There have been some changes.
I know some of you might have come here today in search of the latest Richard Feldman tech…
*Strokes white cat in evil manner.*
I’m sorry, but Mr Feldman isn’t here today.
*Grins evilly.*
He was a threat, you see. This was the real reason I travelled all the way out to GP: Columbus. Unbeknownst to Rich, while we were eating at a Greek restaurant, I slipped a nanotryptic flukewyrm info-probulator contrapulation into his salad. The nanotryptic flukewyrm info-probulator contrapulation (or “Fred,” to those with the technical know-how) was ingested and travelled through Richard’s digestive system. It worked its way through his stomach and small intestine before attaching itself to the colon wall. Once there, it injected part of itself through the colon wall and into the bloodstream. This part travelled to the brain, where it currently resides in Richard’s cerebellum and monitors his thoughts. From the information beamed to my secret satellite network, I learned that Mr Feldman did indeed represent a grave threat to my international ambitions of being the most beloved of all Magic writers (and therefore the most logical choice to usurp Mark Rosewater position at Wizards in order to subjugate R&D into producing a new series of hypno-holographic cards that would brainwash a whole generation of the world’s youth into becoming my personal secret army). Richard Feldman represented a threat to my plans.
Consequentially, I activated the secret cell I had positioned within the CIA to have Richard Feldman kidnapped and then extraordinarily renditioned to my secret island hideaway off the coast of Africa. Currently, Mr Feldman is imprisoned in my hidden lair, while my cadre of mad scientists pick his brains (literally!) and look to splice his genetic material with the remaining pieces of John Ormerod and Ben Ronaldson (having already kidnapped them some years before).
So today, and for every Tuesday from now on, you have adorable me (and you will adore me, otherwise I shall feed you to my lions).
*Strokes cat again for good measure.*
Are we clear on this…?
Good.
[Don’t worry, Feldman Fans… he’s back later in the week. — Editor Craig.]
I did want to mark the move to Tuesdays with a look at Time Spiral Block Constructed. For most people, Regionals are a distant memory, and the next event to prepare for is the PTQ season for Valencia. I spent last weekend testing with Dan Paskins, and while I’d like to talk about it, Dan has a PTQ in Manchester this weekend, and I’m playing in Grand Prix: Montreal. This is the usual dilemma that afflicts any writer. Magic is a game of incomplete information and giving away freebies, while great for reputation, can bite you in the pocket if you’re not careful.
As a writer, I can choose to:
a) Be honest, and then watch all my hard work become undone as the metagame shifts, or as someone nails me in a bubble match because they have that extra piece of information that turns a game.
b) Lie, and batter all the people who followed what I said.
c) Say nothing beforehand, and give a nice informative write-up after the event that both helps people to prepare for similar events in the future (because they’ll see what a hash I made of it) and helps the people intending to play the forthcoming PTQ season (or at least eliminate one of the potential decks after I go down in flames).
I prefer Option C, as it doesn’t torpedo my own chances while allowing me to maintain a degree of integrity.
(Plus, I made so many mistakes during our test games it was embarrassing.)
In the forthcoming weeks, I’ll be looking in detail at Block Constructed (and throwing away a lot of tickets on MTGO). This will cover the fallout from Montreal, and hopefully guide some of you folks to one of those mythical blue envelopes.
But first, we have this week’s article, and I think I’ll turn to an email sent to me after last week’s article by Scott Latham:
Please could you revisit the revisit the murky world of IDs and scoops, as I would like to read about it!
At Regionals last week, you get situations like a 4-2 player being paired down against a 3-3 player. The 4-2 guy can still win and qualify for Nationals – the 3-3 can’t. As I understand it, etiquette is that the 3-3 should scoop. Why? In doing so he denies another 4-2 player – one who is playing a 4-2 opponent, presumably – the chance of qualifying, and also ends his own hope of an event prize. Why does a 4-2 player who is paired down deserve a walk-over while two others battle it out? Or does the 3-3 concede solely because, come next year’s competition, the shoes may be reversed and he hopes for similar treatment?
And what if a 5-0-1 is playing a 4-2, i.e. one player has already qualified? The 5-0-1 clearly doesn’t care any more about qualifying, but still has event prizes, rating, and his mate on 4-2 to think about – why does the 4-2 person deserve sympathy over those causes? Or is it long-term self-interest in disguise again?
Anything that can help me understand it all and avoid an embarrassing scene at a future tournament would be gratefully received!
Yes, it’s that old chestnut, the ID/concession question. The subject has set off many a flame war, and I was about to chip in after Peter Jahn article when I decided my response was getting long enough I might as well make an article of it. Which I then felt reluctant to write, as:
a) The subject has been flogged, quartered, and eviscerated before.
And…
b) The subject is not exactly the cutting edge technology that will get anyone qualified for Valencia next week.
However, an email request is an email request, and so I’m going to throw my hat into the shark-infested waters.
(By the way, if there is something you would like covered please feel free to email me. I don’t do a Jeroen-style column, and I’m not intending to, but a spot of inspiration here and there – especially on the quiet weeks – is always welcome.)
At this point, most people are moving the cursor towards the back button. Please don’t. Stick with me and I promise you a naked picture of Jessica Alba at the end of the article*.
When it comes to the debate about IDs and concessions, there are two highly polarised extremes.
In the one camp we have those that believe that the game is sacrosanct, and all this horse-trading that goes on in tournaments is an affront to fair play and taints the game. Actually, that’s probably an unfair parody. Some people turn up to tournaments because they want to play Magic. IDs and concessions are an alien concept to them because if they agree to it, it means they don’t get to play Magic for an hour.
In the other camp we have the hardened, sell-their-own-grandmother-for-a-pro-point/invite tournament players. The tournament is all, and IDs and concessions are just a means to advancing as far as possible. They would rather buy their way to their aims, as it means that way they don’t have to worry about all that icky random stuff present when your game of choice involves a deck of cards. This, of course, is another parody.
However, on reading forum responses about this subject (usually when I should actually be doing real research), there do appear to be a couple of myths that should be quashed.
The first is that IDs/concessions are evil and inherently taint the game. Fixing results doesn’t happen in other sports because they would reflect badly on that sport.
Then there’s the question of why would someone knowingly simply concede. They must have been bribed somehow, as why else would they just roll over?
I’m going to argue that IDs and concessions are a fundamental by-product of the way tournaments are set up, and I’m going to argue why.
I won’t pretend to know anything about Psychology, but I did study a little Behavioral Ecology while I was studying for my undergraduate degree. Personally, I’m a cold-hearted bastard that believes we aren’t that much different from other animals, no matter how much we try and delude ourselves with our enormous brains. The same observations that can be drawn from other organisms can also be applied to humans.
I remember one of my lecturers at Bristol contrasting the two disciplines with an explanation of why parents don’t get on with their teenage offspring (in humans, anyway).
The Behavioral Ecology explanation for this friction is that the parents have to invest a certain amount of resources to maximise their offspring’s chance of survival. Once you pass a certain point, the additional investment is not going to discernibly improve their return (an offspring that survives to reproduce), and so it makes sense to kick them out to fend for themselves. You’ve given them enough, and it’s time for them to stand on their own two feet rather than moping around the house all day and drinking all the beer.
The Freudian explanation is that the father is worried his son is going to kill him so he can have sex with his mother.
Dunno about you, but I know which theory makes more sense to me.
I like Biology, and evolutionary theories in particular. While Biology is sometimes regarded as one of the woollier sciences, I’ve always seen it as providing some nice down-to-earth explanations for the madness that surrounds us.
Everything is an optimisation problem. Every organism, whether consciously or unconsciously, is striving for the most optimal way to exist. The ones that manage to do this survive, and the ones that don’t… well, we get to dig up their fossils.
People are organisms, and aren’t any different in this regard. It always amuses me when people fail to take this into account when dealing with other people.
A favorite example of mine was when a building was being redeveloped at the university where I work. The pavement (or “sidewalk” to our colonial cousins) was blocked off for the building work, and a sign was put up to redirect people all the way around the building.
You can guess what happened.
Everyone just walked down the road because it was quicker than walking all the way around the building.
The lesson from this is that people are always going to take the path of least resistance. We’re hardwired to do so because evolution and life is the most complex optimisation problem in the existence of existence. Anyone dealing with people should take this to heart, as guiding the flow is always a better plan than trying to block it.
So what does this have to do with Magic, and IDs/Concessions, you might ask…
Everything, as it happens, because everything is an optimisation problem.
When most players are playing in a tournament, they are trying to optimize their performance within that tournament. Tournaments are usually run using a Swiss system, which means that a certain score is required to reach a certain cut-off point. The outcome of any game can be a win, loss, or draw, and at some point both players will find themselves in a position where a draw is enough for both of them.
At this point, you have two options: play or draw.
Assuming equal skill if you choose to play, then there’s a 50% chance of eliminating yourself from the tournament. Even if there is a large difference in the level of skill, then there is still a percentage chance of eliminating yourself from the tournament.
This contrasts with the draw option, where there is zero percent chance of eliminating yourself from the tournament.
Obviously the draw option is considerably more optimal than playing, and so we shouldn’t be surprised that IDs exist, given that we have evolved to seek optimisation wherever possible.
So, under the current tournament situation, IDs are bound to exist as they represent an optimal strategy. By optimal strategy, I mean a behavior that would result in the most tournament success over time.
You can try and counter this by throwing up barriers, i.e. make IDs illegal, but remember the example I used before. People are always going to take the path of least resistance whether you want them to or not. In this case, they’re just going to play real slowly until the game times out, because that would be the most optimal strategy under the given framework.
That explains why intentional draws happen, and why they aren’t going away.
Now onto the trickier subject of concessions.
How do these occur? “Taking the fall” is obviously not an optimal strategy for one player, as it almost certainly eliminates him from the tournament. Because of this, some people assume that concessions cannot occur without something shady like a bribe taking place.
I’m hopefully going to explain why this is a fallacy, and that why a concession is an optimal strategy even if you receive absolutely nothing from your opponent. For this, I’m going to delve a little into game theory and evolutionary stable strategies. An Evolutionary Stable Strategy (ESS) is one that cannot be successfully invaded by another competing strategy.
For this, I’m going to go into some fairly basic Game Theory (it will need to be basic, as I’m not an expert). The classic example is the Prisoner’s Dilemma. If you’re not familiar, then the basic explanation is this:
Two people are in custody as suspects for an unsuspected crime. They have the choice of ratting the other person out, or staying silent. If both stay silent, then both are sent to prison for six months. If they both rat each other out, then they both serve five years each. However, if one chooses to stay silent while his mate rats him out, then the silent one goes down for ten years while the treacherous bastard walks free.
The interesting thing is that while the most optimal strategy for both would be to stay silent (as the combined jail time is less), the most optimal strategy for an individual is actually to rat his mate out. So, even though you know it would best for both of you to stay silent, you still have to rat your mate out because it’s the strategy that would give you the most reward.
However, things get more complicated when the same dilemma is repeated again and again with the same protagonists. Now there is a history to affect the decision whether or not to cooperate or betray. Given the same dilemma, how do you behave if your partner betrayed you last time?
I imagine the natural response from most of you out there is to turn the swine in this time, as they knifed you in the back last time.
But what if they were loyal last time?
I imagine most of you would choose to be loyal this time.
This is natural, as research has shown a tit-for-tat strategy with initial cooperation is the most optimal strategy overall. You stick by them if they were loyal last time, and shiv them if they ratted you out.
Again, we go back to Magic, and the Swiss system of a tournament. Because of the way pairings work, it is inevitable that you will get situations where someone who has already reached the cut-off point will be paired down against someone who hasn’t. In our simplified, abstract example we have a situation where the outcome — win, lose, or draw — doesn’t matter to one player (A), while the other player (B) needs a win.
If there were only one tournament ever, then it wouldn’t really matter what player A did. He’s 100% in, and could pick any option here at random. It wouldn’t matter. Player B still needs that win, and the only way he could make the loss option more advantageous than the other two for Player A is to offer some reward. This leads some to assume that bribery must take place for a concession to occur.
However, our one tournament example is grossly simplified. Obviously, players get to play in multiple tournaments, and – more importantly – they get to play in the same tournaments multiple times.
Now we examine the same situation again. Player A has the choice to play or scoop. We assume that neither affects his gain from this tournament. However, he gets to play in multiple tournaments with Player B, and there is every reason to assume that at some point in the future this situation will be reversed. We know that tit-for-tat is the most optimal strategy for this kind of circumstance, which means that Player B’s decision in the future will be affected by Player A’s decision now. If Player A cooperates (scoops) then he would expect Player B to do likewise in the future. So while there is no overall benefit to scooping over playing for this particular tournament, over many tournaments the optimal decision for Player A (the one that would give him most Top 8s) is to scoop to Player B, regardless of whether B offered him any external benefit or not.
As you can see from our abstract examples, IDs and concessions exist because they are optimal strategies.
But Prof, this is so cold-blooded… isn’t Magic supposed to be about fun?
This is where it gets complicated. The examples I’ve outlined are totally abstract, but obviously in real life there are other factors.
For a start, I’ve talked about optimization in terms of tournament performance, but this is not necessarily the most important criterion.
Different players seek different things from the game. For somebody, the most important aim at a tournament is to enjoy a game of Magic. The value they are optimizing is the amount of fun they get from playing a game of Magic. In this case, they are almost certainly going to choose to play, because you can’t have any fun playing a game of Magic if you are not actually playing.
For other people, the tournament itself is the game. The value they are trying to optimize is their performance at the tournament. In which case, they will ID and concede when appropriate, as these are the best strategies.
It gets more complicated when we step even further outwards and look at the Pro Player club. While fantastic at encouraging players to travel to GPs on different continents, the one flaw with the Pro Club system is that it will promote cooperative strategies that will see players concede in tournaments they still have a chance of winning.
A further complication arises when we consider that, often, players don’t enter tournaments as individuals. They have friends and team-mates. This gives rise to more complicated goals, such as maximising the overall team’s performance.
And then there are other rewards external to the tournament itself, such as rankings and the benefits that sometimes gives.
Okay, so now it’s time to pull my head out of abstract mode and into real life.
One of the points often raised against IDs and concessions is that you don’t see it happening in other games. Unfortunately, I’m going to have to stick with what I know and use British football (soccer) teams as an example, because I don’t follow any of the U.S. sports. There are two relevant scenarios here based on the last match of the season. In one, a team has already won the league and is playing against a side threatened with relegation. Why don’t they just throw the game? In the other scenario, a side is gunning for the title and playing against a side that is mid-table (safe from relegation, but no chance of winning the title either). You’d expect the mid-table side to throw the match… right?
Except there are many more factors in play than simply “winning the league” or “surviving relegation.” For starters, a team is not a single unit but a collection of different individuals who all have their own measures of success they are trying to optimize.
A concrete example would be to look at Manchester United playing West Ham United in the last game of the season. Manchester United have already won the league title, but still have the FA Cup Final to play. West Ham United need to win to avoid relegation.
For West Ham, the goal is clear. They need to win. For the individual players, the goal is clear. They need to win.
For Manchester United, the goals are more nebulous. The result doesn’t matter, but soccer is a spectator sport. Fans have paid money to watch the game. If Manchester United’s players stand on the sidelines and watch West Ham bash in twenty goals, those money-paying fans aren’t going to be happy and will then be less likely to return, reducing Manchester United’s income. However, manager Sir Alex Ferguson also has one eye on the FA Cup Final. His goal will be to make sure none of his star players are injured for the bigger game in a week’s time. So, predictably, he didn’t field all of them. But then we have the crucial difference between a team and individual sport. For the individual players, the goal is to ensure regular first team selection. For this, they need to play well to attract the attention of both their current manager (and other managers that might hire them), so correspondingly their goal, especially if they’re only playing because another player is being rested, is also to win!
So even though the result doesn’t matter to Manchester United the club, it does matter a great deal to the actual players who are on the pitch wearing Manchester United shirts!
For this reason, trying to compare Magic with other sports is clearly a case of apples and oranges.
Now I’m back to Scott Latham’s email, and the last line in particular.
Anything that can help me understand it all and avoid an embarrassing scene at a future tournament would be gratefully received!
First off, there is nothing to be embarrassed about. The most important thing to be aware of is that people have different measures of success they are trying to optimize. Mostly it is tournament success, but this is not the only measure.
While I don’t want to make any generalisations, I would hazard a guess that the majority of the staunchly anti-ID camp are predominantly playing Magic tournaments because they enjoy the game. For them, the “fun factor” and the “moral factor” are far more important than the “tournament factor.” Trying to explain to them that it is better for both of you to ID is futile, they’re looking for different things from a tournament.
Other than that, the ID decisions are largely trivial. If both players are in, then most of the time they’re going to ID. Exceptions to this occur if a player thinks he has a strong enough matchup to risk playing in order to improve his overall standing (there is usually a $2,000 difference on the PT between going out in the quarter-finals and finishing 5th as opposed to finishing 8th), or if the player has a team-mate he’s trying to help out.
Concessions are a little murkier. If you are aiming to optimize tournament success, then the correct strategy is to concede. However, there are a number of complications.
Again, there is final positioning. Billy Moreno dumped Bernado Da Costa Cabral out of PT: Los Angeles because he wanted in to finish as high as possible in the Swiss because of the difference in payout between 5th and 8th.
Then there’s the friends dilemma. Kenji Tsumura did the same to Sebastian Thaler at PT: Geneva to increase Katsuhiro Mori’s chances of making it.
Then consider the strategy/short term tournament success aspect. If it’s a bad matchup, and you know it’s bad, then it makes sense to try and knock it out now when you can afford to lose, rather than risk being eliminated by it in the Top 8. Kai Budde did this to Stefano Fiore years ago at GP: Firenze, because Fiore’s sideboard Hunted Wumpus trumped Kai’s Negators in the Trix mirror.
We also have the Prize puzzle. For some people, winning boosters is a big thing. Even if they can’t make Top 8 or qualify, they still want to try and pick up packs. If you concede to someone, they may make up the difference in what you could have won in any case, but they aren’t obligated to… and you must in no way ever EVER talk about this before the game. Not unless you want to take a short break from it, in any case.
Then you have Ranking. Most players don’t really care about this, as ranking rarely matters for anything. However, some players do care about it, and it can make a difference in terms of qualification for Nationals and Byes at GPs. From what I can determine the biggest cause of the clash between Zac Hill and Tuan Nygun was that Nygun valued his ranking more than tournament success in the long term… which, of course, he is entitled to do.
In all of these occasions, the player is perfectly entitled to force the game to be played. They have a valid reason for not just rolling over.
However, if you are the one who has the power, I would caution against being greedy. Tit-for-tat is an optimal strategy because it can defend itself against cheats (in this case, people who don’t cooperate). If you dump someone out of a tournament for short-term gain, you will decrease your overall tournament success measure. Why? Because at some point the situation is going to be reversed, and you’ll be paired against that player or one of their friends. At that point, it’s time to pay the piper…
Zac Hill calls for people to try and beat Nygun out in future actually aren’t that unreasonable (although they are a little unfair, given that Zac has an online reader base). While they might seem vindictive, it’s just behavioral strategy in action. He’s been “tatted,” so the next iteration of the game you can bet he’s going to play the role of the rat.
I had a similar situation occur a few years back at a PTQ. I was known as one of the better PTQ players, but I got saddled with a weak Sealed deck for this one. I limped to a point where I needed a win in the last round, but got paired up against a monstrously good deck. My opponent could have scooped me in, but he recognised I’d be a danger in the draft and took the opportunity to take care of me right then (which he did, as my deck was much weaker). On one hand, I don’t begrudge him for that. He weighed up the pros and cons and opted for immediate gain. However, the next time we play, if the situation is reversed, I will do my utmost to beat him out, because that is the optimal defence to “keep people honest.”
The reason this subject generates so much heat is because people don’t recognise that other people are trying to get something different from the game.
Just as in real life, people need to stop imposing their own beliefs and values over others.
People who believe the game is everything should not moan when they see IDs and concessions taking place. The people involved in these decisions are doing so because that is the most optimal way to maximise their tournament success. Ultimately it is their decision, and if it knocks other players out… well, those players should have performed better in the first place so they didn’t have to rely on this result. Chances are it will be their choice to make the ID next time round.
And vice versa, people who are looking to maximise their tournament success should not get too upset if they get drawn against someone who isn’t and refuses to play ball. People are looking for different things from the game. If you’re at the tournament because you want to play, then that’s your decision, and don’t let anyone browbeat you into conceding if you don’t want to do it.
Although IDs and concessions are an unavoidable part of tournament Magic, They are pretty much a consequence of the current tournament framework. They survive because they are optimal strategies. There are ways you can make it so this is isn’t the case (and by directing the flow, rather than trying to stem it), but I’m over my word count again, and Craig is fuming at home waiting for me to email this article, so it will have to wait for another time.
Thanks for reading,
Prof
* I lied, obviously. You know I’m evil right?