Metagame.com - Take Your Game to the Next Level

Various famous statistics are being accepted as conventional wisdom these days. The term “conventional wisdom” is a turn of phrase that rather changed the meaning of conventional to be synonymous with “rubbish.” Right after someone suggests that something is conventional wisdom, more often than not, what follows is a little soapbox shouting in which it is made clear that the conventions are wrong. After it all, everyone is his or her own little unique snowflake again, no longer pandering to the opinions of “reputable sources,” and order (through chaos) is restored.

My favorite “bad” statistic isn’t really that bad at all. Apparently, the most dangerous part of flying in an airplane is the drive to the airport. There are variations on this along the lines of “you are X times more likely to die in a car crash than in a plane crash” and “flying is the safest way to travel.”

Unfortunately, as happens so often in life, things just aren’t that simple. On any individual journey, you actually have about the same chance of dying, whether in a car crash or a plane crash. Fortunately, this possibility is pretty remote. Very remote, in fact. Quite astronomically unlikely. You would have every right to feel put upon if your life flashed before your eyes in any kind of crash. This isn’t to say that you can fight for some kind of immortality by spending the majority of your time traveling about (as I’m sure John Travolta is doing in his own private airplane), but let’s just say that it isn’t worth worrying about meeting the Reaper every time you go somewhere.

So, how did this simple little statistic come into being? Someone wanted to allay some fears that jumping into a big metal tube weighing hundreds of tons and full of highly explosive fuel (Maybe snakes too? - TW) and expecting it to fly without a visible means of support was a good way to meet one’s maker. Closer to God up in the air? Not a great marketing angle. Instead, they went with a slightly misleading statistic. Unless you are John Travolta or an airline pilot, it is pretty unlikely that you fly more than you drive.

Let’s say that dying in a vehicle is about a 1 in 1,000,000 possibility.* You’d need to fly thirty-seven times a day for seventy-five years in order get up to the 1,000,000 mark. Realistically, it’s very unlikely that you, or anyone you know, or anyone that they know, has died in a commercial plane flight. I don’t drive a great deal; indeed, I’ve probably only been in a car on the move five or six times in the last week. Most people I know would typically go on more than one car journey a day. Between people I know, and people they know, that is millions and millions of car journeys. Thus it stands to reason that I am aware of people who have died in car crashes. The fact that I know a lot of people does shift the chances of my knowing someone who has died in a car crash up a bit, but ultimately, shift happens.

The key to my point is that without a little perspective, numbers don’t really tell you a whole lot. The reason that the plane statistic is of any value comes down to human emotion—a fear of the unfamiliar and a fear of death. People are used to the risks of driving and accept them as part of normal life. Because flight is still a little bit of a mystery and a novelty for many of us, it is inherently scarier. Anything that makes it seem no more worrisome than something we do every day has to be a good thing.

Stats aren’t just used by airline marketers and John Travolta, though. They are everywhere, and it would seem that relatively often, they give rise to all sorts of bizarre “conclusions.” Stats can be really useful in Vs. System; you just need to be careful about what you do with them.

Playing the Percentages

We’ve just had a Pro Circuit, and all of a sudden, there are reams of new numbers to look into. What was the best deck? Which matchups are good for which decks, and by how much? Do players with surnames that could be forenames do better than their competition?**

For a snapshot, looking at the Top 8 seems like a good plan. But unless you are Craig Gibson, Pro Circuit photographer extraordinaire, it is unlikely that looking at just the Top 8 would work out for you. A picture might be worth a thousand words, but it takes more than a Top 8 photo (and more than a thousand words) to tell you how a big tournament went. With the split Constructed/Draft format of the Pro Circuit, basing decisions about what deck to play on what was in (and did well in) the Top 8 would be making decisions based on faulty information. While all the decks that made it through to Sunday were fine examples of top-caliber decks, you have to remember that the players piloting them also put in strong performances on the Draft day to secure their spots on Sunday.

So, if the nineteen Donkey Club players played a naughty discard deck that does unfair things, and then a fair proportion of them found themselves near the top of the standings with it, that has to be a compelling argument that their deck was the best choice for the weekend, right? The fact that all nineteen members of the team chose to play it actually spoke more to me about the power of the deck than their performance did. When you take some of the best players in the world, you will find that they do well. Some of this will be a product of having a better deck, but some of it will just be due to them finding and making plays that others might not and making fewer mistakes in complicated situations. As anyone who has ever tried to play any deck designed by Dean Sohnle or Jason Hager has discovered, decks do not just run themselves, and until you reach a very high level with some of these complicated decks, they can be mediocre at best.

There does appear to be a bias among some of the better players on the Pro Circuit toward playing complicated decks, even when, to the casual observer, they aren’t significantly different in terms of raw power from more straightforward beatdown decks. I can only assume that this comes down to the internal desire to outplay opponents and to have a feeling of greater control over how your deck is performing by virtue of playing lots of search effects and making lots of decisions. In short, the better players like to play games of skill wherein the importance of luck is minimized and the quality of play is critical.

There are some raw percentages from Day 1 of the Pro Circuit that I am always interested in, and these begin with the metagame breakdown. Fairly often at Pro Circuits, it is yours truly who is sifting through decklists, trying to label them for the purposes of a breakdown at breathtaking speed.*** Unfortunately, this is often done without as much due care and attention to some of the finer distinctions between decks as might be ideal in a world where the number of things I had to do could be counted on digits without having to resort to borrowing extras from any nearby individual with pretty limbs. Some mistakes are made, but in general, the story told by these numbers is pretty useful when looking forward. Typically, a format can get distilled into five or so decks, plus a hefty contingent of “other.” If you weigh the performance of the two of those five that did well at the Pro Circuit, and throw in any of the “other” decks that made Top 8, you’ll end up with a fair idea of how your metagame will start out in PCQs for the next season.

I have purposely left one set of percentages until last in my little statistical hootenanny, and those are the win percentages. To say that I have problems with the idea of win percentages is to put things very mildly. We do not get along. Looking at matchups in terms of win percentages can only ever serve to be frustrating. That is all. In order to get an accurate set of win percentages for two decks in a given matchup with a given initiative, you need to play a lot of games between two players who have equal skill with their own decks and perfect knowledge of the matchup and how to play it. If you have played enough to get to that stage, you do not need a number to help you, and its existence is more likely to get your hackles up when things don’t go according to plan than to be of any real comfort. If you haven’t reached that level in a matchup, then the number is not just irrelevant, but it actually moves to being genuinely misleading. Without equal skills on each side, matchups will often be won by the better player, regardless of what cards are in the decks. Suffice to say, this can be a little perturbing for the player who feels that he or she just lost a favorable matchup. Finally, win percentages don’t actually tell you anything about how to go about winning or losing a game.

If the Deep Green vs. Ivy League matchup from the quarterfinals of Pro Circuit San Francisco were to be broken down, we’d see that it clearly favors the Ivy Leaguer. Within Deep Green, there is only really Slaughter Swamp; Sage, Xavier’s Secret Weapon; and Fatality, Flawless Victory to disrupt the discard plan significantly. Ahmed Samsarra, White King is a huge liability with Deadshot around, and in Ian’s build, the only other relevant search effect is typically one of three copies of Enemy of My Enemy. Without both Sage and Fatality, the matchup becomes very tough (nigh unwinnable), and even with it, if the Ivy League deck draws a suitable collection of Straight to the Graves and discard outlets, then Deadshot can still break things up regardless. Many gave Ian ten percent odds to win two out of three games, and some would call that generous.

At what point is that percentage ever useful, though? To go along with a bad beat story when Vincent put together his two wins with a few slight play errors from Walls and a fair amount of luck to thank? The useful information simply came from knowing the matchup to a point that the guess on a percentage could be made. With that information, each player gains some sort of edge in pulling off a win. Ian sat down knowing that his mulligan condition would be to look for Sage, Fatality, or Enemy of My Enemy. His odds of hitting just Sage, assuming that he was mulliganing this way, were actually very good. Getting Fatality as well was a lot tougher, but far from the ten percent he had been given by the pundits on the floor. Rather than focusing on the numbers, Ian focused on what he needed to do to win. Part of this was to be pretty lucky, and this part he achieved. But all the luck in the world would not have helped if he hadn’t set himself up with the outs to start with.

From looking at the Pro Circuit, you can get a feel for how decks might face off against each other. And certainly, with the right feature matches, you can find some good strategies for them (against aggressive Squadron-esque decks, Deep Green likes to hide Merlyn, Deadly Archer with Rook Control). But the sample size isn’t big enough to reach any hard conclusions. With a game like Vs. that is in a constant state of flux, there is no conclusion, as the metagame continues to play out. The best plan you can make for a big event is to practice enough that you know the important cards and plays in matchups and the points of inflection at which the game is won or lost. All thoughts of numbers can then go toward counting your winnings.

And assuming it’s a long way, you might as well fly to the event. It’s about as safe as driving and doesn’t take as long.

* It isn’t. This is a number I just pulled out of my head. Quite a lot of statistics seem to be arrived at by this method. It isn’t the best, but it illustrates my point fine, so that’s okay.

** Ian Vincent and Michael Jacob certainly did pretty well for themselves.