I’d like to thank Sam Chapin, for his excellent sportsmanship in playing so many games of Catchup.
To skip the intro, and get to thoughts on the game click here.
I’m going to start this essay off by saying that I’m a Nick Bentley fan. I’m a little nuts for his work. You might as well imagine me wearing a Nick Bentley for President shirt while shouting this review on a street corner. That’s the enthusiasm (bias) I’m coming to this with.
Nick seems like the kind of guy to put his keyboard where his mouth is. He wanted an interesting book about games, so he went and made BoardGameStrategies.com. He wanted to see deeper more involved assessments of boardgames, so he went out and marshaled the troops on BGG by coordinating a fascinating list: “The 100 Play Challenge.”
I want to try to follow this kind of lead. So — I’ve done something like the 100 Play Challenge with Nick Bentley’s game Catchup. I’ve also posted photos of the completed hundred games below, as well as a printable PDF of the board that I played the last ~90 games on. I didn’t follow the exact prescription for how to complete the challenge, but I’ve had a lot of fun doing it.
Firstly, this is a game that grows on you. It slowly reveals its depth to you, game by game, and after a hundred games you will still feel like there is something to be learned from each game. We are a long way from understanding this game.
Luckily, when explaining the game, people get it. Catchup balances intuitive appeal and non-intuitive strategy. People know what they need to do. Moments later, they realize they can’t do what they want to do too fast or else their opponent will drive them into the ground. A little while later, they discover basic group management. Dozens of games down the line, things will keep on clicking.
My experience was that Catchup has a long gently sloped learning curve. You play a round, you learn a thing. You play two rounds, you learn a new thing. You play four rounds, you learn a bit more stuff. Games with a relatively steep learning curve would be Mahjong / Riichi experienced by non-Kanji readers. You’ll play a number of games before figuring out all the suits, the turn order, and a tenth of the scoring hands. Perhaps this long lightly sloped curve in Catchup is an artifact of the way I played through the 100 Play Challenge. There were several large gaps, and I didn’t take time to deeply study the game. Maybe, in a couple hundred years, there will be elaborate Catchup opening theory, end game theory, group adjacency theory, and all kinds of impediments to starting to play the game competently. Luckily, we didn’t get hung up on that or its absence.
When I first learned the rules to Copolymer, I was thrilled. Here was a pure abstract game with a crystalline balancing mechanism that would provide all kinds of interesting growth competition. Unfortunately, the opening proved to be hopelessly opaque and always felt very drawn out. Catchup fixes those problems. The opening is pretty clear, and the middle game is there before you know it.
On the issue of clarity. During most moves, one places two stones on the board. Initially I was worried that this would lead to the game being hopelessly unclear. There are many places to put stones, and placing two stones must certainly be more complicated than placing one. In practice, I think most moves get played “locally,” by which I mean that both stones get placed near each other on in the same part of the board. This significantly cuts down on the amount of things one needs to consider and makes the game clearer. In the fertile twilight of the opening, a global move can come as a shock, but generally the double placement mechanism didn’t render the game hopelessly unclear. Usually one used both stones to achieve the same purpose (connect up two groups, block off a corner, etc.) and once one got a feel for the possible uses of two stones one could readily keep track of possible moves.
Thematically Catchup is about warring bacteria culture trying to outgrow each other in a petrie dish. They surround each other, sequester off parts of the dish, and cut each other up. It’s a blood bath.
Throughout the game, multiple groups will vie to be the largest. Bottle necks will appear. There is always a lovely interplay of local and global strategy. The main drama of the game comes from the lead passing back and forth between the players. Especially in the opening, there will be several moves where both players feel that someone must grow, and the next move will decide the game, but neither player wants to give the extra stone to their opponent. This build up to the enlargement of the largest group provides a regular build up of tension that keeps the game interesting from opening to end game.
Catchup’s end game condition is a subtle beast. Ties cancel each other off, and one is always worried about the second groups. In my particular experience, almost all games were decided by first or second groups. There is one freak game of the hundred that was decided by fourth groups, the ever diligent and curious reader can find it for himself. Yet, in most games where first groups decided the game, second groups still played a strong role in steering the course of the game. One doesn’t want let their opponent’s second group get too large for two reasons: firstly, if your opponent gets two large groups its likely they’ll connect them up and gain a decisive lead, secondly, if you successfully defend against the joining up of those groups, you run the risk of tying first groups. So — Even a minute second group is a concern. This two tier strategy will probably only become more crucial with more play. My gut tells me that if the community moves to larger boards, we’ll need to consider three or four tiers.
Christian Freeling has pointed out that a game is doing well if players start to develop jargon for playing it. Nomenclature started to develop when we were playing. A couple of times we tried to import terms from Go, and then after a bit more play found that it simply did not fit. If one has a small group surrounding a corner, that corner is not territory, it has only been sequestered off from your opponent. At best the group you’ve used to cut it off might join up with your larger group, at worst, it’ll restrict the size of your opponent’s largest group.
About the board size, the standard side length five hex grid provides a very tight space for a war. Things in the middle bump in to the corners, etc. If people move to playing on larger boards, I think that we’ll see very rich strategy about corner management emerge. That said, there’s no need to move up to larger boards since side length five is a very fertile game. I haven’t tried out larger boards, and don’t think I will feel a need to any time soon.
One feature of Catchup that appeals to me is that the board gets more and more full until the final move. Moreover, nothing is moved, nothing is altered. ZÈRTZ is exactly opposite this. At the end of a game, one just has a mostly empty table and a crushed soul. At the end of a game of Catchup, you have this elegant picture of warring colonies surrounding each other. Even if you lost, you have something aesthetically pleasing to point at. See the photos below.
Nick has written about the ecstasy and agony of scoring tracks. I’d like to thrown in my two cents about the scoring track in Catchup. Firstly, use it. It’s there for a reason (Catchup with Scoring Track). We tried a couple of games without it and it was a terrible mess. We kept recounting things over and over. Secondly, it’s not as distracting as you might think. Things usually change in pretty controlled ways. They also change rarely enough that it is not a constant nuisance. When I made my own board for play, I made a “split scoring track” that you can see in some of the photos below or in the attached PDF. The idea being that one uses one track to track the leading digit of your largest group and another to track the second (‘units’) digit. In practice, this was perfectly clear. More often than not both players have roughly equal groups and so knowing the units digit was perfectly fine. Note that the scoring track only goes up to thirty since if you have a group of size thirty-one (on a side length five board) then you’ve won by default.
TL;DR: Catchup is an exceedingly well designed game that everyone who loves abstracts should try out at some point.
Edit: Fun fact, Nick Bentley sometimes plays fourty games of Catchup a day.
Edit: I found some brief notes I kept after the first few games. I include them here for the sake of completness.
Nick Bentley’s 100 Play Challenge
Nick put out an interesting GeekList a while ago asking board game geeks to play a game one hundred times and meticulously review it. This project might help to combat some of the surprisingly shallow game analysis that occurs on BGG. I can’t count how many times I’ve seen someone who is reviewing a game after a few plays hail it either as: a work of genius, a piece of junk, or a childish triviality. Of course this really takes a chip off my shoulder when it comes to games I really like. If you think Hex is childishly trivial, you’re doing something wrong. Moreover, in the abstract game review scene, which is the tiniest iota of the already infinitesimally small world of board games, there is an baffling tendency to compare everything and anything to Go. This trivialises the whole project of meaningfully comparing games.
I am pretty excited by this whole enterprise of deeply looking into strategy for new modern games. Nick also started a blog/wiki about this endeavour.
Meg and I want to take up the project by playing a lot of Nick’s best game: Catchup. If you have any appreciation of finely crafted games, and any inclination towards abstracts, go quickly and read the rules — your life will be immeasurably better. The famed Argentinian writer Jorge Borges said that to not read Dante’s Divine Comedy is “to ascribe to a secret and obscure asceticism”; this comment also applies to Catchup.
Edit: Borges actually said: “The Commedia is a book that everyone ought to read. Not to do so is to deprive oneself of the greatest gift that literature can give us; it is to submit to a strange asceticism” in Seven Nights, as translated Eliot Weinberger. I prefer my vague remembering, and will continue to imagine it as the original. Back to the review.
One pleasant thing about playing Catchup is that it is a monotonic game. Once the game is done one is left with a pleasantly complicated pattern of stones that has a nice mix of cluttered busyness and bloby order. . We are no where near the point where we can read out a game’s history from its end-state but one can sense a games broad outline. “Those stones were initially filling up enemy territory but were later joined a big group…”
We are playing on a side length five board drawn on a Chessex mat. Initially we started playing without the circular scoring track. This was more mentally taxing and caused us to ‘time out’ several times a round to recount everything. After round two we adopted an unmarked scoring track. This led to awkwardly counting around the boundary when we forgot the exact size of the opponent’s group. Now we are using a numbered circular track. Don’t try to be clever; use that thing from the start. It is handier than you would first expect.
The way we are playing now it is not unreasonable to play two or three rounds in a sitting. Presently a round takes us fifteen minutes.
We are taking a “pics or it didn’t happen” approach to logging our games.
Rounds 1 & 2
In our first rounds I mostly did things that would make sense in Hex. I played a number of bridge moves, creeping all over half the board, while Meg made a large group. In the middle and end game I had enough structure that almost everything enlarged my largest group. This cost me a fair bit of momentum. Meg had two equally large groups that joined in the end and that was pretty shocking.
Rounds 3 & 4
We played two really satisfying rounds of Catchup in the morning over coffee. Over night Meg realised that this early webbing business was making my middle game too strong. She retaliated by thwarting any attempts to construct it. When I played bridge moves she would fill in one half of the virtual connection, knowing that I would not retaliate since that would make the nascent groups larger and give her more moves.
This webbing and counter-webbing business set the stage for Meg slicing the board down the middle around turn five <>.
Rounds 6, 7, & 8
At the close if the day we played the more rounds of Catchup. After two rounds Meg was still quite enthusiastic and wanted to play again.
At the end of the second round I despaired my lack of reserved territory enclosed by my groups. Meg had considerably more territory roped off. She had the define upper hand in the middle game. One needs strategic concept became clear. I discovered this notion of surrounding suave issue enemy territory. I managed to destroy eight points of territory in two turns by making a ‘turtle’ (a shell) along the boundary.
Afterwards, I thought a bit how about one might use the corners of the board to quickly produced shells. Nothing
Something rally clicked for Meg in the third round and she best me terribly through the opening and middle game.
It occurred to us that it is better (for readability) to take a photo just after someone resigns than after the board has been filled up. We are starting to feel comfortable with figuring out when someone has lost, even if that is for or five moves away. That says a lot about how coast this game is.
Rounds 9, 10, & 11
Woo-hoo! I’ve made it one tenth of the way to my hundred play goal. Things are trucking among pleasantly. I keep finding interesting little nuggets of play. The opening is still very murky and we are stuck in a loop of playing very tight symmetric openings or very sparse webbing. Understanding opening theory seems a little ambitious at the moment. We are presently exploring what kinds of connections are virtual, and solid, in what contexts. There are several things to keep an eye on while evaluating a connection. Of course, if two group of stones share four neighbours then they are strongly connected, but those conditions are so idyllic that they rarely happen. Instead one is left evaluating groups with two shared neighbours and three shared neighbours.
The realisation that one can create enclosed territory has changed the game quite a bit.
We played a couple really good games this evening. Meg won by keeping two large groups active and growing the whole game. She always had three spaces to possibly connect them, in case she needed a big push. Of course she managed to perfectly arrange that I would never get the chance I needed to separate the two groups.
The following pseudo-puzzle occurred to me this morning:
Are there situations where moving hurts one’s position? Put another way, is there a position such that Black moving causes White to win?
During these rounds I tried deploying men just above the corners early on in the game hoping that later on I could secure some territory. This was not as useful as I thought it might be; I only got to fill in the hexes neighbouring the corner once which didn’t win me much.
One charming thing about the rounds this evening was that Meg won a round by a point and so did I. We perfectly matched at this point. Close games have a very appealing mix of joy and frustration; knowing that you are only one step behind someone is much better than getting crushed by their virtuosity.
Played four rounds, two against Sam and two against Megan.
Sam won twice. Meg won the first round and I won the second round.
Meg points out that I should always play one colour so that I can know which rounds I won/lost at a glance. From now on I will play black.