[Article] Theory Games

[Bardo]

"Theory Games," (http://www.wizards.com/default.asp?x=mtgcom/daily/fk56) by Frank Karsten.

While not having anything specifically to do with Legacy, this article was awesome. A little clip:


Should You Expect Optimal Play or Mistakes from Opponents?

Imagine the following situation. You are playing a Gruul aggro mirror match. Your opponent led out with a Stomping Ground and a Mogg Fanatic. You made a Forest and a Llanowar Elves on your first turn. On his second turn, your opponent attacks with his Mogg Fanatic. Should you block?

If you decide to block with your Elf, he will stack damage and fling the Mogg for 1 final point of damage to your head (End result: no creatures in play, you are at 19).

If you decide not to block with your Elf, your opponent has two choices. He can either decide to keep the Mogg in play (end result: Mogg and Elf still in play, you are at 19), or he can ping your Elf (end result: no creatures in play, you are at 19; that’s the same as if you had blocked).

So if you block, you make the choice for your opponent. If you do not block, your opponent makes the choice for you. It would seem that blocking is the smart choice. Giving your opponent extra options for free does not make a lot of sense; you want to constrain him as much as possible... if everyone behaves like optimal, rational beings, that is. But humans are fallible. It may be advantageous to give your opponent bad options in order to induce a mistake. If you do not block, your opponent may make the mistake of not pinging your Elf, and then you can start casting your Troll Ascetics one turn earlier (your opponent does not know you have it of course, but if he is smart then he can guess you have something like that, because you don’t block). If you expect him to make that mistake, it is probably correct to not block there. But this comes at a cost.

By responding to potential weaknesses in your opponent’s strategy, you leave exploitable holes in your own strategy. It’s just like in poker; the strategy that the other players at the table cannot improve against is not the one that has the best success (as it doesn’t try to exploit opponent’s mistakes). Perhaps your opponent was really hoping you would not block, so that he could keep his Mogg Fanatic in play. He may have multiple Moldervine Cloaks in hand and no other creatures. And then not blocking will have grave consequences.

Anyway, Karsten has been in my Top 5 MtG Writers pile for a while and this is an exceptional piece by him.

[Maveric78f]

And by not blocking, you show that the best option, according to you, is to make the elf live, that is to say that you will probably take advantage of your extra mana. From this observation, your opponent may infer that the best strategy plan is not to play his creatures but it is preferable to play land destruction.

For instance if the mogg player has in hand forest + birds + tarmogoyf + mountain + creeping mold + random stuff. The fact that you did not block may make him decide to play bird instead of a better mana optimized play of tarmogoyf.

Edit : I mean that the choice of blocking or not is the choice to enter in a bluff mode.

[zulander]

Or that hey may mogg your birds, then follow up with goyf and put pressure on you.

[T is for TOOL]

I don't think the scenario outlined in that excerpt illustrates the point the author was trying to make at all. The situation presented would have been more applicable to the point if the victor of the game was at stake and not a mere point of damage. Such an example would better emphasize the importance of the information war and its affect on determining the correct play.

[Maveric78f]

With my example, I only mean that anyway, a smart player will burn your elf and the fact that you have taken the option to leave the opponent the choice would only give him some information about your hand and your deck, except if you bluff... But as it is likely that you reveal quite soon your hand by playing it, the bluff interest in this example is only worth 1 or 2 turns. Personnally I would never leave the option to my opponent, but I don't claim to have the best play. Trying to go psychologic is nice, but it has to be worth.

A good parallel with legacy of this kind of play is to begin a game with a fetch. To use it to search for a tropical island and pass. Your opponent may have wasteland in hand, but your play shows that you don't care about wasteland. Maybe you don't care because you have stifle, because you have 4 other lands in hand, because you were more afraid of getting your fetch stifled or simplier because you are a noob and you don't even know the card wasteland. The wasteland player will have a bad expectation of the utility of playing wasteland and crack it right now, because he has a lot to lose (wasteland + land drop), against nothing to gain (if you have 4 other lands in hand) and a small probability of having a real profit (a player who fetches for the dual in the 2 best colours of legacy does not look like a noob).
Personnally, I fetch more and more for duals when I'm on the play before ending my turn, even if I have nothing to play.

[T is for TOOL]

With my example, I only mean that anyway, a smart player will burn your elf and the fact that you have taken the option to leave the opponent the choice would only give him some information about your hand and your deck, except if you bluff... But as it is likely that you reveal quite soon your hand by playing it, the bluff interest in this example is only worth 1 or 2 turns. Personnally I would never leave the option to my opponent, but I don't claim to have the best play. Trying to go psychologic is nice, but it has to be worth.
The correct play in that scenario is to not block the Fanatic. Your opponent already had control of the decision to trade when they opted to attack. Since intent to utilize the elf became self-evident when you cast it, not blocking merely signals that you are giving your opponent the opportunity to make a mistake and not kill it. There is no reason to force your opponent to play correctly.

[FoolofaTook]

It's a decent article. I thought it was interesting that he picked the chance element in a Magic match as being approximately 33% after analyzing data that suggested the same. This is the number that was widely accepted a dozen years ago when tournament Magic first arrived on the scene. The interesting thing is that despite the introduction of 11,000 cards in the interim and the change to a more forgiving mulligan rule the chance factor in the game has remained relatively constant.

The variables that people looked at then were the decks being played, then the players playing them and finally the random chance that anybody could be knocked out by a bad draw at the wrong time. You'd hear somebody say: "Sean is playing such and such a deck" and you'd know that he was a very strong candidate to top 16 at the 512 person single elimination tournament and then you'd be surprised when he went out in the round of 32 to an inferior deck or player. The odds had caught up with him today and the chance element had eliminated him.

[DragoFireheart]

It's a decent article. I thought it was interesting that he picked the chance element in a Magic match as being approximately 33% after analyzing data that suggested the same. This is the number that was widely accepted a dozen years ago when tournament Magic first arrived on the scene. The interesting thing is that despite the introduction of 11,000 cards in the interim and the change to a more forgiving mulligan rule the chance factor in the game has remained relatively constant.

The variables that people looked at then were the decks being played, then the players playing them and finally the random chance that anybody could be knocked out by a bad draw at the wrong time. You'd hear somebody say: "Sean is playing such and such a deck" and you'd know that he was a very strong candidate to top 16 at the 512 person single elimination tournament and then you'd be surprised when he went out in the round of 32 to an inferior deck or player. The odds had caught up with him today and the chance element had eliminated him.

So how do we know for sure that a deck is of great quality and simply did not have a great deal of luck on their side? [Assuming the player that used the deck is a great player].

[frogboy]

http://www.starcitygames.com/php/news/article/6843.html has a better example of this idea.

[troopatroop]

http://www.starcitygames.com/php/news/article/6843.html has a better example of this idea.

That was a good fucking article.

[DragoFireheart]

That was a good fucking article.

That article was 11/10 and I would read it again.

[FoolofaTook]

So how do we know for sure that a deck is of great quality and simply did not have a great deal of luck on their side? [Assuming the player that used the deck is a great player].

Well, at least back then, the decks appeared in top 8's a bit more often than the players. When the Fastbond Recursion decks were in style they top 8'd more often than any of the players who ran them. Then you had the first wave of permission decks (Zak Dolan variety), which did the same. Then Balance and finally the answer decks (Brian Weissman variety) that just looked for card advantage and to wear the opponent down.

There's a reason that unlike Tennis or Golf or any of the individual sports that rely solely on skill and ability Magic has never had a dominant player for more than a year or two. Nobody has just won a third of the large events each year for three or four years running. It's the chance factor that stops that from happening.

The competition among decks for top billing is a different proposition. The top deck changes fairly frequently because the meta looks for answers to it and generally finds them. But the top players change because of luck. And that's not saying they're not very talented and knowledgable, because they are, it's just they'd need more luck to stay on top than any human being can produce.

[frogboy]

Nobody has just won a third of the large events each year for three or four years running. It's the chance factor that stops that from happening.

It has more to do with Finkel making seven figures playing blackjack and sportsbetting and Kai finishing school and playing WoW but hey continue blaming bad luck.

[Maveric78f]

The correct play in that scenario is to not block the Fanatic. Your opponent already had control of the decision to trade when they opted to attack. Since intent to utilize the elf became self-evident when you cast it, not blocking merely signals that you are giving your opponent the opportunity to make a mistake and not kill it. There is no reason to force your opponent to play correctly.

Well you're right, my reasoning was faulty. In order not to give information to opponent, one only needs to always have the same behaviour. In this case, not blocking is the best as it allows misplays from opponent.

[DeathwingZERO]

It has more to do with Finkel making seven figures playing blackjack and sportsbetting and Kai finishing school and playing WoW but hey continue blaming bad luck.

Not saying that it's going to make much of a difference, but didn't Finkel come back to the game a number of times in the last few years only to scrub out, because his heart's not in the game anymore?

And I guess that bit about Kai sums up why he's not around anymore. I was wondering where he went off to, he fell off the face of Magic faster than Finkel did.

EDIT: Also, both articles were great reads. It's amazing to see someone actually taking the time out to explain how each option has it's setbacks, and how some of the most obscure plays and a little dumb luck with a forgetful opponent can swing out a win.

[FoolofaTook]

It has more to do with Finkel making seven figures playing blackjack and sportsbetting and Kai finishing school and playing WoW but hey continue blaming bad luck.

Nobody gives up something they love and are good at if they think they can still compete at the highest levels.

The money playing blackjack is fine but if Jon Finkel thought he could stay absolutely on top in Magic he would have.

I don't know Finkel well at all. I met him a couple of times in the mid to late 90's before he had won a lot. I know Zvi at lot better than him because I used to practice with him when he was 15 at Neutral Ground in NYC. Zvi's probably the closest you'll ever get to somebody capable of staying on top for a significant period of time, because he loves the game and he was motivated to keep playing it against significant opposition in his personal life.

I had more than one conversation with Zvi that centered totally on the random nature of the game and how even the best players were brought down by it routinely. It's not humiliating but it is humbling to walk into a venue where everybody knows your name and expects a lot from you and to finish 69th out of 512 players despite that. That's the rule, not the exception.

BTW, I'll edit this to ask a question: where are all the 40 somethings who played Magic as 30 something when it first came out? I sat in tournaments where 20% of the registration was people my age, some of them quite good. Now I look at even the Legacy top 8's and everybody is south of 30, mostly south of 24. That tells me that despite the appeal of Magic something turns people off after a relatively short playing career.

[DeathwingZERO]

As far as I can tell, the younger generations are the ones that are playing the tournaments now. We've got much younger people now than we used to that are grasping the tournament level play, and are dedicated completely to it. I don't think we're seeing that in the older generations, I personally think they just want the tournament for the excitement, not 100% dead set on making a top 8. I think with all the newcomers being so young, the "dinosaurs" are becoming extinct, for lack of a better reference. The recreational aspect of olden days tournies are way more fast paced and technical now.

I know it took me forever to actually want to break from casual to competitive play. I've been playing well over a decade now, and had just gotten into competitive around 03-04. When I finally did, I felt that I was already way behind the rest of the field. It took me to hitting 32nd in a 130+ Regionals (my first) with a deck I swapped to literally the night before to realize that I could actually grasp the game on the same level as most competitive players, but even then I still feel that I'm no better than the average person that walks into a tournament because it's fun. I think that also has something to do with it, older guys just don't like the pressure involved with keeping up to date on everything to stay competitive.

[Artowis]

Nobody gives up something they love and are good at if they think they can still compete at the highest levels.


Barry Sanders down?

[FoolofaTook]

Barry Sanders down?

Barry Sanders hated losing. Point made.

[kirdape3]

Jim Brown, and his team was at the top of the NFL and he ran for well over 1000 yards his last year.

[FoolofaTook]

Jim Brown, and his team was at the top of the NFL and he ran for well over 1000 yards his last year.

Jim Brown got into an argument with Art Modell about playing football vs acting. He'd have stayed in the game for a couple of more seasons if Modell allowed him to finish filming the Dirty Dozen. Modell said no so he retired.

Bad move for Modell and probably for Brown as well given he was a much better football player than actor.

[T is for TOOL]

People stop playing magic professionally because they move on in life and become busier or simply change their priorities. Playing magic professionally isn't very appealing to most people anyway, particularly since your salary is guarenteed to be mediocre at best. Magic is simply a cardgame, not a lifestyle.

[Wallace]

Spammery sauce--and not the yummy kind. - Bardo

[TeenieBopper]

Not saying that it's going to make much of a difference, but didn't Finkel come back to the game a number of times in the last few years only to scrub out, because his heart's not in the game anymore?


Yeah, except for the part where he went into a limited PT with no experience and made top 8. Jon Finkel is hands down the best magic player ever. He's literally the Michael Jordan of Magic.

[Nihil Credo]

I remember Dave Williams saying that he still played Magic for the enjoyment, even though he'd make a hell of a lot more money playing Poker for the same amount of time.

[Wallace]

I remember Dave Williams saying that he still played Magic for the enjoyment, even though he'd make a hell of a lot more money playing Poker for the same amount of time.

At GP Richmond I reged My deck right next to Dave Williams. This was right after he took 2nd at the WSOP to Raymor. People would not leave him alone.

[DeathwingZERO]

People stop playing magic professionally because they move on in life and become busier or simply change their priorities. Playing magic professionally isn't very appealing to most people anyway, particularly since your salary is guarenteed to be mediocre at best. Magic is simply a cardgame, not a lifestyle.

Obviously you haven't paid much attention to the PT players who are playing strictly in their own countries, a few times a year, and comparing/contrasting with those who do it year round.

The Japanese players who are PT calibur are averaging about $30k a year on just playing when things hit Japan or nearby, for example. A lot of the guys that are in the high end of the PT "Payout" program are making somewhere in the neighborhood of $60k/year along with the fact they get everything comped once high enough, and the ones at the top also get a "showing fee", basically paying them to continue to play regardless of if they hit finalists. This is nothing in comparison to how some of the players have gone on to poker and made multiple digits in much less time, but those players are even fewer than the PT Points guys.

While most players aren't good enough to hit that, the ones dedicated enough to actually make it a lifestyle are making damn good money for it. Not to mention they can always get themselves into actual good jobs like Rune and Zvi, etc have done.

[T is for TOOL]

$60k/year for the few best of the best? Many companies offer comprable starting salaries for new workers in addition to others benefits like retirement and insurance plans. These salaries only increase with time, and they don't have nearly as much pressure to perform. Follow one of these people and a Magic pro for a few years and see who is better off at the end of that time.

[Maveric78f]

I am really sorry that such a thread on Game Theory discusses only about the MTG stars and not their play.

If I understand well your debate, you only want to find who knows the best the life of magic champions.

[T is for TOOL]

I am really sorry that such a thread on Game Theory discusses only about the MTG stars and not their play.

If I understand well your debate, you only want to find who knows the best the life of magic champions.

I do not understand this quote. As far as I know, that is not the topic that is being debated here. Bardo included a snippet of an article about theory, but I felt that since the example used in the article had a correct play that didn't depend on bluffing or additional information, it was a poor example to illustrate the point of the article. Frogboy then linked to another example which did a better job of demonstrating bluffing although they relied on opponent play mistakes as well. FoolofaTook then made the claim that people don't stay dominant in Magic for long because the luck factor of the game keeps them from staying on top. My response to that was that professional magic players don't make a lot of money, and their priorities change after they've been on the scene for a little while. Deathwing then responded that the high-end pros can earn upwards of 60k/year and that some get themselves good jobs afterwards. My most recent response to that is that 60k/year is hardly impressive for professionals that are the best in their field. I will further add that those that actually play Magic for profit soon realize that they can make greater profit doing other things.

[FoolofaTook]

The thread is off course and for that I apologize. Maybe we should have a separate thread to discuss the subtopic of chance and its overall effect on the Magic community and metagame.

The springboard post for the chance discussion was actually one that centered on what the chance factor was for aggro decks and control decks and how to express that chance factor in a way that it could be evaluated.

This is a very valid part of the Theory Games discussion and I'd like to return to it.

Threshold seems to be the overall most predictable deck in the format in terms of it's ability to guarantee its effect in a timeframe that is germane to winning the game against most opponents.

If the chance factor is really as high as it seems to be, approaching 33% of the matches, then shouldn't Threshold be the deck of choice to play in this meta?

It seems to me that either the chance factor is not that high or alternately that people are making irrational decisions in terms of what to play when an obvious best deck is presenting it self to the meta. Similar to what happened at the Flash Grand Prix when upwards of 85% of the attendants chose to play clearly inferior decks.