups of your own tiles on the board, you took pieces off of the board as the game progressed. At the beginning of the game, the board was seeded with a different set of tiles displaying numbers from -5 to +10. On your turn, you’d play a letter-tile from your hand near the edge of the board, and then you’d collect the lowest number-tile from the row or column matching the color of the letter-tile you played. Your goal was to collect as many points as possible in this fashion.

After a play-through, we concluded that the tile-collecting aspect of the game was uninteresting, but the row-and-column idea was promising. I suggested that the game would be more fun if you were building contiguous groups on the board rather than taking tiles off of the board. After mulling that over for a minute, Jacob Davenport suggested just giving each player one of the tile-colors, and on your turn you play a tile from your hand into any empty space in the corresponding row or column. That mechanism sounded great, and we decided to try it using the simplest win-condition we could think of: the winner is the player with the largest connected group on the board at the end of the game. By the time the game was over, I knew we were onto something, and I went home feeling that familiar game-design excitement.

Initial Problems

Of course, we were already aware of a few problems. For one thing, the game was too short, and the tile-set was too limiting. Each player only had sixteen tiles to play, and the groups that we created tended to all look like staircases or straight lines. Furthermore, the win-condition was somewhat broken. It was too easy to get into a situation, even in the early-game, in which you knew for certain that you couldn’t win.

After mulling over these problems for a day or two, I decided that I needed to add a third “layer” to each tile-set. By this time I’d already changed the prototype so that there were number-tiles for the columns of the board and letter-tiles for the rows. My idea was to add “icon” tiles to the mix, and put corresponding icons into the spaces of the board. Now each space on the board could hold one of three different tiles from your tile-set, which made the tile-placement much more flexible. Furthermore, the extra tiles would extend the length of the game.

Of course, I had to figure out how many icons to add and how to arrange them on the board, but after a minute’s thought, the answer was obvious: I should extend the size of the board to 9×9, and then divide it into nine 3×3 regions, each with its own icon. Now there were nine letter-tiles, nine number-tiles, and nine icon-tiles, and each tile could be played in one of nine spaces on the board. Perfect!

(Although hardly anyone believes it now, I’d never heard of Sudoku when I came up with this idea. In fact, I only learned of it after people started playing my early prototypes. This kind of convergent evolution happens all the time in game-design.)

For a new win-condition, I decided to try the idea that your score is the size of your smallest group times the size of your largest group. With that rule, even if you’re behind in the mid-game, you can always hope that everyone else will have to play singletons at the end and “crash”, allowing you to eek out a win. Furthermore, this system forces you plan ahead and think about every one of your plays, rather than just make throwaway plays when you can’t increase your largest group. In general, the idea promised to add a dose of tension to the game.

A little less than a week after the first playtest, we got a chance to try the updated rule-set, and, as expected, the changes had greatly improved the game. By now I was certain that this was a keeper.

PowWow Problems

That weekend, Dave Chalker and I were heading down to Virginia to attend a game-design event called PowWow that was being run by our friend Stephen Glenn, designer of Balloon Cup. We’d recently met Stephen at Protospiel, and we were looking forward to seeing him again. In addition, we’d caught wind that game-designer Alan Moon might be in attendance. I’m not the gushing fanboy type, but I was definitely excited about the chance to pick the brains of both of these great designers.

We arrived at the hotel after a sweaty three-hour car-ride, and while Dave headed to our room to take a shower, I wandered over to the gaming area to see what was going on. Alan Moon, Stephen Glenn, and a few other folks were hanging out at table chatting, so I joined them. I figured I was going to have to wait a while before I got a chance to pull out my game, but, amazingly, when we discussed what we should do next, no one else seemed to have any particular agenda. So I mentioned that I had something we could try, and before I knew it, we were playing.

Things went even better than I’d hoped. The game moved along at a terrific pace, and everyone was smiling and having fun. Because the game was so quick, we were able to play it a couple of times in a row. Stephen loved it, and Alan said that he enjoyed it in spite of his usual dislike of abstract games. Over the course of the weekend, they and many others were generous enough to spend extra time working on the game with me.

Despite its positive reception, it was still clear that the game had some problems. The biggest was the “hand-clogging” problem. As the game progressed, rows, columns, and regions of the board began to fill up, causing other tiles to become unplayable. My end-condition reflected this fact—we just continued the game until a player had no legal play. This was functional, but I didn’t like the way that your options diminished further and further as your hand became clogged with unplayable tiles. And the problem was exacerbated by the unforgiving nature of the scoring system. You could drop from a score of 50 to a score of 5 on the final turn of the game because your meager options forced you to play a singleton.

All of our playtesting that weekend focused on the hand-clogging problem. We tried many “obvious” solutions—allowing players to discard and replace unplayable tiles, and so on—but they all felt like band-aids to me, and they usually caused unwanted side-effects. When PowWow ended, I knew that I still had work to do.

The Capture Solution

After thinking about it for a while, I decided that the root cause of the hand-clogging problem was the static nature of the tile-play. Once a tile was placed, that space was locked up for the rest of the game, which caused other tiles to eventually become unplayable. Therefore, I began experimenting with different ways of allowing pieces to be removed from the board. For instance, I tried various Go-inspired mechanics which caused entire groups of tiles to “die” when completely surrounded, but those ideas made the game too volatile.

My favorite idea was to give you the ability to “capture” an opponent’s piece by simply replacing it with your own. However, it was clear that the game would fall apart if this ability were not restricted in some way. While discussing the problem with the core playtesters (Jacob Davenport, Dave Chalker, John Cooper, and Kristin Matherly), Jacob asked if it would be inelegant to disallow any capture that would split an opponent’s group into multiple groups. I’d been considering the same possibility, and I had the same worry about its possible inelegance. However, we decided that the idea sounded promising enough from a gameplay perspective to make it worth trying. And indeed, the mechanism not only completely solved the hand-clogging problem, but it also made the basic gameplay substantially more dynamic and interesting. It felt right to me.

End-Game Problems

The new capture rule forced us to reconsider the game’s end-condition and win-condition. We could no longer just end the game when a player had no legal plays, because now players always had legal plays. We could continue the game until all tiles had been played, but then almost everyone would end up playing singletons at the very end of the game. The rule we settled on was that you play until you draw your final tile, and then you get one more turn, leaving four unplayed tiles in your rack. This gives you a little breathing room at the end, while guaranteeing that you’ll have access to all your tiles.

(Incidentally, this new system allowed me to add something that I’d wanted to add since the beginning—a “wild” tile for each player. This idea didn’t mesh well with the earlier rule-sets, because the game would sometimes end before you drew your wild. In the new system, you’re guaranteed to draw your wild eventually, even if it does come later than you’d like.)

The new capture rule also didn’t mesh well with the old win-condition. Since in that version your score was directly related to the sizes of your groups, the ability to capture made it too easy to eat away at an opponent’s score. We realized that we were going to have to come up with some other win-condition that wasn’t based on group-sizes. After experimenting with various systems, we hit on the idea of focusing on the number of your groups rather than their sizes. This was a great fit with the capture rule, because it meant that players could not directly reduce your standing in the game by capturing your tiles.

Of course, with such a granular win-condition, ties are virtually guaranteed, so we needed to come up with a tie-breaker. Once again, we wanted to avoid using the concept of group-sizes. Therefore, we decided to break ties based on the number of tiles you’ve captured—more specifically, on how few tiles you’ve captured. This not only meshed elegantly with the “fewest groups” win-condition, but it also kept the game from being too aggressive.

We played many games with this rule-set, and it was clearly the best version so far. I kept an eye out for additional problems, but nothing cropped up, so eventually, about six months after the original playtest, we declared it done. I added a roaring 20s theme, named the game Uptown, and started looking for a publisher.


After PowWow was over, Alan Moon was kind enough to invite Dave and I to his annual Gathering of Friends. So in April of 2005 we made our way to Ohio, and I brought a bunch of prototypes with me. I played the game with various publishers, and I put my prototypes into the hands of the right people. Everyone seemed to like it, and I figured I’d have no trouble landing it somewhere. But, somehow, it didn’t happen.

Almost a year later, in an email conversation with Alan, he expressed his own surprise that the game hadn’t sold. He mentioned that Funagain Games was getting into the business of game-publishing, and that the game seemed like a good fit for them. He put us in contact with each other, and I sent them a prototype. Since the next Gathering was only about a month away at that point, I figured I’d just meet up with them there to see what they thought.

Well, I did meet up with them there, and found that they’d been pretty underwhelmed by the game. I started to wonder if I was going crazy. Was it possible that the game just wasn’t as good as I’d thought? But further probing uncovered the truth: a badly-worded sentence in my written rules had caused them to play the game incorrectly. We played the game with the correct rules, and, gratifyingly, they liked it. In the following weeks we negotiated a contract, and a year later, at the next Gathering, we got to see the first pre-production copies of the game. The game was finally released in the fall of 2007.

From Uptown to Blockers

Uptown’s print-run was relatively small, and relatively low-budget. Funagain began looking for ways to bring the game to a wider market. Eventually they partnered with Briarpatch, who began working on a new production. We ditched the roaring 20s theme in favor of a more abstract presentation, and renamed the game Blockers.

In the meantime, we’d been getting lots of feedback telling us that the didn’t play very well with three players. The board wasn’t crowded enough, so it was too easy to find safe plays. I decided to try to fix this problem for the new version of the game.

I thought about putting “blocks” onto nine spaces of the board, causing everything to be more constricted, but that seemed like an ugly solution. A better idea was to tweak the scoring rules to make capturing more painful, which would cause the board to fill up more quickly and (hopefully) make the three-player game more interesting. An obvious possibility was to give you a point for each group and a point for each capture, but that made capturing *too* painful. We started searching for a less extreme version of this idea, and after much discussion, a playtester at that year’s Gathering suggested the rule that you get a point for each tile of the color you captured the most. Not only did this fix the three-player problem, but it added the fascinating element of encouraging players to spread their captures among all of the other players, which improved the game in general. Blockers was eventually published in North America and Europe with this new rule, and it went on to make the 2011 Spiel des Jahres Recommended Games list, and win the 2012 German Lernspielpreis award.


As you can see, Alan Moon was instrumental in helping me get this game published. He essentially acted as my agent, free of charge. Alan cares about the industry, and he seems to really enjoy helping designers and publishers connect.

Jeff DeBoer and Rick Soued (and others) were a joy to work with at Funagain, and they did a great job with the product.

Stephen Glenn, Brian Leet, W. Eric Martin, Jared Scarborough, and lots of other folks at the first PowWow helped me immensely by playing the game over and over and letting me tweak it to my heart’s content. At the two following Gatherings, Brian must have played the game with me a dozen times as I demoed it to various publishers.

And, of course, I owe many thanks to Jacob Davenport, Dave Chalker, Kristin Matherly, and John Cooper—the best boardgame development team on the planet.