Quick PSA: sometimes it's OK to be greedy

[DirkGently]

Saw a meme that reminded me of a fairly stupid conversation I had last time I played EDH.

4-player game, still in fairly early turns. I'm doing not much (playing Zirilan with too little ramp...I really need to upgrade that deck one of these days), two of my opponents are playing Ayli and Golos, with Golos getting kinda scary with their commander in play that no one seems to be able to remove and activating it every turn. The fourth player is playing Kadena with what looks to be a fully precon deck, or at least barely upgraded.

Kadena casts tempt with discovery.

The Ayli player says "don't take the land, everyone knows you're never supposed to take the land, it just helps him." Despite the fact that (1) the dude was playing a precon, and (2) he was pretty far from the lead, and (3) we could really use the help to catch up with Golos. Despite me pointing this out, he declines. Me and Golos both take the land, because duh. The Ayli player ended up way behind and the precon player got to stand a fighting chance. He still thought it was the right play to not take the land, even when the Kadena player was essentially fetching basics. SMH...

It reminded my in general of how annoying it is when people (who usually think they're super good at the game) overapply strategy rules without thinking critically. "Always pay for rhystic study" is another one I see a lot that always drives me nuts. For one thing, in a 4-player game a single opponent drawing a card is much less bad than in 1v1. For a second thing, depending on board state they may not even currently be a threat to you at all. For a third thing, it's going to depend a lot on what kind of deck they're playing, how strong it is, how badly you need the mana etc etc to determine what's the best play for you.

The one thing that's certainly true, though, is that you paying extra mana to deny another player a draw is definitely the optimal result for the third player telling you how it's DEFINITELY ALWAYS correct to pay for study and you're making a BIG MISTAKE because EVERYONE KNOWS that you ALWAYS PAY.

Anyway, that's my PSA. Don't just keep repeating "rules" without thinking critically about them, and sometimes the greedy choice is the right choice. And don't let idiots bully you with bad advice that "everyone knows".

[lyonhaert]

I wouldn't even say that's being greedy. Just that the usual reason to refuse Tempt (the caster's benefit outweighing anybody else's benefit) was not an issue, so why not get the free land?

But yeah, some people stick rigidly to rules/techniques because they haven't internalized the underlying reason.

[DirkGently]

But yeah, some people stick rigidly to rules/techniques because they haven't internalized the underlying reason.

One I see in all formats, especially limited, is people that think "instant speed removal should always be played at the last possible second". So they sit on their removal spell, and let the opponent untap with their absolutely-must-kill-bomb, and then when they try to remove the creature on attack, their opponent has a counterspell that they couldn't have paid for if they'd just used the removal on their own turn.

[pokken]

I think you could probably call these things heuristics. And they really do have value because it lets you reduce the mental load of playing the game and offload it to pre-defined algorithms essentially, but people need to know that learning the rules is just a step on the way to playing well.

What's that old saw, something like you ought to know the rules so you can know when to break them? Sounds like Twain or something in my head.

It is interesting how susceptible magic players are to turning general guidelines into rules. I think magic players are prone to pretty aggressive pattern-finding that tends to support this kind of behavior (such and such happened twice, therefore it is a principle, etc.).

[Maluko]

Pokken made a very good point about this subject. I share the same philosophies with the player you described (never be tempted, never pay for Rhystic Study), but I only practice them as a default. One of the qualities needed to become a pro player, whether you play in highly competitive environments or at a kitchen table, is to know when to break your defaults. Pro players know how to properly access the situation at hand and whether they should use their default response or not. In the situation you described, I think the Ayli player made a poor choice. And if he/she is failing to realize it even after the game ended, that's because he/she hasn't gotten that skill yet.

[toctheyounger]

This was a good read. Context is everything, and it's why these hard and fast rules become softer and slower when you apply them outside of a vacuum. In this case, what's it to ya if the precon player fetches a couple more basics, or a couple more lands that enter tapped? No big deal. If you were up against a tuned deck you can expect to see Cradle or similar, yeah, different scenario there.

This feels like a teaching moment in terms of how as a player you assess the table you're playing at and what you can expect to see from a deck. A lot of that is unknown, especially if it's a random matchup, but every commander has archetypes and staples you could feasibly expect to see commonly. Then there's cues from the plays you see from the table around you. You might only get a couple of turns to take a punt at what a person's angle might be but that often gives you enough information to know whether you should give in to temptation or pay the Rhystic cost.

[hyalopterouslemur]

But yeah, some people stick rigidly to rules/techniques because they haven't internalized the underlying reason.

One I see in all formats, especially limited, is people that think "instant speed removal should always be played at the last possible second". So they sit on their removal spell, and let the opponent untap with their absolutely-must-kill-bomb, and then when they try to remove the creature on attack, their opponent has a counterspell that they couldn't have paid for if they'd just used the removal on their own turn.

Yeah, I've seen that one a lot. Another is "play all lands immediately". This is actually why MLD is seen as so unfun, I believe: You need to have either alternate sources of mana or a hand full of land to rebuild. Or just, you know, play MLD when you're about to win. (That's a whole other example, people who play MLD when they're about to lose.)

Another thing players do is add cards that don't really work in their deck. Aura Shards and Mycosynth Lattice is a "nice" thing to do with the lattice. (As opposed to, I don't know, Null Rod.) And token decks with green and white tend to play it as a result. You know who never should? Rith, the Awakener

[WolfWhoWanders]

Good points here, appreciate the read. I don't really like statements like "always" and "never". Seems like there are usually exceptions to a rule.

[DirkGently]

never pay for Rhystic Study

Whoa, hot take 😆

Tbh I think it might be more often correct to not pay (when it slows you down, especially when developing). If you're sandbagging your 3-mana rock and playing nothing t3 to avoid letting someone draw a card, I'd say it's a mistake. If you've got good alternative plays then do that, but tempo is usually more important than CA in the early turns.

[Maluko]

never pay for Rhystic Study

Whoa, hot take 😆

Tbh I think it might be more often correct to not pay (when it slows you down, especially when developing). If you're sandbagging your 3-mana rock and playing nothing t3 to avoid letting someone draw a card, I'd say it's a mistake. If you've got good alternative plays then do that, but tempo is usually more important than CA in the early turns.

As explained previously, it depends on the situation. But in this case, it also depends on the typical profile of players that run Rhystic Study. In my experience, decks that run that card are usually borderline – not super competitive but also not enjoyable to play against, usually running lots of staples and goodstuff. For one, I'd rather not give those decks more advantage than they already have. And two, sometimes it pays off to hold your stuff for another turn: you don't attract much attention, and you avoid overcomitting to boardwipes.

[DirkGently]

As explained previously, it depends on the situation. But in this case, it also depends on the typical profile of players that run Rhystic Study. In my experience, decks that run that card are usually borderline – not super competitive but also not enjoyable to play against, usually running lots of staples and goodstuff. For one, I'd rather not give those decks more advantage than they already have. And two, sometimes it pays off to hold your stuff for another turn: you don't attract much attention, and you avoid overcomitting to boardwipes.

Yes, clearly when I'm talking about playing a mana rock on T3 I'm talking about overcommitting into a board wipe...

Btw you do realize you said NEVER pay for rhystic study, right? lol.

[tstorm823]

I think you could probably call these things heuristics. And they really do have value because it lets you reduce the mental load of playing the game and offload it to pre-defined algorithms essentially, but people need to know that learning the rules is just a step on the way to playing well.

If you'd like a slightly more specific term from a different game entirely, go has the word "jōseki" for this concept.

[Mookie]

Going to one of my favorite sayings - All models are wrong. Some are useful.

Re: Tempt with Discovery - this is a very game theory-ish card. If we treat all actions players take as a zero-sum game, then any one player being tempted could be described as '+2 to the tempted player and the caster, -2 to the other two players'. If everyone is tempted, that is a net +6 to the caster, and -2 to the other players. While if no one is tempted, it's net +0 to everyone. Which means the other three players shouldn't be tempted....

And to be honest, that's sort of the whole point of the prisoner's dilemma - it's optimal for every participant to defect, even though it results in net loss. But just because it's globally suboptimal doesn't mean you shouldn't be tempted - it just means you need to convince the other players to not be tempted, while taking the offer yourself.

The other situation is that it's really easy for the weightings to be different - if the tempter is far behind, them gaining a few points doesn't matter nearly as much. Alternatively, if they're fetching basics while you're fetching Gaea's Cradle, you can certainly come out ahead. I regularly take the offer if I'm down on lands or playing colors that have difficulty ramping, or if there is a particular utility land I want to grab. On the other hand, if I have a bunch of mana and only have fixing lands to fetch while the caster has stronger things to grab, I'll be less tempted.



Re: Rhystic study - if it's out, you effectively have three options: cast spells and pay the tax, cast spells and don't pay the tax, or don't cast spells at all. Not casting spells is usually bad, but worth consideration if you expect the study to go away or have mana sinks to use instead. But as for paying the tax vs not paying the tax, you're essentially asking the question of whether a short-term benefit (extra mana) outweighs a long-term benefit (your opponent having fewer cards). Which, in turn, depends on how long you expect the game to go... and if you expect it to go for a while, you need to ask the question of whether you can leverage your own short-term mana advantage into a long-term benefit.

I'd certainly consider not paying the tax for a spell like Rampant Growth or Phyrexian Arena that is going to give me a significant advantage over time. Alternatively, I won't pay if I expect the game to end before the extra cards matter (such as if I really need to cast a board wipe to not die, or if I'm comboing off). On the other hand, if you're casting Opt or another transient spell, you should really either pay the tax or rethink your action. It may still be right to not pay the tax, but the question is harder to answer.

[bobthefunny]

Going to one of my favorite sayings - All models are wrong. Some are useful.

Re: Tempt with Discovery - this is a very game theory-ish card. If we treat all actions players take as a zero-sum game, then any one player being tempted could be described as '+2 to the tempted player and the caster, -2 to the other two players'. If everyone is tempted, that is a net +6 to the caster, and -2 to the other players. While if no one is tempted, it's net +0 to everyone. Which means the other three players shouldn't be tempted....

And to be honest, that's sort of the whole point of the prisoner's dilemma - it's optimal for every participant to defect, even though it results in net loss. But just because it's globally suboptimal doesn't mean you shouldn't be tempted - it just means you need to convince the other players to not be tempted, while taking the offer yourself.

...Actually, reading your answer like that, it sounds like you've just argued that it's actually correct for exactly one player to take the temptation. If everyone takes the land, then everyone is behind other than the caster. But if the first two players decline, the fourth player now has a chance to be "+2 to the tempted player and the caster, -2 to the other two players." - thus, they'd be even with the caster, and put the other two players on a deficit.

Unfortunately, the first player can't really take that greedy decision, because if they do, then the other two players won't take the deficit (better to be behind on one player, than multiple), Which means that if player 1 tries to take the greed, they end up putting themselves in a worse position. Same with the second, but to a lesser degree, but the last player can theoretically take the temptation and pull one over on the other two players.

Fascinating.

[schweinefett]

I think you could probably call these things heuristics. And they really do have value because it lets you reduce the mental load of playing the game and offload it to pre-defined algorithms essentially, but people need to know that learning the rules is just a step on the way to playing well.

What's that old saw, something like you ought to know the rules so you can know when to break them? Sounds like Twain or something in my head.

It is interesting how susceptible magic players are to turning general guidelines into rules. I think magic players are prone to pretty aggressive pattern-finding that tends to support this kind of behavior (such and such happened twice, therefore it is a principle, etc.).

yea, actually, that's what sorta makes great players so great. They understand the heuristics, but also understand situations where that isn't applicable. It's like playing around red elemental blast/pyroblast or double blue open mana in typical 1v1 games. there are times when it's right to jam your cards, and other times when it's not.

I gotta say though, these sort of rules are a lot more applicable to 1v1 than EDH. like playing turn 1 petal into dark ritual to 'fake' being land-less and drawing a counterspell.

How many of these heuristics are there really in EDH? or like.. 'rules' that people use? I know one we usually do is X/0 piles, but on the 0 pile, offer a can of beer.

[DirkGently]

Going to one of my favorite sayings - All models are wrong. Some are useful.

Re: Tempt with Discovery - this is a very game theory-ish card. If we treat all actions players take as a zero-sum game, then any one player being tempted could be described as '+2 to the tempted player and the caster, -2 to the other two players'. If everyone is tempted, that is a net +6 to the caster, and -2 to the other players. While if no one is tempted, it's net +0 to everyone. Which means the other three players shouldn't be tempted....

And to be honest, that's sort of the whole point of the prisoner's dilemma - it's optimal for every participant to defect, even though it results in net loss. But just because it's globally suboptimal doesn't mean you shouldn't be tempted - it just means you need to convince the other players to not be tempted, while taking the offer yourself.

The other situation is that it's really easy for the weightings to be different - if the tempter is far behind, them gaining a few points doesn't matter nearly as much. Alternatively, if they're fetching basics while you're fetching Gaea's Cradle, you can certainly come out ahead. I regularly take the offer if I'm down on lands or playing colors that have difficulty ramping, or if there is a particular utility land I want to grab. On the other hand, if I have a bunch of mana and only have fixing lands to fetch while the caster has stronger things to grab, I'll be less tempted.


Re: Rhystic study - if it's out, you effectively have three options: cast spells and pay the tax, cast spells and don't pay the tax, or don't cast spells at all. Not casting spells is usually bad, but worth consideration if you expect the study to go away or have mana sinks to use instead. But as for paying the tax vs not paying the tax, you're essentially asking the question of whether a short-term benefit (extra mana) outweighs a long-term benefit (your opponent having fewer cards). Which, in turn, depends on how long you expect the game to go... and if you expect it to go for a while, you need to ask the question of whether you can leverage your own short-term mana advantage into a long-term benefit.

I'd certainly consider not paying the tax for a spell like Rampant Growth or Phyrexian Arena that is going to give me a significant advantage over time. Alternatively, I won't pay if I expect the game to end before the extra cards matter (such as if I really need to cast a board wipe to not die, or if I'm comboing off). On the other hand, if you're casting Opt or another transient spell, you should really either pay the tax or rethink your action. It may still be right to not pay the tax, but the question is harder to answer.

I like these sorts of semi-mathematical dissections of cards. I'll give my two cents.

As far as tempt goes, from each individual player's position, it's usually correct to take the bait. The only time it wouldn't be (in a 4-player game) is if you knew that the tempt player was going to get a LOT more mileage out of it, probably either because they were going to get way better lands than you, or because they are in a better position to take advantage of the mana. The way to play against tempt cards - assuming that the table agrees that they strongly don't want the tempt player to get the lands (I bring this up because in the OP that was definitely not the case) - is for everyone to forgo the tempt, even though it would be beneficial to each player to break the deal. It really doesn't matter what the order is, either way anyone who breaks the deal you get an advantage, but if no one does you all benefit (except the tempt player, obviously). Classic prisoner's dilemma.

When, as in the OP, the tempt player's lands aren't much of an issue, then it's the same situation minus any serious incentive to band together. Sure, if you all pass then it's fine for everyone, but there's no reason to trust the other people to follow through so you may as well do it. And in the situation where the tempt player can actually help against an archenemy, then you should definitely, definitely take it.


As far as rhystic study, I think your analysis is both too complex and too simple. There aren't really three choices, because you're not going to not pay the tax just for S&G, so it's really just play or don't play, and if you play you're going to pay tax if able. But then that's applied to all the cards in your hand, so you're going to play some number of cards, and pay as much tax as you can afford (barring some unusual situation where you really want the rhystic study player to draw more, while you're both temporary allies let's say). So you'll usually have a choice between a very mana-optimal turn, that lets your opponent draw the most cards, a very inefficient turn that denies draw, and some number of choices in between. And it's all just a matter of how much better you think the mana optimization is, and whether it's worth letting the player draw that many cards.

It's worth keeping in mind - let's say, as a general heuristic, that one opponents card is worth roughly 1/3 of one of your cards. So if each player draws a card, 1/3 * 3 = 1, so everything is balanced as you'd expect (obviously this is just in general, in practice the value of a "card" is pretty fluid). That means, an opponent drawing a card is as bad for you as losing 1/3 of a card. So if that rhystic study player draws 3 cards off you while you cast 3 spells, you've "gained" 3 mana, and lost 1 card.

Hmmm......3 mana for 1 card....where have I seen that before...

Maybe it's not such a bad deal after all to let the rhystic study player draw a bit, eh?

[hyalopterouslemur]

never pay for Rhystic Study

Whoa, hot take 😆

Tbh I think it might be more often correct to not pay (when it slows you down, especially when developing). If you're sandbagging your 3-mana rock and playing nothing t3 to avoid letting someone draw a card, I'd say it's a mistake. If you've got good alternative plays then do that, but tempo is usually more important than CA in the early turns.

With Rhystic Study and Smothering Tithe, you'll often see them paired with Winter Orb and friends. (Yeah, it's evil.) In such a case, paying 1 sets you back a lot more than an extra card on your opponent's side of the table.

[Hipster Genocide]

Always *clap* pay *clap* for *clap* Rhystic *clap* Study. I can't stand when someone at the table is like, "pfff, whatever, I don't care, draw your card." Then a few turns later..."whoa, how did he steamroll us like that?"

Well, the fact that you just had to play out all your stuff on curve and gave him 20 cards might have had something to do with it. I mean, even if you play out everything in your hand, when you give the Rhystic Study player a bunch of cards he will be able to Wrath it all away anyhow.

So yes. Slow down, play your stuff one turn later, use politics by constantly pointing out that it's the Rhystic Study player causing these problems, and if we could just team up and take him out first, the problem will go away.

And if the guy pairs it with Winter Orb, that just means I won't play against him anymore.

[Maluko]

As explained previously, it depends on the situation. But in this case, it also depends on the typical profile of players that run Rhystic Study. In my experience, decks that run that card are usually borderline – not super competitive but also not enjoyable to play against, usually running lots of staples and goodstuff. For one, I'd rather not give those decks more advantage than they already have. And two, sometimes it pays off to hold your stuff for another turn: you don't attract much attention, and you avoid overcomitting to boardwipes.

Yes, clearly when I'm talking about playing a mana rock on T3 I'm talking about overcommitting into a board wipe...

Btw you do realize you said NEVER pay for rhystic study, right? lol.

I would like to hear how using hyperboles to distort my arguments and taking quotes out of context favors your point of view.

[DirkGently]

[mention]Hipster Genocide[/mention]

You *clap* are *clap* grossly *clap* oversimplifying.

For more details, read the thread.

Another thing worth pointing out about rhystic (especially when played in the early turns, where it's difficult to pay the tax without drastically slowing your play) is that draw has diminishing returns. Draw 10 is not twice as good as draw 5, especially if they don't have reliquary tower (or similar). Mid/late it becomes much easier and much more important to pay, but early it's rarely worth it imo. Nobody pays, they discard a bunch, their hand is pretty sculpted but that's usually much less bad than straight-up skipping turns because you think you NEED to pay the tax.

I would like to hear how using hyperboles to distort my arguments and taking quotes out of context favors your point of view.

I was specifically talking in the early turns. Holding stuff back on the early turns - especially mana ramp, which is what I mentioned - to avoid overcommitting doesn't really make sense. Mid/late game, sure I agree, you're more likely to have near-optimal turns that allow you to have extra mana to pay the tax, but early-game you probably can't afford to hamstring yourself by always paying the 1, and that's what I was talking about. You quoting my post implies that you're responding/disagreeing with it, but your point wasn't really relevant to what I was saying.

As far as the "never pay for rhystic study"...I'm pretty sure you meant to say "ALWAYS pay for rhystic study", way back in your first post. Just a simple typo but I thought it was pretty funny so I pointed it out. Doesn't actually have any import in terms of the conversation, I just thought it was an amusing (if accidental) take.

I'm actually not sure what you think I'm hyperbolizing(?) or distorting about your argument.

[3drinks]

I mean, I'll always attack the U player first. That's just on GP. Is it correct every time? Surely, it's not. But as another point earlier in the thread about these predisposed play patterns, this is just another in the long list of them. Even though I recognize and admit here that it likely isn't always correct. I will still continue to repeat this behaviour over and over, continue to hold up Boil mana with Red Elemental Blast backup because that's "just what you do".

[hyalopterouslemur]

Always *clap* pay *clap* for *clap* Rhystic *clap* Study. I can't stand when someone at the table is like, "pfff, whatever, I don't care, draw your card." Then a few turns later..."whoa, how did he steamroll us like that?"

I gave an example of when you shouldn't. In the shadow of a Winter Orb, you really should reconsider, just because the mana is a more precious resource. (And until opp gets a Seedborn Muse or Clock of Omens or something, that card stays in their hand.)

It's like cards exist in the context of other cards and not in a vacuum.

[KarnDaddy]

it's just like everything else in this game, there is a rule of thumb, but it all depends. when i am teaching someone to play, i tell them straight up "I can teach you all the general rules, but you WILL come across a card or board state that breaks every single one of them. my playgroup has learned, i don't always pay for Study. usually do, but if i don't, they should worry

[Rumpy5897]

It's complicated. There are actually some decks that scale well into absurd draw totals, so saying drawing 10 is not twice as good as drawing 5 is still a simplification on some level. Still, the validity of rule of thumb holds true, as it lets you take them as a reasonable prior, processing game states easier and moving the game along. It's interesting to see those for different people - within my group, one guy skips paying for Rhystic 90+% of the time, while another one actively goes out of his way to pay Smothering Tithe to the point of kneecapping his own turns over it.

My system's pretty easy - only pay Smothering Tithe if there's absolutely nothing better that could be done with the mana (to the point where I count bluffing interaction here), but try to pay Rhystic Study if it doesn't completely sandbag the turn. Playing nothing and passing because the 1 wasn't around is a little extreme.

[DirkGently]

As far as the double draw goes, I guess it gets tricky since there are situations where the difference is winning and losing, so then it's hard to say what "double the effectiveness " really means. But early-game with non-cEDH 75%-ish decks, I'd argue the function of draw-> effectiveness is pretty asymptotic. Which is the main time study is so burdensome,

Generally I agree with your system, though I suspect I'd go less far out of my way than most to pay. For example, t3 if I've got worn powerstone or fellwar Stone, I'm definitely jamming powerstone. Tbh I default to not paying on turns 1-5 unless it has no impact on my plays, unless it's reeeaaaally minimal.