Friday, June 16, 2017

The MVP of WBC

In the last few months my main streaming game has become Blood Bowl. There's a decent computer implementation of the board game (which I have been playing off and on for 18 years) and it turns out I'm pretty good at it and people like to watch good but not perfect play. (It gives them a chance to spot better plays and feel smart!) One of the mechanics of that game is that each team gets assigned a random MVP after a game which is worth a bunch of experience to help level your players. One of my viewers from the pre-Blood Bowl days saw a discussion about the MVP and asked if I'd ever won MVP of the World Boardgaming Championships.

Now, he was being silly, as many Twitch chat people are, but the question has festered in my mind for months. There is an MVP of sorts at WBC. It is definitely not assigned at random, though. Every event at WBC awards points to the top 6 finishers based on size of the event. Sum up all of these laurels and whoever has the biggest number is the MVP. They actually have two such awards. Caesar is awarded to the person who earns the most laurels over a 12 month period, Consul is awarded to the person who earns the most laurels at WBC. They care more about Caesar because they want to encourage people to play in events outside of WBC (a bigger deal when they actually ran a second convention) but I don't much care for play by email tournaments of wargames so for me Consul is the interesting thing.

Looking at the totals needed to be Consul in the last 10 years we see 130, 129, 100, 151, 133, 133, 108, 128, 120, and 156. The most you can get from a single game is 60 and that's for winning one of the 11 biggest events. Available laurels peter off pretty quickly, with second place in a huge event being worth 36 and winning one of the next 12 biggest events being worth 50. Something like 3rd place in a 4 class event (24-53rd biggest events) is only worth 12 laurels. So to get Consul you're looking at needing to win 3 events, or maybe 2 with some other good finishes.

I did come close once, when I earned 99 laurels in 2008. I won a 6 event, won a 3 event, came 4th in a 2 event and 6th in a 3 event. That was good for 3rd place that year (and 16th for Caesar, to show how many points could be earned outside of WBC) with the winner having won 4 different events and an extra 4th place thrown in for fun. I also came 9th in 2012 where I had 2 1sts, a 2nd, and a 6th all in 3 events. That was good for 81 points where first had 133 with 3 wins, 2 seconds, a 4th, and a 6th.

So it's not outside of the realm of possibility that I could have a really good year and come out on top, but it's not actually very likely if I don't make some sort of change. In recent years I've been spending less and less time at WBC actually playing in tournaments or expecting to do well when I do. Moving to New Brunswick meant I both didn't play any games and cared more about hanging out with friends at WBC than playing in events. So while in previous years I may have done things like randomly played a heat of Tigris & Euphrates (in which I somehow came 2nd in 2010) to boost my laurels I wouldn't have done so the last couple years. Which did mean that last year was my first year in 10 where I didn't win a single plaque. I'd averaged 46 laurels per year for my first 9 years; last year I got 2. It felt a little bad. I should be better than that.

Now, I've been playing more games in the last year and in particular I'm hyped about my ability to play a new game for the first time in a long time. The format for Star Wars Rebellion sucks, which has dampened my enthusiasm, but I still have reasonable hopes of being able to win. So maybe I can just use that as my motivation for this year, but I want to think more on trying for Consul. At least think about how to best position oneself for doing so even if I don't end up actually doing it.

There are a couple of variables at work when trying to max out opportunities for laurels. Generally speaking the prize level of an event is based on the hours spent by all players on the event. So if an event takes longer it'll earn you more laurels but cost you more time. If an event has more players it'll earn you more laurels but the competition will be stiffer. These should all even out in the wash so that where you spend your time isn't terribly important... Winning one of those 11 6-prize events is a huge boost, but they should be a large time investment with a small chance of pulling it off.

Stone Age, for example, is one of the 6-prize events. It gets around 160-200 players, has 3 heats, and runs a quarterfinal. If you wanted to put in the best chance at winning the event you're probably looking at playing at least 5 games. (Either play all 3 heats to try to earn a bye through the quarterfinals or play 2 heats to get a win and then win the quarters and semis to make the finals.) So you'd be looking at investing 10 hours to get a smallish chance at the 60 laurels.

Stone Age is actually a fairly skill intensive game I think, and one I'm decent at, so I'd probably give myself a 40% chance at winning a 4-player semi and maybe a 30% chance at winning a 4-player final. So if I asserted I could get a bye I'd be looking at a 12% chance at getting 60 laurels for 10 hours. With some smaller payouts down the line too. Probably not a bad idea.

What might make it a bad idea is when those 10 hours take place. The scheduling game at WBC is not an easy one! The first heat of Stone Age conflicts with History of the World (one of the other 6-prize events) and the single elimination tournament for Innovation (an event I've won in the past). The second heat conflicts with Empire Builder (another 6-prize event), Castles of Burgundy, and Concordia. The third heat conflicts with the single elimination tournament for Star Wars Queen's Gambit (an event I've won in the past). The elimination rounds for Stone Age conflict with all kinds of other semis and finals since they start at 9am Saturday morning.

Which leads to one revelation... Find games that have no conflicts. If the thing that matters is spending time playing games (that you can play through to the finals) then playing games with no conflicts is a good plan. So things that start at 11pm and go past midnight aren't going to have any conflicts and are basically a freeroll. Play Slapshot because it doesn't cost you time you could be spending on a higher payoff event. *sigh*

Certainly one way to gain a big edge in terms of laurels earned is to play a game where you're much better than the average player. Back when Le Havre was an event at WBC I made the finals all 6 times. Even with a smaller prize level than Stone Age, that would be a much better play for me.

The flipside to that is avoid games where you're significantly worse than the average player. My chances of winning a semifinal of Agricola are likely to be in the single digits. They use extra cards that don't ship with the game and with which I have played exactly one game. If I'm trying to earn laurels I'm probably much better off playing Seven Wonders, Love Letter, Scythe, Ra!, and Las Vegas all in the time I'd have to spend playing 3 heats of Agricola. Then the next morning I could play San Juan instead of playing the Agricola semi.

Some events are basically random. If no one is better than average then you just need to understand the game enough to be average and show up. Someone has to win Can't Stop. Why not Zoidberg?

One other thing to consider is advancement conditions. Some events advance plenty of alternates or don't require you to even win a semifinal to make the final. (Top second in a semi has advanced in plenty of lesser attended games over the years.)

Then there's also the fact that prize levels are quantized. Around 50 of the events are going to be at the 2-prize level, but some of those games are going to have significantly more player-hours than others. The ones with fewer player-hours are likely to be more efficient uses of your time than the others. They may only be worth 20 laurels, but if it only takes a couple hours and there aren't many people to compete against, well, it could be a good idea.

I think if this is something I want to do the next step is to go through all the events and estimate the hours it would take to do well, and estimate the chances of actually doing well. Use this to identify a few events to focus on and then build a schedule filling in the gaps with other events that won't have elimination rounds that conflict with the core games.


Andy said...

I don't think Can't Stop is a crapshoot, though the fact that I've laureled it in two consecutive years may influence my opinion. While I agree that it's not difficult to play well, most people in the tournament, and even most people playing in the second round, play it very badly. They play way too conservatively, don't modify their level of risk enough based on how well they are doing, and don't understand how big the difference is between winning a column and being very nearly winning it.

One question that is worth asking yourself is whether maximizing your chance of being consul is the right goal; will it maximize your enjoyment of the week? Suppose you win a heat of some game whose semifinals conflict with round 2 of Star Wars Rebellion. If you want to maximize your expected laurels, it might be right not to play Star Wars Rebellion at all, because playing round 1 takes a lot of time, which will be wasted if you lose, and partially wasted even if you win, since your expected laurels from round two of SWR are probably higher than the semis of the other game, but not by much, and they'd require much greater investment of hours. But you should play SWR anyway, because you love the game, and the chance to play it against other players who also love it and are almost as good as you is why you come to WBC. I love Amun-Re, but the set of people who play it at WBC is such a shark tank that I'm very proud of my one fourth-place finish. If I was maximizing my expected laurels, it might have been correct to skip the finals! But the chance to play against David, Kevin, Eric, and Mike was a big win in expected happiness. I've even played a heat of Titan, where my chance to win a 4-player game is probably in single digits, because I enjoy playing the game even though I'm terrible (and unlikely to improve much; even if I got better at other aspects of the game, my memory isn't good enough to play competitively).

Nick Page said...

Can't Stop has been an event for 11 years and it has 62 different laurelists. There isn't a single person with more than 30 laurels in it total. There's no way that I can see to interpret this information except by saying it's basically random. If you look at data from websites that have it as a game you can play (like Yucata) the TrueSkill system they use makes it painfully clear that there's no such thing as a great Can't Stop player. (The highest TrueSkill at the moment for Can't Stop is 1286 in a system where 1200 is the base. My Stone Age TrueSkill is only 8th on the site, but is 1536. Top Stone Age players are way better than average compared to top Can't Stop players.)

Definitely agree that maximizing laurels is not the same as maximizing happiness. But I've been spending the last few years on maximizing happiness and maybe it's time for a change? I'm not sure! I haven't played Titan in forever so maybe it's time to play that again even though my laurel chances there are basically nil.