fbpx

Chasing Victory – Being Results Oriented

Being “results oriented” means that you focus on the end result and not necessarily the in-between it took to get there. For example, if you make a certain play and win strictly because of that play, it means that play was correct. After all, how could it not be? You won the game because of it!

The biggest problem with being results oriented is that it relies on the assumption that results are constant. Just because you won the game because of a certain play does not mean that it was correct. Results oriented thinking is illogical. Just because you flipped a coin and it came up heads does not mean that the coin flip will always come up heads. Nor should you think that just because that play won you the game that it will yield that same result every time. You should instead consider your plays, and make the play that has the best expected value.

Just imagine if everyone was results oriented in every facet of their life. Whenever someone failed a test in school, they would drop out. After all, why wouldn’t they? If they can’t pass a single test, how could they ever hope to graduate?

Similarly, if you are in car accident, you don’t automatically assume that every time you drive, you will be in an accident, therefore you refuse to ride in cars for the rest of your life.

I enjoyed the movie Taken, but Liam Neeson’s character was extremely results oriented. His daughter wanted to go to Europe, but he didn’t want her to. As an ex-government [card Throat Slitter]ninja[/card] he had seen exactly how dangerous the world could be, and he would rather have his daughter safe at home. She goes anyway, gets kidnapped, and Neeson goes on a killing rampage to eventually get her back.

The entire movie is basically based on Neeson saying “I told you so.” That is just ridiculous. Imagine how many people get kidnapped per year. Just because he told her not to go, because she would get kidnapped, and then she did, doesn’t mean that every time she leaves the house she will get kidnapped.

You need to realize that whether you fail or succeed isn’t nearly as important as why the result happened in the first place. Once you figure that out, it becomes much easier to deal with losing and frustration, since if you are making the right decisions and losing, that’s fine. If you’re doing everything right, in the long term, you will get the results you worked for.

Let’s take a look at the different aspects of Magic, and how people are incorrectly results oriented.

Playtesting

You might be thinking “If results mean nothing, then what’s the point of playtesting?” Well, that isn’t exactly what I said. Results matter to some degree. If you play a million games or flip a million coins, chances are you will end up with a result that is close to correct. Just remember, a coin flip can come up heads a million times in a row. It’s certainly not probable, but it’s definitely possible, so just because it was heads 60% of the time in your playtesting doesn’t mean that is how it will always be.

Say you did a draft with your friends, using the shiny new M10 set. One of your friends beats you three games in a row with his deck, always ramping up to five lands, but then drawing all spells afterward. Clearly that is hard to beat. Afterward, you ask to see his deck and you realize that he was only running 12 lands. His deck ran great for him, and having only 12 lands “seemed to be working for him,” so you decide to try running a low land count in your next draft.

You get mana screwed.

How is that possible? You “playtested” running only 12 lands, and it “worked fine” for your friend, so why not you? Well, as always, the problem was in the execution. Your “playtesting” only consisted of three games. Anything can happen in a sample size that low. You need more results to be able to make a more informed decision.

I see a lot of people using the “well, it works for me” or “it’s worked fine so far” for their reasoning behind card choices, and that is never a compelling argument for me. If you want to make a case with me, you must make a decisive stance and defend yourself. Say that you need card X for matchup Y, which is otherwise a difficult matchup. That is a reason for including that card. If there is no reason for including the card, why defend it? Just admit that it might not be right and do a little legwork to find out what the correct card would be.

In the above 12 land scenario, you can easily do a little math to figure out that 12 lands simply won’t be enough for your deck to operate on. A limited deck generally needs three or four lands to run smoothly, and you should probably have a few higher end spells or stuff like [card]Merfolk Looter[/card] to make use of your excess lands should you get flooded.

You should be using playtesting mostly as a guideline, and not as a definitive result. Too many variable go into a game of Magic to play one game, ten games, or even a hundred games, and claim that deck Y beats deck Z X% of the time.

Players who complain that “my deck beats your deck 80% of the time, so I should have won” greatly annoy me. First of all, how do you know that it’s 80%? Maybe your opponent is this tournament plays a lot better than your playtesting partner. Maybe he knows a better line of play. Maybe he has a lot of sideboard cards for your matchup. Maybe he hit his 1/5 chance to beat you. Either way, your percentage is probably wrong, but 80 most certainly doesn’t equal 100. There is always a chance to lose.

Instead of focusing on your previous results and why you “should” have won, you should figure out what went wrong. Figure out if there was something you could have done differently, or figure out if your playtesting was flawed. Maybe you didn’t prepare for card W and it completely wrecked you.

Mulliganing

If your opener is bad, throw it away. That’s easy enough. If your hand is a risky one lander, consider whether or not you’re on the draw, whether or not this hand automatically beats your opponent if you “get there,” whether or not you need to get lucky in the matchup to win, and whether or not you would be extremely happy to mulligan into that hand minus one card.

If you decide to mulligan, don’t look at the top of your deck. Who cares if there’s a land there? Odds are that it wouldn’t have been.

If you have a one land opener in your 24 land, 60 card deck, that leaves 23 lands in your 53 card deck. You don’t need to draw a land on the first turn, but you probably do on the second, so that leaves you with 23 outs into 53, and then 23 outs into 52. You are worse than 50/50 to draw a land, but it’s pretty close to a coin flip.

Most players just think that they are flipping coins for the game win, which would usually be fine in some circumstances, but that isn’t the case. You are on the bad end of a coin flip to even get to play Magic! Just because you draw a second land doesn’t mean that you automatically win most of the time.

Either way, looking at the top of your deck is stupid. The result of whether or not you would “get there” isn’t determined by what was actually on top of your deck, but by the math. Math is constant, pure, and never wrong. If you had a 42% chance to get there in two draw steps, whether or not the land was on top, you probably made the right decision, the decision with the most positive expected value.

If you do look, you develop bad habits. You will generally remember that time you “got there” or didn’t “get there” and be influenced by what happened in the past, rather than doing the math and focusing on what is “right.”

Also, just because you flip a coin and it comes up heads, doesn’t mean that it will be tails immediately afterward. 50/50 is not an indication that something will happen, and then alternate immediately to the other possible result. A coin flip can come up heads three times in a row. It’s somewhat unlikely, but can obviously happen.

If you “got there” the last game, that doesn’t make you any less likely to “get there” this game. You are on a clean slate. You still have a 42% chance, and that should be what you take into consideration, not what happened in the last similar situation.

Winning the Game, or the Tournament

Just because a certain deck won a big tournament doesn’t mean that it’s the best or even that it’s the best build of that archetype. Again, variables.

In poker, when a bad player scoops a giant pot after playing bad cards poorly, the losing player will often make fun of their bad play. When the bad player can’t defend his play, or make a coherent argument, he will often say, “Well, who’s got the chips?”

Well buddy, you do, but that is irrelevant. That doesn’t change the fact that you played poorly.

Winning tournaments with bad decks doesn’t mean that your deck is good, or that you were right.

Playing and Making Mistakes

Say you are in a situation where you have two 3/3s to your opponents 3/3. He is at nine while you are at three, it’s your main phase, and you each have no cards. What do you do?

Well, it kind of depends, but the main thing here is that you are dead if you attack with one of your guys and he draws a removal spell. However, if you don’t attack, you give him more turns to draw out of a situation where you are clearly favored.

If you attack, he draws a removal spell and you die, so what? He probably had 3-4 left in his deck, out of 25
or so cards. Those odds are in your favor.

If you don’t attack for fear of removal, and he draws it, that doesn’t make your play correct.

Paulo Vitor wrote about a play he saw, where a guy activated [card]Blinkmoth Nexus[/card], cast [card]Pyroclasm[/card], killing his own creature, and then ended up winning the game on two life while his opponent held a useless [card]Molten Rain[/card]. Does that make his play correct? He did straight up win the game because of it, but that is irrelevant.

Card Choices

I’ve seen a lot of players cut the Faerie hate from their deck, only to play against the one Faerie deck in the room round one and get demolished. Naturally, that person goes on to complain about how they knew they shouldn’t have cut the Fae hate from their deck. Obviously they were going to play vs. Faeries.

No, they weren’t. There was one Faerie deck in your hundred person tournament. The odds are greatly against you playing against that deck. You might think that you got unlucky, which is reasonable, but luck runs both ways. You take the good with the bad.

If you cut the Faerie hate from your deck, you are making the best decision possible, as 99 times out of a hundred, you won’t play against Faeries (at least round one).

Likewise, I saw a player absolutely demolish the swiss rounds of a PTQ, only to fall to [card]Time Sieve[/card] in the Top Eight, a bad matchup. That player, confident that he could beat everything except for Time Sieve, added ten cards to his sideboard to beat the bad matchup that ended his tournament last time.

He 0-2ed the next PTQ. What happened? Well, he probably lost a ton of percentage points by cutting all of his sideboard cards for the “good” matchups to make room for the anti-[card]Time Sieve[/card] cards. [card]Time Sieve[/card] wasn’t a big part of the metagame, and he didn’t play against it, and instead lost to two “good” matchups.

The problem was that he turned his “good” 65% matchups into 50% matchups by cutting a bunch of useful sideboard cards, therefore making those “good” matchups not so good anymore.

The moral of the story is to not play scared and focus on what matters. The numbers are what matter. If you expect Time Sieve or Faeries to be 1-2% of the metagame, don’t waste sideboard slots on them, as odds are you won’t play against them at all. If you do, you still have a shot! It’s not like you are drawing dead because you didn’t play with 4 Stags.

Also, don’t become resigned to your fate before the match has even started. So you’re playing against Faeries round one after cutting all your Stags. Who cares? You still have a shot, you just have to keep your play tight. Thinking about what you should or shouldn’t have done last night is irrelevant. The only thing that matters in the game you are playing right this very second. If you aren’t going to give it your full attention, you might as well concede right now.

Sideboarding

You are playing in Top Eight of your Nationals with your trusty GB Elf deck against your enemy’s UB Faerie deck. You should probably side out your [card]Nameless Inversion[/card]s, but you forget and leave one in on accident, draw it, and curse your luck. However, as the game goes on, you sculpt a turn where you can use that Inversion to pump your Tarmogoyf and get in for exactly lethal damage.

Congratulations, you just won a game based on a mistake you made! You realize the error of your ways and side out the Inversion for something better. However, there is still some merit to the situation. Your opponent, Paul Cheon, now has to consider what you sideboarded out for those Inversions, as he fully expected you to cut them from your deck, and rightfully so. He now probably has to play around them, while they are all sitting in your sideboard.

Sam Black, the Elves player, realized that just because he won the game with the Inversion didn’t mean that it should have been in his deck. The end result of the Inversion being in his deck was irrelevant, as he knew that Inversion was worse against the Faerie deck than anything else he could have brought in, and corrected his mistake after that game.

Now, what if Sam did all of that on purpose? What if he kept in a single Inversion, hoping to draw it and trick Cheon, either into playing around further Inversions or because he knew Cheon “knew” that Sam would side them out. That, my friends, is called “leveling” someone. You know exactly what level of thinking your opponent is on, and therefore you can make plays that would be “wrong,” but actually quite correct in the abstract.

My point is that everything that is right or wrong is based on perception. What might be the right play for an FNM player (siding out Inversion because it’s bad against his opponent, and he knows his opponent doesn’t try to play around anything) might not be the correct play for Sam.

What might be an 80% matchup for you against your testing partner Jimmy is going to be a coin flip against your local PTQ end boss Calosso Fuentes. Calosso plays better than Jimmy, he’s going to be a jerk and fluster you, and he focused on his deck being able to beat yours because he knew your deck would be popular.

Draft Picks

You are in the draft pod of a Grand Prix Top Eight, and your opening pack has Doom Blade, Pacifism, Overrun, and Merfolk Looter as potential options. You of course, take Jackal Familiar, get fed an insane mono red deck, and easily 6-0 the top eight.

Now, it’s pretty clear that that pick is wrong, but maybe you know something nobody else does. I know that in Shadowmoor limited, I took Intimidator Initiate over much “better” cards, but I rarely lost with mono Red. In M10, that isn’t quite the case and not taking any of the four cards I mentioned is probably a mistake, if only because that Jackal Familiar would probably come back 9th pick.

If taking Jackal Familiar first pick works well, does that make it correct? What if your neighbor is attempting to draft mono Red also, does that make it wrong, simply because you got cut out of Red?

Similarly, I see a lot of players stick to two colors when it’s perfectly reasonable, only to see a lot of Black cards in pack two and three, and then talk about how they “should” have been Black. Why should you have been black? There was no indication that Black was open until very late pack two, at which point it’s too late to switch colors.

Yes, Black was open and it sucks that you have a solid two color deck instead of the insane mono Black deck you could have had, but that doesn’t mean you should jump ship for that Dread Warlock you see fifth pick pack two. Even the sixth pick Tendrils doesn’t warrant an audible. Nor does the seventh pick Royal Assassin, or the eight pick Nightmare. At that point, you are too far committed, and not jumping into black is hardly a mistake.

If we can’t rely on results, what else is there? Well, you can rely on the results of several tournaments, but I wouldn’t want to assume that the results of one tournament, one small sample size, are entirely correct. If you start testing a matchup like Faeries vs. 5cc, you will probably be able to see in a few games that playing from the 5cc side is very frustrating. Faeries always has the counterspell, they always have the Mistbind Clique. That matchup is hard.

Bam! Three games and you’ve figured out that it’s a bad matchup, all just by thinking on your own. You didn’t have to play a hundred games to realize that Faeries is a difficult matchup, or that Mistbind Clique is one of their best cards against you. All you have to do is pay attention to what’s important and to what is causing each deck to succeed or fail. At that point, you can go about making changes to the decks and trying it all over again.

Second, there’s math and expected value. As I said earlier, as long as you are making the right plays and decisions, eventually the results will come. It’s much better to focus on that rather than your short term failures. You cannot win every game of every match, let alone every tournament.

Magic is not designed to be like that. If it were, I suspect eventually there would only be one person left playing it. No one likes being mana screwed, but everyone loves winning, and as long as anyone can win at Magic, that will keep people playing. The problem is that you, as the PT hopeful, would like to win a PTQ and probably a PT right now.

Until that happens, good luck!

GerryT

69 thoughts on “Chasing Victory – Being Results Oriented”

  1. While most of those things have already been said and repeated over and over by different sources, it’s still nice to see them all summed up like this. I will be deffinitely referencing this article for some of the points made there.

    “The moral of the story is to not play scared and focus on what matters.”

    I have this – call it a syndrome – that I don’t want to auto-lose matchups and I often have “the fear” – going overboard, just to be prepared for everything. Sometimes it is possible to go that route, other times your 75 cards are not flexible enough or even if they are – being too flexible causes you to fall prey to a more focused and straightforward strategy.

    In this particular respect your advice has helped me strenghten my resolve when selecting decks and I thank you for this.

  2. “If you have a one land opener in your 24 land, 60 card deck, that leaves 23 lands in your 53 card deck. You don't need to draw a land on the first turn, but you probably do on the second, so that leaves you with 23 outs into 53, and then 23 outs into 52. You are worse than 50/50 to draw a land, but it's pretty close to a coin flip.”

    I didn’t finish the rest of the article, but I got to the mulliganing section and just wanted to point out that 23/53 + 23/52 is not worse than 50/50. You actually have an 87.6% chance to draw another land.

  3. GerryT,

    Article was definitely worth the hype, or perceived lack thereof. Either way, I’m glad to finally have read it, and the logic in this article is undeniable. It’s hilarious to me how some individuals can parade around thinking the way they do. Shallow statements I know, but it’s 12:40a.m and I have an AIS class in the morning. Looking forward to the next one.

    Keith “MckHick” McKimmy

  4. I want to say this is a good article, but we will have to see how many hits it gets. If it doesn’t get a lot of traffic, it was probably pretty bad.

    😉

  5. Pingback: MTGBattlefield

  6. (This is a minor nitpick.)

    In the example where you’re on the draw with 23 lands left in 53 cards and need one by turn 2, the math I did gave the result that you’re 68% to get there. I haven’t been good at math in years, though, so can someone who IS good confirm/deny this?

  7. “Leveling”? Flores called it Catachresis, and that is easily ten times more ridiculous-sounding.

    (As we all know, whatever sounds more ridiculous is better to use for Magic slang)

    Article’s all right. I can’t help feeling that most of it is sort of obvious, and the rest, well… is Catachresis. Sure, it’s five years later and not talking about Onslaught block Rochester drafts. But it’s well-written, and there is nothing actively wrong with it. I’m sorry that I don’t like it more because the only strike I can find against it is “Flores did this five years ago with a sillier name and a dumber story”.

  8. Your definition of “results oriented” is a strawman, and a joke at that. It certainly is easy to bash down your imagined definition, but c’mon.

    That said, the rest of the article is okay. Just don’t assume that results oriented = a complete fucking moron about statistics.

    Results oriented + statistics = quite good, in fact!

  9. Odds of NOT drawing a land first turn = 30/53 = 0.56 (a)
    Odds of NOT drawing a land second turn = 29/52 = 0.55 (b)
    Odds of not drawing a land in either is a*b. So the odss of drawing your second land by the second turn is 1- a*b = 68%. The point remains, though.

    I liked the article

  10. who cares about the NAME? its the concept that matters, and he described the concept. Whether it’s called being results oriented or Sbrubbles, it doesn’t matter as long as it’s clear what he is talking about. That said, most people DO use his results oriented definition.

    I liked this article – as someone else has said, most of those things have been written about already, but you summed everything up pretty well and accurately

  11. Absolutely fantastic article, Gerry. This is one of your best, and is one I will be linking people to for a long time to come.

  12. Tim: You are right. Was probably thinking about the one shot drawstep instead of two.

    Seeker: I don’t remember that article, and I imagine that a lot of other people don’t as well, especially new players.

  13. @1024

    ya, his math is technically off in his 23 lands left example. statistically you figure out the chance of not drawing the land either time, and then subtract that from 1:

    1-((30/53)*(29/52)) = .684

    Ofc it’s just nitpicking cuz if you really wanna go in to it, it often matters what kind of land you get. His point gets across either way…

  14. I agree with most of the advice in this article, but it’s so hard to read. It’s just one long machine-gun-fire-paced rant. The tone, and repeated use of the pronoun “you”, also comes across as snarky and pretentious. Again, the *content* of the article is great, but the delivery leaves something to be desired. I mean, I agree with almost 100% of the info here, but I feel like I’ve just been argued with and called an idiot for the past five minutes or so while I read.

  15. At grinder before Nationals I took out fae hate, because I thought playing them at this tournament full of Mono Reds/Blightnings was plain wrong.

    Then I played fae in the first round (one of two fae players on 90-people tournament)

  16. I think the second-person works very nicely. And anything that’s an update on Flores is pretty good to me.

  17. if you have 23 lands left out of 53, that leaves 30 non-lands……the chances of drawing a non-land is 68%….the chances of drawing a land is 31%

  18. Great article. I can’t recall the number of times I’ve heard something to the effect of “well, it won X tournament” used to justify deck choice or “well, I won the match” used to justify questionable play. I approve of any article that attempts to cut these sort of crutches out from under people and replace them with careful analysis.

  19. This article is great. The thing that I think is poorly understood by most magic players is statistical confidence levels (which you indirectly refer to by talking about how small sample sizes aren’t reliable indicators).

    I’ve read several variants of this article across the years, but none that actually take a stand and tell you when you have enough results to be confident that the results you are seeing correspond to ‘reality’ (10 games sets? 30? 50?). Is that because the players capable of writing such an article don’t understand the statistics, or that the statistics would be too confusing to most magic players, or magic is too complex a beast to be able to say decisively, or what?

    “In poker, when a bad player scoops a giant pot after playing bad cards poorly, the losing player will often make fun of their bad play. When the bad player can't defend his play, or make a coherent argument, he will often say, "Well, who's got the chips?"”

    Although, sometimes your opponent is just being a jerk and giving the ‘vs opponent’ version of a bad beat whinge. I have said in the past “you’re right, I am terrible. How does it feel to lose to someone this bad?” If the opponent is sledging and trying to get into mind games, this has been an effective comeback for me.

  20. 42, 87, 68, 31. wow, that’s a lot of different odds. Either way, the point he was making still stands.

    And as I have not read anyhting like this before, I think it was a great article. Straightforward and, dare I say it – RESULTS ORIENTED!

  21. Lots of different odds … but that’s because those #s represent different outcomes. The 87 could technically be correct (it’s the odds of drawing ONE land in the first 2 turns), but the 68 is probably more correct for these purposes. 68% represents the odds of drawing at LEAST one land in the next 2 turns, the “1 minus” rule in probability theory means the P(at least 1) = 1 minus P(none at all).

    And yes, I realize this doesn’t matter at all and the point gets across either way, but as a math major I felt compelled 🙂

  22. So when you next mulligan against Gerry, pretend to look. It might take a small amount of the edge off your mulligan by putting him a little on tilt 😉

  23. Good article. I thought i would do most of the things you said correctly but upon further thought is realised, that even being good enough at statistics to get my probability of drawing out of the situation in under 2 minutes and thus should apply it to MTGO i solely rely on my gut feeling.
    I could not shake off the feeling i was in some sort of magic boot camp though. It read like being screamed at the whole time while being told that im scum and unworthy and should just run 2 hours through pounding rain to make up for my inferiority. I guess you were actually enraged about all the idiots out there that do all these simple things wrong and you can not take it any longer :-P.

    About statistics: I learned a lot of statistics in about 2 weeks for an examn (including tests and confidence ranges, as well as the basics like hypergeometrical and poisson distributions – i hope the names are correctly translated… im bad at scientific english) and the only hard thing was understanding what you actually do, not how to do it. But it is hard to make it look interesting, which might be the reason nobody writes about it.

    On playtesting: You are right. Playtesting is deceiving and should not be used as a method to get win% but as a method to see why deck X wins vs deck Y and how you have to play. In the last extended season my playtesting showed that my fearie deck had a 10% matchup in monoR. Guess what i never lost to the whole season?

    On mulliganing: Yes, the chance of at least drawing one land in the next 2 draws is about 68,5% but your point is valid nonetheless.

    On winning: There is this saying “you can´t argue with success”. Thats a damn stupid thing to say! Magic involves so much “luck” (by which i mean statistical variation) that there is no way of deducing playskill from number of victorys. Even LSV *could* just be the luckiest man alive because the sample size is still too small to reduce variance to an acceptable level (just saying this to make a point, not because i really believe LSV is in any way not a god of magic).

    On card choices: People tend to have this strange urge to have an answer to every deck in the format. Thats fine if you have only about three decks that steamroll everything but each other but in a diverse metagame, thats just nuts. You just can not beat everything. Thats a hard truth to accept because we want to win all of the time, but if a deck is below 2-5% of the metagame you should ignore it in favour of the popular decks. You have to live with the fact that there are bad matchups and you can get unlucky be facing the one guy playing your nemesis but because that is highly unlikely, why bother?

  24. So, are the odds of getting another land on your first 2 turns – are they 68% or 44% on the first turn and 43% on the second turn ?

    My friend says that if it’s only 1 game, then the 68% for 2 turns doesnt apply, and it is 44 and 43 individualy. Therefore you should always mulligan a 1 lander.

  25. Pretty fuzzy math going on in some of these comments. Some people would be served well to google ‘probablility without replacement.’ You are .50 to draw exactly one land in the first two draws, and .68 to draw at least one.

  26. Gerry, I would like you to read my response to your article where you talk about me not scooping to Oli in r15. Maybe it would help your understanding.

  27. Learning basic statistics will definitely improve your ability to play any game involving probability. This clearly includes magic but is far more broad of a skill….

    You can approach this question from either direction.

    I will use the notation P(X) to denote the Probability of X. I will use the notation !X to indicate X not happening.

    First, the obvious:

    P(X) = 1 – P(!X)

    Written another way:

    P(X) + P(!X) = 1

    Either X or !X must happen, so their probability must sum to 1.

    Next, conditional probability:

    P(X and Y) = P(X) * P(Y)

    P(X or Y) = P(X and Y) + P(X and !Y) + P(!X and Y)

    this can be simplified to:

    P(X or Y) = P(X) + P(!X and Y)

    this can also be expressed:

    P(X or Y) = 1 – P(!x and !y) = 1 – P(!X) * P(!Y)

    Using these basic formulas it is very easy to compute the probability of drawing at least 1 land over the first two turns…

    L1 = land on turn 1
    L2 = land on turn 2 ASSUMING no land turn 1

    P(X) = P(L1) + P(!L1 and L2)

    P(L1) = # lands / # cards = 23/53 = 43.4%
    P(!L1) = 1 – P(L1) = 56.6%

    P(L2) = # lands / # cards = 23/52 = 44.2%
    P(!L2) = 1 – P(L2) = 55.8%

    P(X) = 43.4% + (56.6% * 44.2%) = 68.4%

    OR

    P(X) = 1 – P(!L1 and !L2) = 1 – (56.6% * 55.8%) = 68.4%

  28. @ Kevin Boddy, Tmaggio, and especially Devin (who despite being a math major, thinks it more likely to draw exactly one land than at least one land): please come play against me!

  29. Math and realism are 2 of my favorite subjects on this planet. Thus this article makes me happy. Well done.

  30. This is a hugely important subject, and magic and in regards to everything dealing with probability (aka.. just about everything).

    There have been a number of articles about this in the past and this one adds to the discussion. It doesn’t really deal with the “why” of the “grand problem” though, and that is that is:

    ***It is very difficult for humans to avoid being results oriented.

    It is how we are hard wired. Our brains are trained to make deductions from single events and very small sample sizes (ie. ouch! red burner = f&*#ingHOT!) Most of us are not kidnapped the first time we travel abroad – but if you are you brain is trained to “learn” NOT to travel abroad.

    We deal in relative absolutes very well – unfortunately* we are trained to turn “middling probabilities” (ie 25%-75% … certainly 35%-65%) into absolutes. This manifests itself regularly in Magic, I am always amused at how any matchup that approaches 65% positive becomes classified as “unlosable.”

    * Well, evolutionarily, speaking in terms of carry the species forward, this is almost certainly a long term advantage – but it is a disadvantage when it comes to dealing with “middling” probabilities.

    Well, this post is turning into an article, but the point is going from a results oriented mind set to a process oriented mindset is not “natural”. You have to learn to objectively look at situations and parse out their components. This is a skill that must be practiced in order to be easily applied because it is against our nature, and magic is a great way to hone it.

  31. Since we’re all mentioning mistakes in mathematical comments…

    If you are in a 100 person tournament, and there is one faerie deck, your chance of playing that faerie deck in round one is 1/99 (since you are apart of the 100 players and you can’t play yourself). So, your chances of playing some other deck is 98/99 instead of what you said, “as 99 out of a hundered, you won’t play against faeries.”

    Other than that, enjoyed the article!

  32. I really liked this article. espcialy because the only fight i’ve ever had with my best friend was over the subject lol.

    in a risk game, i had 12 men defending and he had 8 and he decided to attack me… he won the battle, broke me, andi ended up being eliminated a few turns later, and i said “i can’t believe i lost from such a bad play” and he thought i meant one of my own plays, i said it was his bad play, and he said “how could it be bad if i won that battle?”… i tell ya, i was furious!!!

    lol, anyway, great article.

  33. I don’t exactly understand how my math is flawed, to be honest with you. If you pull two random cards out of the deck, it would be 23/53 + 23/52 to draw a land, which is 87%. Although I realize that I am wrong because if you were to take another card out, then you would have over 100% according to my logic, which obviously doesn’t make sense, but I fail to see how my first statement is wrong.

    @Mark Conkle: What does that even mean?

    Sorry to go way off-topic, Gerry. I enjoyed the article a lot, as I know that I’ve tossed the whole “results-oriented” statement around quite a bit. It was very interesting to read about stuff like Shadowmoor limited, as I remember forcing that archetype quite a few times to success. And about the draft part, where you mention that “all the black cards are coming so I should have been in black,” that’s literally all I hear at local drafts after the draft is over. Everybody just talks about how cards were going super late so they should have been in that color if they were a “good” drafter, and I now realize why I found that very annoying. The P2P7 Tendrils isn’t getting your UW deck anywhere, so why worry about it? Great article.

  34. @Kevin Boddy

    That isn’t how probability works. You can’t just add them.

    Please see my post above for a walk through of the correct math.

  35. The reason your way doesn’t work Kevin is that when you are looking at whether two events happen simultaneously you multiply probabilities (assuming they are independent) not add them.

    Why do we multiply instead of add? Other then the trite “thats how we make it work”, it basically comes down to how you define probabilities formally.

  36. Quick note on the math discussion: yes, it is 68%, and yes, it’s still right to mulligan. No, Kevin, you don’t have 130% chance of having a land within 3 cards =P. Assume the matchup is a 50-50 matchup, but you have a good hand. Even the depreciation of a 6-carder should only reduce your chance to win to about 45%. Even with an optimistic stance on winning of 60% given 2nd land, your chance to win keeping is still only 68% * 60% = 40.8%. Toss that chaff.

    Gerry: Great article! I really appreciate the tone and the candor in this. Glad to have read it.

  37. Pretty good stuff…however:

    “Winning tournaments with bad decks doesn't mean that your deck is good, or that you were right. ”

    Actually…it does. You can’t take the idea of not being results oriented so far as to say “This deck has won 10 tournaments…but it sucks.” really? So what constitutes a good deck? because the ‘pros’ say it’s good? That’s stupid and elitist. Cascade Swans is a great example of a ‘bad’ deck that stole a tourney. Those who played it were correct to do so, even though it never did anything else anywhere.

    While being results oriented can be bad, the goal of going to a tourney for most people IS the result: winning. So if I pick a ‘bad’ deck but a deck positioned well for that tourney and win, then I made the correct choice, barring ridiculous luck. (However, in a sample size as big as a PTQ or Worlds, saying pure luck won the tourney for me is pretty sketchy).

    But aside from that, I tend to agree with most of this article.

  38. There is a flaw in this article, though it took me a few re-reads to figure out exactly what it was. The flaw is that neither games nor tournaments happen in a vacuum. There is a lot of information going on with each play, a lot of which you do not have. Because of this, what may have been a bad play in a similar situation may in fact be the right play in this one.

    I’m coming from the opinion that success tends to add a lot of credence to a particular decision. A lot of our “gut instinct” is based on information our subconscious has picked up and trying to lead us in the right direction; we should not be ignoring it just because the numbers tell us something different.

    That being said, it is also important to understand how and why the game ended the way it did; it provides more information for next time.

  39. if Patric Chapin mulligan he will look the card of the top. He says that helps him on the next mulligan choices.

  40. Lol at some of the math in these comments. How would you be less likely to draw one OR two lands than to draw exactly one land? Maybe that’s not what you meant or maybe I misunderstood.

    Anyway, explicitly, this is NOT what results oriented means anywhere else. Maybe it can be adapted in magic to mean this, but if you are being interviewed and are asked to demonstrate results orientation, pleasepleaseplease don’t think that this is what it means.

  41. ” Assume the matchup is a 50-50 matchup, but you have a good hand. Even the depreciation of a 6-carder should only reduce your chance to win to about 45%. ”

    Where did the 45% number come from? I feel like a mulligan puts me much further behind than that, particularly in a mirror.

  42. “You can't take the idea of not being results oriented so far as to say "This deck has won 10 tournaments"¦but it sucks." really? So what constitutes a good deck? because the ‘pros' say it's good? That's stupid and elitist. Cascade Swans is a great example of a ‘bad' deck that stole a tourney. Those who played it were correct to do so, even though it never did anything else anywhere.”

    Here is another crux of misunderstanding in this argument. Being “results oriented” is bad when it excludes other sources of data – particularly the much more telling “process” data.

    Example: In baseball these days we can often tell if a expected production has gone down or if a player is simply getting unlucky by looking at core stats like how often they swing at pitches outside of the strike zone and line drive percent. If these numbers (among others) have stayed constant then we can surmise that a player is simply getting unlucky. Whereas simply looking at a classic results oriented stats (like average) is less impressive (OMG he is hitting .600 against sinkerball, left handers, at night on grass!)

    On the other hand, if these numbers have moved in a negative direction then we might surmise the a player is hurt or cannot perform at the same level that they have in the past (assuming the sample size is large enough).

    To bring this back to magic. If a deck “wins 10 tounaments” it very easily could be “bad.” Take your example of swans. What if swans had broken right before states, but it hadn’t been proven. It easily could have won many state tournaments over the course of a week before the sideboard hate and knowledge of the deck relegated it to “bad” status. This is why the *process* of how results were achieved is ALWAYS more important than simply looking at the results.

    “if Patric Chapin mulligan he will look the card of the top. He says that helps him on the next mulligan choices.”

    He may feel that that is true but knowledge of the you would have drawn does not help predeict what you will draw in the future, unless you are picking up on subtle differences in the card sleeve or something – which of course would be unconscious, because if you were doing it consciously it would be cheating 😀

  43. Your logic, wrt being results oriented is wrong. Results are the key to success.

    “Being "results oriented" means that you focus on the end result and not necessarily the in-between it took to get there.”

    Suppose you get into a car accident while driving, you can either think, “well driving car = accident” i.e. be results oriented and never drive again.

    Or you can not be results oriented, look in-between, and come up with the result “driving a car and not applying the break in time = accident.” Then you correctly apply the break whenever you drive.. but you just looked at the result, so you are result oriented? Sounds like a contradiction!

    You probably want to say something along the lines of, “Being results oriented is fantastic! Just make sure that the action that gave you the correct result is the action that will give you the correct result most often.”

  44. Your logic, wrt being results oriented is wrong. Results are the key to success.

    “Being "results oriented" means that you focus on the end result and not necessarily the in-between it took to get there.”

    Suppose you get into a car accident while driving, you can either think, “well driving car = accident” i.e. be results oriented and never drive again.

    Or you can not be results oriented, look in-between, and come up with the result “driving a car and not applying the break in time = accident.” Then you correctly apply the break whenever you drive.. but you just looked at the result, so you are result oriented? Sounds like a contradiction!

    You probably want to say something along the lines of, “Being results oriented is fantastic! Just make sure that the action that gave you the correct result is the action that will give you the correct result most often.”
    Oops, should have mentioned great post! Waiting on the next one!

  45. Great Article!

    This gives a great argument as to why Phyrexian Dreadnaught is not good in Legacy despite its success.

  46. @Pete – I’m pretty sure Swans “never did anything anywhere else” because seismic assault rotated out, not because it was a bad deck.

    Oh, and nice article.

  47. When I mulligan I look at the top cards of my library and say something about how I would have destroyed my opponent if I kept. Then when I draw my six I tell them these 6 are better. I do this to put my opponent on tilt and because I’m a jedi – and because I’m stupid and don’t know math.

  48. @ Robin: “To bring this back to magic. If a deck "wins 10 tounaments" it very easily could be "bad." Take your example of swans. What if swans had broken right before states, but it hadn't been proven. It easily could have won many state tournaments over the course of a week before the sideboard hate and knowledge of the deck relegated it to "bad" status. This is why the *process* of how results were achieved is ALWAYS more important than simply looking at the results.”

    I don’t deny that. But the flaw here is he is basically saying “results don’t matter.” Which is clearly flawed. To use another sports analogy, the year the St. Louis Rams won the SB with Warner at QB, NOBODY had them as being a serious contender for the playoffs, let alone win it all. Yet nobody would say they were a bad team. Why? The results. I was merely pointing out the dangers of being completely non-results oriented. If a deck loses 50 times out of 50, chances are it’s bad. That’s not being results oriented in a bad way.

    @Nathaniel: “I'm pretty sure Swans "never did anything anywhere else" because seismic assault rotated out, not because it was a bad deck.”

    I’m pretty sure it had 2 other GPs before Assault rotated and didn’t do squat at either. I’m pretty sure there were probably SCG 5ks and Channelfireball tourneys in there that it did nothing in either. The deck isn’t good and is easily hated out. It won because if nobody knows how to stop it, it can win. But if simply knowing a decklist means you can easily beat it with nearly any deck in the format…that’s a bad deck.

  49. “I don't deny that. But the flaw here is he is basically saying "results don't matter." Which is clearly flawed. To use another sports analogy, the year the St. Louis Rams won the SB with Warner at QB, NOBODY had them as being a serious contender for the playoffs, let alone win it all. Yet nobody would say they were a bad team. Why? The results. I was merely pointing out the dangers of being completely non-results oriented. If a deck loses 50 times out of 50, chances are it's bad. That's not being results oriented in a bad way.”

    But the results DON’T matter. The Rams won because they had and amazing offense that exploited holes in the pass defenses that were currently used in the NFL, and had a surprisingly efficient defense as well – NOT because they went won a ton of games. No one is saying that results are not what we strive for, but looking at results without the process can lead to terrible decision which is why results oriented analysis is a trap. Sometimes you have no choice – all you have are results (this dove tails nicely with Jon Loucks call for info this week) but looking at a process always is more productive than simply looking at results.

    If a deck lost 50 games in a row (hugely unlikely for even a tier 4 deck) you might be turned off. But what if it legitimately had a hugely unlucky streak of mulligians (almost the only way a loss streak of 50 could occur to anything resembling a cohesive deck) and it was play testing against it’s worst matchup (perhaps it is 80-20 vs. the rest of the field but a 20-80 dog in this matchup) in order to get a feeling for what options it could possibly have – surely that process should be the crucial part of making the decision about whether the deck is “bad” or not.

  50. I know it’s a bit late and that most of the people who had something to say about this article have already said it and moved on, but imma throw my thoughts out there anyway.

    I think that the biggest hurdle of not being “results oriented” is the terminology. The thesis of GerryT’s argument, as I understand it, is that to simply support a deck if it won a tournament is a mistake because one will not learn why it won and will likely not be able to repeat that success in the long term.

    However, one needs to have some kind of results to base an opinion on, like those of a playtesting session. Without any kind of testing or practical information it is significantly more difficult to pick a deck that will perform well.

    My suggestion would be to rename this process “event oriented” instead of just “results oriented.” We need SOME kind of results or information to base our opinions and back up our hypotheses. However, modifying the terminology to “event based” or some such could more effectively communicate to the laymen that the conclusions shouldn’t be based on a singular event because that might be an anomoly without adding layers of confusion.

    Hope that makes sense. If not and anyone is still reading this/cares, lemme know and I’ll be glad to elaborate.

    -Andy

  51. The whole point of that article was that you don’t need results to hypothesis anything. If you play a matchup like Faeries vs 5cc, the results shouldn’t matter, but the way the matchup plays out should. Figure out what cards are important and how to be prepared for them.

    (GerryT posting on LSV’s account)

  52. “My suggestion would be to rename this process "event oriented" instead of just "results oriented." We need SOME kind of results or information to base our opinions and back up our hypotheses. However, modifying the terminology to "event based" or some such could more effectively communicate to the laymen that the conclusions shouldn't be based on a singular event because that might be an anomoly without adding layers of confusion.

    Hope that makes sense. If not and anyone is still reading this/cares, lemme know and I'll be glad to elaborate. ”

    What we need to do is look at *process* (I think this is what you are saying) – the “why” and “how” a match played out internally is much more important to discovering the quality of a deck or the intricacies of a matchup than which deck won, or even what a given decks winning percentage is.

    Process, process, process… ignore the “results” whenever possible. Results only give you a second hand look at the REAL information, and are only useful in the absence of adequate information (or ways to track) the process.

    As Gerry says: “Figure out what cards are important and how to be prepared for them.”

  53. This won’t really be seen, but I don’t particularly mind it. . . Since reading the 3 or 4 articles on this site about “the aura,” results orientation, and things of that nature I’ve had this itch to explain this idea, nowhere better to put it, it seems, so here you go GerryT.

    I’m a math major, I want to be a math teacher, mathematics means a lot to me, its my way of solving almost every problem I come across. I’m not religious in the slightest, and only have some slight feelings towards any kind of spirituality. All that being said, I’m a tarot reader. Contradictory? Yes. Unrealistic? Maybe not.

    There is an idea amongst logical people who do tarot reading, and it is bound within the concept that the subconscious is a lot more powerful than the active mind. Its why we who don’t believe in fortune telling, per se, still do tarot reading, because its not fortune telling for us, its an insight into our inner minds.

    How it works is this: You have a tarot deck, you have studied and practiced to be able to give a traditional reading, you know every card in your deck and know the meanings of each combination of cards. You perform a ritual, in which you focus on the cards, and think about them as you shuffle them, the purpose of the ritual is to let go of any fleeting thoughts, and focus only on the cards, and the subject of the reading you’re about to administer.

    The theory is that your subconscious takes over during this ritual, you are no longer just shuffling a pile of cards, but your mind is putting the cards in order without you actually knowing it. Are you curious as to which path in your life you should follow, because you don’t know for sure whats best for you? Guess what, your inner mind already knows, because it takes care of crap like that for you, so while you shuffle, your mind is making tiny adjustments to your hands as you shuffle cards, trying to put the cards that will give you the answer on top of your deck, in the order to send you the right message.

    It is not terribly unlike playing Magic, you know the cards in your deck, and most people shuffle their cards the same way every time, sure, its designed for randomization, but the same theory can be applied, that as you shuffle your cards, you put a perfect curve on the deck, you set up your cascades, you hit your land drops. Sure your opponent comes by and mucks the whole thing up, but the cards are still in an order for which to be most desirable for you.

    If your opponent sits there and shoves the deck on top of itself a good 5-10 times, no amount of focus will set your deck up perfectly, but focus good well enough and your opponent just cuts it a couple of times, and you might find yourself not having to mulligan. . .

    Long story short, this mathematician believes in some kind of an ‘aura’ and its worth a shot to throw some Zen at your deck from time to time.

  54. You’re largely right about everything you’ve written here, and you’ve done a good job articulating the basics. Magic players generally have an extraordinarily poor grasp of game theory and probability–to the extent that I often feel like I’m beating my head against a wall in discussions. Thanks for helping to correct that, GT.

  55. BTW the term “results oriented” is a direct port from poker and in particular from the community on the TwoPlusTwo forums, where it never provokes confusion. If you’re too confused by that to understand the point, ditch the term, but it’s pretty retarded nit to pick.

Comments are closed.

Scroll to Top