Tuesday, June 25, 2013

GM Bidding System Obligations

I've been thinking a fair bit recently about how bidding systems should be dealt with at the World Boardgaming Championships. I managed to rope myself into being a GM for a game so this is more than just theoretical musing at this point, but I mostly care about the theoretically 'right' way to do things.

The primary problem that is trying to get solved by a bidding system (where before a game starts the players bid some sort of currency in order to pick which starting position they want to have) is that some games have very unbalanced starting positions. As the rules are written the USSR will beat the USA most of the time in Twilight Struggle. The players that start in the corn seats in Puerto Rico will tend to finish higher than those that start with indigo. Princes of Florence has bidding tempo advantages for some of the seats. A Few Acres of Snow is an auto-win for the British under optimal play. In other cases the bidding systems exist to allow people to play their preferred side, like in Star Wars: The Queen's Gambit. I've also seen systems which allow the game to be played at all like 1776 where a full campaign game would take days. So they restrict the game to a certain subset of the war and you bid for how many cities you need to control at the end of that time period in order to win.

How should the GM handle advising the players about these bidding systems? A GM that tells new Puerto Rico players to bid highly for the indigo seats is doing them a great disservice. In 1776 the 'consensus' right bid was something like 12 or 13 cities, but I imagine it took them a lot of plays to arrive at that number, and I believe it did evolve over time even though the game itself didn't change. If the actual right number is 14 and they started off recommending 10 is that a big problem or an honest best try? How is the GM even supposed to know what the actual right number is? In some cases the GM may actually be the best player of the game and have spent the time to really nail the number down but I suspect it's way more likely the GM is merely average or hasn't invested the time.

I believe the year I won Puerto Rico the GM posted historical seat data, but not bidding data, to let everyone know what had happened in the past. I ended up winning with a bid of 2 for corn which had apparently never happened before. Is that a case of getting lucky or did my little play group figure out a better bid for corn than anyone else? Maybe some of both? But if the GM had actually made a big deal that a bid of 2 or more is unplayable he may have swayed the way I bid and may have made it so I didn't win.

A Few Acres of Snow, the game I will be GMing, is not very balanced. I strongly believe the British will win every game between top players. There was a bidding system in place last year and I just copied those rules for this year too. I know at some point the game switches from an auto-win for the British to an auto-win for the French. 39 in particular! So the 'correct' point is going to be somewhere between 0 and 39. But I have no one to test with and I can't use this rule variant on Yucata so I have no way to know the real number. What should I be telling people?

Here's a final wrinkle... If you remove the games I played from the equation it looks like the British won 12 games and the French won 17. So while I firmly believe the British are guaranteed to win it turns out the French actually won 70.6% of the games I didn't play. And since my bids of 5 and 6 for the British were pretty much the highest bids in the event it wasn't because people were paying too much for the British. On the other hand 9 of those 17 French wins were from the three people in the Henning group who had a really refined French plan which I'm sure took less experienced British players by surprise.

The problem is if the French win most games between inexperienced players then advocating everyone bids really high for the British means the French are apt to win the vast majority of the games, while the British will keep winning the games when a really good British player is playing. As a really good British player I like this situation, but I don't think it is appropriate to cause it to happen as the GM. But on the other hand if I don't let people know that high bids are probably the right way to go in top tier games that's also an advantage for me. And I don't even know what the right bid is because I can't test it at all. I've played a total of 6 games with this rule set, all at last year's WBC!

Also, do I have an obligation to teach new players how to win the game or just the rules of the game? If I could convince everyone to just play British military games I could probably get a good bid value by next year. But I don't think I can convince everyone to just play military games in part because lots of people like to play games for fun and not just to win and in part because there's actually a lot of subtle little things going on that the good British player needs to be aware of in order to play optimally.

I feel like I can't teach optimal play in a demo, or during the event itself, even if I thought it was a good idea. So I should let that slide... But if I let that slide, and accept that many games will play out with a points focus should I advocate that people start bidding for the French? In a game between two newer players the French will actually win most of the time I think because they start with many more points on the board and are better able to handle a bloated deck. Is it even fair to come out and basically say anyone who loses as the British is terrible at the game?

My feeling for right now is to just give a rules centered demo, point out that each side has advantages, and mention that last year's event was won by a British military strategy with a bid of 6. Let people do what they want from that point. Ideally the event will end up with higher bids for the British among the experienced players and we'll eventually evolve into a fair game. It's just so weird that the game is weighted in favour of the French if people don't know what they're doing and in favour of the British if they do.

1 comment:

Sthenno said...

I have never been to WBC so I don't know much about the etiquette there. It might be a good idea to use precedent to decide whether to teach people how to play or how to win. Personally, I would be inclined to teach only how to play and leave the "how to win" to James Coburn.

Ultimately people who are new to a complex and skill-intensive game are not going to win. But if they are going to win it is usually going to be by catching people off-guard. A friend who was very successful in inter-university Kendo tournaments talked about how he was more scared in the first round than in the last one. Sure the last round opponent is very good and might be able to beat you, but the first round opponent is just going to swing at you like a berserker and it would be foolish to predict what will happen.

Those berserker swings are the only edge new players have. I recall a PTQ where Josh and I broke the format and I got beaten in rounds two and three by rogue decks that had no chance against the format.

Unless you are playing tic-tac-toe you can't actually be sure you know the game perfectly. I don't know a Few Acres of Snow, but it is easily imaginable that when you change the rules of a game from "try to win" to "try to not lose by more than 8" there is a different optimal strategy. While trying to win may make you lose by 10, trying to not lose by more than 8 might be possible (e.g. maybe you change your strategy from trying to score points to trying to make the game as low scoring as possible for both sides).

By telling people how to win I think you might reduce the amount of exploration that new players do, which is probably more valuable to them than doing well in the game. Presumably if they were there for a chance of winning, they would be playing games they already know how to play.