Thursday, September 12, 2013

A Few Acres of Snow Stats

Part of being a GM at the World Boardgaming Championships is agreeing to submit a recap of the tournament for inclusion on the website and in the printed yearbook. Personally I read every recap for every event every year so these things have a lot of importance to me. And yet earlier this week the convention director for WBC had to send me an email because I have, in fact, missed the deadline for submitting my recap for A Few Acres of Snow. Because I'm a lazy bum, and a terrible GM. The deadline doesn't actually mean anything because there are no immediate consequences and if there's one thing I've learned about myself over the years I'm either going to get things done right away or they're just going to slip my mind and get left to the absolute last minute. I don't like that this is true, but I'm 'absent minded' and that doesn't seem likely to change. It's why I had to go to a 'post once a day' format for my blog (even if I've fudged what a day means) because when I just posted when I felt like it I never actually posted.

At any rate, there is a second deadline for submitting a recap, and if I don't hit that one then my event loses a prize level for next year. There's even a third deadline, and if I don't hit that one then my event loses a status level for next year. I'm pretty sure A Few Acres of Snow is sunk by attendance issues anyway, but I still want there to be a recap and if somehow it does get back in the century I really want it to have 3 prizes instead of 2. Whoever gets 3rd place will be glad I did! (Which was me this year!) I really don't want it kicked out entirely for next year. Two events which I'd previously won got the boot last year and that made me sad.

The first step to getting this done is parsing through all the data I have on the bidding system since that's probably going to be the key thing discussed in the recap. I had every game fill out a form indicating the bid and result of the game, so I need to enter that into a spreadsheet...

Having now done so, I have data for 25 games. 14 of the games were French wins, 11 were British. The 14 French wins came from an adjudication, a cube out, 8 disk outs, a spoils win, and 3 military wins. The 11 British wins were 2 disk outs, 4 military wins, and 5 spoils wins. One of the disk outs there was a French disk out with the British having the high score. (This was my semifinal game, where I played my last disk out and Nick Henning had one chance to start a siege to keep the game going and did. Winning that siege swung the point total in his favour.) The interesting thing compared to last year is the military win ratio with the French winning almost as many games by military as the British did. I think part of the problem is going to be with the spoil wins since I'm pretty confident the British spoil wins were earned through military means and the French one was likely through raiding.

How about if I split out by bid? Well, bidding doesn't seem to matter. Twice the bid was for 7, and each side won one of the games. Twice the bid was for 6, and each side won one of the games. The British had a 3-1 lead when the bid was 5. The French won 11 of the 17 games when the bid was 4 or lower.

My assertion has always been that the British will win every single game with no bidding system, but the numbers don't back that up at all. If anything the stats make it seem like the French win with no bidding system while the British win once some bidding happens. And that might make sense if people were bidding for the French, but every single bid was for the British!

What I think is actually overshadowing the bidding system is the large difference in skill levels between some of the players who showed up. I had several people play their first game after the demo, and I suspect most people in the event had played only a handful of games in their lives. I've played 315 games on Yucata and a few more in person and I think the two finalists this year had played quite a few games against each other over the past year. Someone good is going to beat someone bad regardless of side choice. And when two newer players play against each other the French have an overwhelming advantage because they start with such a huge point and military lead. The British have an unstoppable line of play to overcome those advantages, but if you don't take that line of play you're in trouble.

If I restrict my numbers to just the last three rounds I have 6 games where the British won 5 of 6. The bids still don't make a whole lot of sense, with there being British wins at 1, 2, 3, 6, and 7. The French won a game at 4.

What I do know, from watching and playing the games, is the 6 and 7 wins for the British were both very close. In both of those games the French were able to play all their disks. The British had to start an attack on their last turn to win. The British had one card in their deck to use to start that attack, and had drawn about half their cards on this shuffle. (And needed to also have a boat and some military cards.) Both games were probably close to 50-50 on the last turn of the game to see which side would win, and it just happened that both times the British got the card they needed. I know in my game at least I put those 7 free draws to good use just getting to the point where I could disk out, and the British would certainly have been able to launch their attack if I didn't have them.

So is the right bid in the 6-7 range? Would having an 8th draw have saved me in my game for sure? Would it have swung the odds a little more in my favour? What about the 9th or 10th? The French did win a semifinal game with a bid of 4, so maybe the number should be lower? Was the play just different at that table? The same French player didn't win the finals with the bid at 6. So that could be a skill difference, or the whole 4-7 range could just be enough to introduce some randomness. It certainly feels like there is probably a bid range where neither side is guaranteed to win which is not something I was convinced was possible after last year, and which may be enough to convince Robb that this should actually be a WBC event. If only enough people had shown up with copies of the game so I didn't have to turn so many people away. Losing almost a third of our attendance in a year where attendance was up in general is probably a death knell for the event.

Well, now I have the data in a reasonable spot. Now to actually try write a reasonable recap for the website...

No comments: