Sitting around at Bordercon on Saturday afternoon, Iain asked me whether I wanted to try a new game he’d picked up. “It’ll only take about an hour,” he said, basing his estimate on the playing time of 45 minutes.
[As it turns out, this was the first of several mistakes we would make with this game.]
The game was Circus Train. It’s about running a circus out of Canada during the Prohibition era, and is based on Sara Gruen’s novel Water for Elephants (or think of the TV series Carnivale). Potentially an interesting theme.
The version we played was produced by Victory Point Games. They’re a small, volunteer-based publishing company. As I understand it, the games they publish are often created by students of game design.
That makes a lot of sense in the context of the game. It had familiar mechanics drawn from a variety of games:
- “Best Work” scoring, similar to Princes of Florence or Colosseum
- Card-based actions (hand management), where you choose which card to play from your hand each turn, from a limited pool
- Rounds and Uber-rounds (phases/stages/eras). And Uber Uber Rounds.
- A modified pick-up-and-deliver component, that sort of lost half of the deliver part in that you kept the items you had collected
- Multipliers for having multiple copies of a single item (more Princes of Florence here)
- Random event cards that cannot be mitigated against (think Agricola X-Deck but even worse) (I can say that, I was one of the X-Deck designers)
- End-of-round (end of week / end of month / end of 2 months / end of game) administration
- Limited number of turns
So far, that probably doesn’t sound so bad. And in fact I see that this game has some positive reviews on BGG. I am, however, at a loss to see why.
The game is played over 27 “weeks”. Game time, not real time, although we wondered. That’s six months of either four or five weeks, divided into two-month groupings. Here’s an overview of how it works:
- Set up the board: draw 8 cities and randomly place tokens on those cities.
- Week starts
- Each player, in turn, plays a card and does what it says. This may let you perform a circus or pick up some performers.
- End of week administration: Count how many cities have tokens and bring the number back to 8. Move the week counter.
- Repeat Steps 2 to 4 x3 (or x4 in a 5-week month)
- End of month administration and scoring round.
- Do another month (Steps 2 to 6)
- End of 2-months admin.
- Player with lowest score can steal.
- Repeat steps 2 to 9. but now with 10 active cities
- Repeat steps 2 to 9 but now with 12 active cities.
- End of game scoring
I checked the time just before the end of the first two-month period. We had already been playing for an hour. Only massive self-restraint was saving me from gouging out an eyeball to halt the play. Probably one of my own, although if Fraser had stolen another Human Cannonball it might have been his. We guessed that the “45 minute” playing time was maybe per-player: Given that each player has 27 turns AND six lots of end of month AND three of end of two-month AND end-of-game AND there is administration at the end of each of those 27 turns, there was no way we were going to be done in 45 minutes.
[Note: Further Research shows that 45 minutes is the base game playing time, for 1 or 2 players. The expected time for us should have been 90 minutes. We still took double that.]
So why did I hate this game so much that Iain actually offered to let me burn it after we’d played?
It wasn’t the quality of the game components – all cardboard chits & clip art. I’ve played worse, and will again.
It wasn’t the artwork, although the clip art was all cheerful, modern-looking clowns. If the game was about alcoholic prohibition-era clowns, the chosen artwork was not a good match. Google Images was not my friend when I searched for “alcoholic clown” (If you haven’t seen Shakes the Clown, do yourself a favour … and don’t see Shakes the Clown) but I am sure some more creative searching would do the trick.
It wasn’t the rules booklet, because I didn’t have to read it. There’s a summary sheet that’s actually quite good, but the rules themselves seem to be written in the old-style wargame format, even (I believe) down to the three-level numbering system.
But here are some things it might have been:
- WAY too long (this is both about the inherent length of the game and about our mistaken expectations of a “45 minute” playing time – but it still took too long for a 90 minute game)
- WAY too fiddly (see the various comments about administration – made worse by the fact that the number of active cities changes in each 2-month block)
- WAY too fiddly (three times during the game, you have to pay wages. The wages for each type of show/performer vary: they may cost 1, 3, 6 or 12. Or be modified by another chit.)
- WAY too fiddly (every time you start to score a performance, you have to add up all the different types of act, which vary depending on the performance so you can’t really work it out in advance, PLUS the varying points for the performances you have already done PLUS the varying points for the current performance. Our highest scorer had over 100 points).
- WAY too fiddly. There’s this whole subplot about your reputation that seems important but turns out to just be a way to get stuff. Get rid of it and spend VP instead.
- Ridiculously random. Gregor actually found a strategy (involving having lots of clowns). Then the random month card came out: Your clowns can’t perform because they are too drunk. Maybe this is an accurate historical simulation (really?) but it is also completely random. [Note: As it turns out, the random month cards are optional. I would choose to destroy them so they could never be used.]
- Poorly laid-out. The same chit (or type of chit) is abbreviated differently on different tiles/cards. So there’s HS for Horse Show and BC for Big Cats … but there’s also another place where at least one of these is spelt out (memory says it was “Horses” but I am not sure of the detail).
- Disjointed. This is a feel rather than a concrete observation. The game didn’t flow for us at all. There was a disconnect between moving and scoring and paying wages. Gregor thought perhaps it was the jump from movement to maths.
- Not fun. For us, anyway. And this is the hardest one to forgive.
I suspect that there’s probably a decent game buried in there. It’s certainly an interesting and clever theme.
For starters, I’d dump the random Event cards – they might be OK for a simulation, but they don’t make the game fun. Or foreshadow them – deal 3 face-up so players have a while to prepare for them, then randomly select one that actually occurs. The game isn’t light and isn’t short, so there needs to be a mitigation strategy.
Then I’d just get rid of stuff. Or streamline it.
Make all months 4 weeks long and redraw the card deck after eight weeks instead of nine (so 1 card remains unplayed). It’s OK to say that the wages card must be played every 2 months (ie that it cannot be the unplayed card).Or shorten the game by getting rid of a 2-month period.
There’s an option to rest in Canada for a turn to re-gain reputation, or pick up another act. Get rid of it. I think it was only used once during our game. Let people rest up somewhere else. There was too little opportunity to do anything interesting as it was. (Note: I recognise that Canada is Thematic. But if Thematic=Boring, bring in a wagon train from Canada or send a clown up there; don’t clog the game up with it).
Fix the information design issues. There was also an issue that when chits were on the board it was easy to miss where the train connections really went. I tried to take a flight somewhere and we noticed. I’m pretty sure that another player took a flight during the last round but I didn’t bother pointing it out because he’d already had his turn and it would be too fiddly to undo it. And I wanted to play something good before dinner.
Dump the stealing thing. Maybe it’s thematic, but when the game annoys its players, it’s always going to be bad.
But then I see that lots of people really seem to like the game. Which makes me wonder what I am missing.