A few weeks ago, I got in my first play of Urbanization, a city-building game from Queen released last Essen.  This is a title that has largely flown under the radar, possibly because of some questionable physical design decisions by the publisher which can lead players to play by the incorrect rules.  I enjoyed it quite a bit and it made me realize that four of my favorite designs from last year are games of perfect information.  By perfect information (which I’ll abbreviate as PI from now on), I mean a game with no random factors or hidden elements.  All of the players know the exact game state at all times and can plan and react accordingly.

The other three 2012 titles I’m talking about are:

• Terra Mystica, the hit fantasy-themed building game, that currently sits at #8 on the Geek’s list of top-rated games;
• Agricola: All Creatures Big and Small, a 2-player spinoff of Uwe Rosenberg’s magnum opus, which is nevertheless its own, independent game; and
• Le Havre: The Inland Port, another 2-player spinoff of a hit Rosenberg game, which is even further removed from its parent game than the previous example.

The Agricola game won the IGA award for best 2-player game last year and the other two games have been nominated this year.  In addition, Terra Mystica made the SdJ recommended list.  So clearly this is a group of well regarded games.

Then I realized that another 2012 game I’ve played, Splotter’s The Great Zimbabwe, is also a PI game.  (I actually thought TGZ would be a personal favorite after my first game, but my opinion soured considerably after my second game.)

All of this got me wondering if having this many notable PI games in one calendar year is an unusual thing or a typical one.  It’s hard to create a PI design, since with so much information available, it can be difficult to make the game challenging to play; without random factors, replayability is also a concern.  So it seemed like five games was a lot, but I wasn’t sure.

Fortunately, in this day and age there are resources available to investigate such questions; in my case, this includes my fellow OGers.  I did some research and consulted with the other folks here (Tom Rosen was particularly helpful).  Together we came up with a list of 33 notable games that are PI.  This is not intended to be a complete list by any means, but the goal was to summarize the well known games of the past few decades that are perfect information.

Before I show you the list, let me mention a few restrictions I imposed on the process.  I’m not including 2-player abstract designs, such as Go and Chess, since perfect information is kind of the norm for those titles (Backgammon is an example of a classic 2-player abstract that isn’t PI).  Similarly, I’m not considering speed or dexterity games, since I’m trying to focus on more complex games that still manage to be PI.  Finally, I’m choosing not to include games with simultaneous selections that are otherwise PI.  Examples of these include Maharaja, In the Shadow of the Emperor, and Diplomacy.  I can appreciate an argument that these really are PI games, but since there is some hidden information at the time you have to make your choices, I think it’s best to exclude them (as they just don’t feel PI to me).

With those ground rules out of the way, here are the games we came up with.  I’ve listed the year of publication after each title and they’re shown in reverse chronological order.  I’m also using asterisks to further subdivide the selections.  Games with no asterisk have a constant setup, so that the initial conditions are always the same for each play.  Games with an asterisk have a variable setup, accomplished either randomly or through player choice.  These have the potential to give the players a different set of conditions at the beginning of each game.  However, once the setup is determined, there are no further random elements in the gameplay.

• Agricola: All Creatures Big and Small (2012)
• Le Havre: The Inland Port (2012)
• Terra Mystica* (2012)
• The Great Zimbabwe* (2012)
• Urbanization (2012)
• Ora et Labora (2011)
• Luna* (2010)
• Sun, Sea & Sand* (2010)
• Steam* (2009)
• Steel Driver (2008)
• Chicago Express (2007)
• Hamburgum (2007)
• King of Siam* (2007)
• Imperial* (2006)
• Antike (2005)
• Caylus* (2005)
• Neuland* (2004)
• Hey!  That’s My Fish!* (2003)
• Clippers (2002)
• Mexica* (2002)
• Pueblo (2002)
• Ka-Ching!* (2001)
• Medina (2001)
• Pampas Railroads (2001)
• Blokus (2000)
• Morisi* (2000)
• Stephenson’s Rocket (1999)
• Torres [Master Version]* (1999)
• Through the Desert* (1998)
• Billabong (1995)
• Tutankhamen* (1993)
• Forum Romanum (1988)
• 1830 (1986)

What kinds of conclusions can we draw from this list?  Here are a few:

• First of all, it does indeed seem that five PI games in one year is an awful lot.  No other year even has four games in our little list and managing as many as three PI games in a year is worth noting.  I’m sure other people could come up with comparable lists that show other years equaling or exceeding 2012, so I’m not claiming that last year was necessarily extraordinary.  I just think it’s safe to say that we saw an unusually high concentration of PI games during the last calendar year.
• The split between Constant Setup games and Variable Setup games is just about 50/50.
• There are definitely some designers, and even some publishers, who have shown a tendency towards creating PI games.  These include:
  • Uwe Rosenberg, who’s published three in the last two years alone.  (He also assisted with the development of Terra Mystica.)  Additionally, both Agricola and Le Havre could be converted into PI games with some simple variants (but the same could be said of a lot of other games).
  • Corné van Moorsel, who has released a bunch of PI games through his company Cwali (two of them, Sun, Sea & Sand and Morisi, make the list).  Some of these are 2-player abstracts, but many are multi-player.
  • Martin Wallace is represented with three games.
  • Three of Mac Gerdts’ rondel games are on the list and a fourth, Navegador, is very close to being PI.
  • Even though Kramer and Knizia have 4 and 3 games, respectively, on the list, both create so many games that I’m not sure I’m willing to say that either one of them show a tendency towards PI games.  Still, it’s obvious that both of these designing giants are perfectly comfortable with this kind of game.
  • A lot of the recent Winsome cube train games qualify as PI.  There are two Winsome games in the list; the others have a small, but enthusiastic, following.
• I only included one 18xx game, 1830, since it’s easily the most popular of the genre, but just about all of them are PI games, going back to the title that started it all, Tresham’s brilliant 1829, first released way back in 1974.
• Most of the earlier games in the list are of the simpler, more abstract type, while most of the later games are more complicated and multi-faceted.  The dividing line is just about the turn of the century.  I suspect that abstract designs (and more basic designs in general) were more acceptable to the gaming public during the 80’s and 90’s, whereas the last 10 years have shown a tendency toward more complex titles, particularly from the hobby’s most renowned designers.

One of the reasons that PI games are so rare is that it doesn’t take much to make the information available to the players imperfect.  For example, in Tzolk’in, the only non-PI element is that the buildings come out in random order; in Puerto Rico, it’s the way the plantations are revealed.  In neither case are these major parts of the games, but the elements do affect play and the tiny bit of randomness improves the replayability of the two titles considerably.  So even games which feel mostly PI benefit from the small amount of randomness included in them.  (Then again, you can look at Roads & Boats, where the only random factor is the seeding of the mines, which seems to have more of a thematic justification than anything else.  The game could have easily been designed to be PI, but Splotter chose not to do so.)

An even more revealing case is the comparison of two of Mac Gerdts’ rondel games, Imperial and Navegador.  As mentioned above, the former game is PI, but the latter is not, as the colonies which are revealed when each region is discovered are determined randomly.  This does direct play to some extent, as players will naturally prefer founding cheaper colonies.  But the funny thing is that, to me, Navegador feels much more predictable and deterministic than Imperial does.  Imperial is completely PI, but there is a large amount of player-induced chaos built into the design.  In spite of the randomness level of each game, Navegador feels considerably more like a PI game than Imperial does.  So obviously, there are more things to consider (such as player interaction and how much direct damage can be done to opponents) than just whether a game is PI or not.

Even though our little study seems to indicate that 2012 was a great year for perfect information games, I have no reason to think we’re looking at a trend.  You see accidental groupings of games all the time, so it could have easily been something of a fluke.  What does seem more likely, though, is that we will continue to see a steady stream of PI designs and that many of them will have the kind of complexity and multi-faceted nature that you rarely would have encountered a decade ago.  If nothing else, the tremendous popularity of games like Terra Mystica and Ora et Labora should inspire publishers to continue producing games like this.

So we OGers have done our homework, but we realize our list is far from complete.  We’d love to hear from you about any other PI games you think we’ve missed, or even if you disagree with our categorization of the games we list as PI.  Please feel free to leave your suggestions as a comment to the article.  We’d like nothing more than to have you add to our site’s store of knowledge.  More information?  Perfect!