Playing more than 60 cards

[Cthuloo]

I agree with Gui.

@alderon: In your example, adding a removal spell will decrease the probability of seeing a Misstep in the fist 10 cards by 0.7%, but (assuming e.g. that you moved up from 3 to 4 removal spells) it has increased the probability of drawing a removal by almost 10%, which can be a huge boost if the fields turns out to be more aggro than combo. The point is: if you have perfect knowledge of the field and perfect practice with the deck, then obviously 60 cards is the best configuration. If you don't (which is always the case), going up to 61 can be a more conservative, yet acceptable move.

[Mr. Safety]

(BTW: Neither TES nor ANT run an absurd amount of tutors. Infernal Tutor pretty much sucks when you need a 1-of answer. Burning Wish does this job, but 4 cards out of 60 hardly qualifies as an absurd amount. Currently, there is only two "Tier 1 tutors" in the format that see play - Merchant Scroll and Enlightened Tutor. And in their respective decks, running 61 cards might just be fine.)

I was counting Ad Nauseum as a tutor, along with Infernal Tutor and Burning Wish. That totals up to a minimum of 6, probably closer to 7-8. I shouldn't have said 'tutor'...I should have said 'library manipulation' spells. What I'm saying is it has a lot of ways to find the cards it needs. I misworded it; many apologies. Taking into account Brainstorm and Ponder, I would say that totals up to 'absurd' in my opinion, lol.

EDIT: the TES list in the TES primer actually plays 17 tutors/library manipulation spells...I think that's an absurd amount, lol. Maybe the list has changed slightly, but I don't think by much. Someone with more familiarity would be able to answer that better than I.

[alderon666]

Good players tend to avoid high variance decks, so they capitalize on their play skill and outplay their opponents.

Dredge is a fine example of a high variance deck. The deck is extremely powerful and can even fight hate decently. But the deck has so many dead draws that the variance is ridiculous. You can literally draw several 7 card hands and a lot of the will be unplayable or slowed down to a crawl by a single MM or FoW. On the other hand, there's a decent amount of hands that can basicaly win turn 2 fighting though a Crypt .

I agree with Gui.

@alderon: In your example, adding a removal spell will decrease the probability of seeing a Misstep in the fist 10 cards by 0.7%, but (assuming e.g. that you moved up from 3 to 4 removal spells) it has increased the probability of drawing a removal by almost 10%, which can be a huge boost if the fields turns out to be more aggro than combo. The point is: if you have perfect knowledge of the field and perfect practice with the deck, then obviously 60 cards is the best configuration. If you don't (which is always the case), going up to 61 can be a more conservative, yet acceptable move.

Then the correct thing to do is to remove some card that's bad against aggro and replace it with the extra removal spell. That way you tune your deck to be slightly better against aggro without altering the chances to draw land, creatures, etc.

[.dk]

Then the correct thing to do is to remove some card that's bad against aggro and replace it with the extra removal spell. That way you tune your deck to be slightly better against aggro without altering the chances to draw land, creatures, etc.

I think you just agreed with him... Given perfect knowledge of the meta, then 60 cards is ideal. His point was that if you don't have perfect knowledge, running 61 may be an acceptable trade.

So maybe the lesson here is to get the best knowledge of the meta as possible, and stop worrying about how many cards you run?

[Gui]

Then the correct thing to do is to remove some card that's bad against aggro and replace it with the extra removal spell. That way you tune your deck to be slightly better against aggro without altering the chances to draw land, creatures, etc.

This is true, but only if you know exactly what to cut, and is 100% certain you won't miss that card. If you cut something good against control to add the removal, and face a lot of aggro, the change made sence. But if you do that and end up facing controls, you risked yourself more than I would have by playing a 61-cards deck, and yet my 61-cards deck is almost as suited as yours against aggro. It's impossible to predict exactly against what you will play.

We are not saying "always play 61-cards deck, they are awesome!", but rather saying it is not as impactant as some people judges it to be. And we are actually providing numbers that prove this by showing the difference is barely sensible. But if you have the chance to playtest endlessly against every deck and decide precisely what to cut, that's the best move.

[Koby]

I believe a prime example of the 61st card syndrome is demonstrated by old Landstill - a deck that typically wants 23.5 lands. 23 oftentimes feel too few, and 24 feels too many. Adding the 24th land as the 61st card was a way to balance out the mana/cost ratio in the deck while still providing an additional mana source.

I also remember Jamie Wakefield (of Secret Force fame) always started his deck contruction with 26 lands, 62 cards. Is this correct? probably not. But it does lend some evidence on building better mana bases that are otherwise tricky to get right in decks that require a certain number of lands by a certain turn.

I don't think the above model is fit for Legacy deck construction. Cantrips and fetchland also skew this old theory.

[Shimi]

The only reason to play 61 cards deck is:

A)It's some lucky number or anything stupid like that.

B)You are testing some silverbullets and would like to see what to cut to make you deck a 60 card deck.

C)You are running something like lands + engine + tutors + 1of tutor targets that REALLY can't be cut.(eg. lands decks).

[SpikeyMikey]

This has been discussed before by top-level pros, and i think their general consensus was that given the correct circumstances, it can be right (I remember one deck from GP Atlanta popping up) but a vast majority of the time, it is strictly worse to run 60 cards. There is also something to be said about running 61 if you don't know what to cut, because you don't want to cut the wrong card and be screwed.

Top level pros are generally the last people I want to get deck construction advice from. Very few of them have the first clue what they're doing. What it boils down to is that while we like to pretend that Magic functions like an art, it really is a science. It's just a far more complex version of poker. But in Poker, calculating odds is much easier. You can talk about playing to outs in Magic, but the math is too difficult to do in your head. In hold 'em, you know if you've got AK suited you're ~33.3% against KK. It's far more difficult to, on the fly, calculate your chance of drawing both a land and a removal spell in the next 3 draws in order to stay in the game.

I almost never play decks with more than 60 cards (although I regularly play 41 in limited and occasionally 42 or 43) but that doesn't mean that I feel that 60 is always the correct choice. Deckbuilding, in my opinion, is primarily about ratios. The Bant deck I ran for the first 6 months of this year ran 4 Swords to Plowshares and 1 Path to Exile (although for the last 2 months or so while the format has been slower due to Mental Misstep, Path has become Engineered Explosives). People think of the singleton Path as strange, but it's not a singleton, it's the 5th Swords to Plowshares. It's the difference between being able to remove a creature 50% of the time by turn 3 (average between play and draw) and 59% of the time. By the same principle, if I've got a deck where I want to both maximize my chance of drawing Tarmogoyfs and Knights of the Reliquary but I also want to run 19 blue spells to help ensure I have enough to pitch to Force of Will, it may be necessary for me to run 61 or 62 cards. Cutting a Tarmogoyf or Knight for a junk blue spell, even a filtering blue spell like Ponder or Preordain is not an option. It's going to significantly reduce the number of whatever it was I cut that I draw because when I draw the Ponder instead of Knight, I'm looking at a (very slightly less than)4/(15-18)ish chance of finding a Knight with it (depending on how many cards are left in my library). I could run out the hypergeometric distribution, but let's just say that 75% of the time, that Ponder won't find me Knight. So instead of a 50% chance of having that Knight I need, I now have a 40% chance. That's pretty significant, you're talking once every 4-5 matches or about twice in a major tournament. And that's just for the first Knight, the odds are exponentially worse when you talk about finding a second one (having 2 left vs. 3 left).

[alderon666]

I think you just agreed with him... Given perfect knowledge of the meta, then 60 cards is ideal. His point was that if you don't have perfect knowledge, running 61 may be an acceptable trade.

So maybe the lesson here is to get the best knowledge of the meta as possible, and stop worrying about how many cards you run?

61 cards is an excuse for adding some card you don't need because you don't know what to take out.

Because quoting yourself is cool.
If you don't the meta well you should just tune the deck to play more versatile cards instead of just adding more of X or Y.


Playing 61 cards because you need half a land could probably be achieve in some other way. Like playing more cantrips or even playing less fetchlands (to increase your chance of drawing land cards, LOL).

[Hanni]

This thread needs more Jack Elgin.

[TheInfamousBearAssassin]

Well that's always true.

I kind of feel like I hashed this before. While I concede that there's the possibility that 61 cards is the correct number for mathematical reasons (or even 62), I think the math behind that is going to be so complex that it is pretty much certain that you're wrong for thinking so. And certainly it's the wrong default.

61 is fine if you're not sure what to cut the day of a tournament. I would otherwise much rather run 63 cards than 61 as doing the former will legitimately enrage some people (like running 8th Edition basic lands), and that makes people make mistakes.

But short of trying to psych people out (which I would only recommend in a highly redundant deck), stick to 60.

[Kuma]

Playing more than 60 cards is wrong:

1: The cards in your deck can be ranked in order of overall power from 1 to 60. This is hard to do in practice, but we all know that some cards in our deck are better than others.

2: Running a 61st card decreases your chances of drawing the 60 cards more powerful than your worst card. It's only a small amount, but why willingly sacrifice any chance of drawing your best cards, especially when there's a card in the list that's weaker than all the other cards that you don't have to run? You need to figure out which card this is and cut it.

Therefore, playing more than 60 cards is wrong.

[Julian23]

@Kuma:

I disagree with your first assumption. Therefor I disagree with your conclusion.

[.dk]

@alderon666 yep, that's a fair point. if you can find more versatile cards to play in your 60 without sacrificing much, then that would be better. I think the only point here is that it isn't ALWAYS wrong to run more than 60. If you can't find a more versatile solution without sacrificing too much, and you don't know the field, then perhaps it could be better to run 61.

@Kuma isn't your first point dependent on what your matched against? Again, I think this comes down to meta-game knowledge. If you knew exactly what you'd be playing every time, then yes, you could rank your cards from 1-75 (don't forget your other 15). However, some of those might be relatively weak against some portion of the field, which would change their ranking.

i'm going to stick to my last post and claim that this really comes down to understanding the meta-game. If you know the meta-game like the back of your hand, you should be able to narrow it down to 60.

[Gui]

Playing more than 60 cards is wrong:

1: The cards in your deck can be ranked in order of overall power from 1 to 60. This is hard to do in practice, but we all know that some cards in our deck are better than others.

They can, but then you are taking into account the % of decks played in the metagame.

Let's say, just for the sake of argument, that you can predict 100% of your metagame and this wouldn't require NOAA Super Computers and complex models to be done.

Then you'd have to calculate the % of the time each card wins against each given possible matchup in the field in order to determine which is better, using a weighted mean.

This, without taking into consideration the interaction of that card with the rest of your deck, and the interaction of that card with varied build from different decks. Only then you'd have, precisely, the power level of each card on your deck, and could order then from 1 to 60.

Then you'd have to take out the least powerful card you found, add yours, and do the whole process again, and see if your % winning against the field is now better.

Well, if you really can do all that, suit yourself. I'd rather take almost the same risk by adding the 61st card instead.


No, really, noone knows exactly the % of matchups and the % of time each of his cards wins, the stupid thing is not playing 61 cards, the stupid thing is claiming that you know that much, enough to dismiss 61 card decks all the time.

And just to reinforce, I'm not saying 61 cards should be the starting point to build a deck, just saying that given a few situations, you can as well play 61 cards. Unless of course you have NASA super computers and 300 IQ.

[SpikeyMikey]

Playing more than 60 cards is wrong:

1: The cards in your deck can be ranked in order of overall power from 1 to 60. This is hard to do in practice, but we all know that some cards in our deck are better than others.

2: Running a 61st card decreases your chances of drawing the 60 cards more powerful than your worst card. It's only a small amount, but why willingly sacrifice any chance of drawing your best cards, especially when there's a card in the list that's weaker than all the other cards that you don't have to run? You need to figure out which card this is and cut it.

Therefore, playing more than 60 cards is wrong.

This is the common logic behind the idea that 60 cards is a hard and fast "rule". It's a theory that dates back to the late 90's and has held on as firmly as the card counting rule of "card advantage" (card played is -1CA, card drawn is +1). It is also absolutely and unequivocally wrong. Outside of the ratio argument that I gave above, you can use two general scenarios to show the fallacy of this theory.

Let's follow the chain to its logical conclusion. If card 61 is better than card 62, and card 60 is better than 61 and 59 is better than 60, we would be forced to conclude that the best deck would be a single card, as card 1 is obviously superior to card 2. Obviously, this is impossible due to the mechanics of M:tG (you'd deck, you wouldn't actually be able to cast this card, etc.). Even if we remove decking from the equation, a 13-14 card version of a current deck wouldn't be competitive (if it were legal). There's not enough room to boil Counterbalance down to 14 cards. You're looking at 5 lands (5.13 lands would be equivalent to 22 lands in a 60 card deck) and 9 spells. You want to fit Sensei's Divining Top, Counterbalance, Swords to Plowshares, Force of Will and Tarmogoyf into the deck at a minimum. You need a minimum 4 blue cards to support Force of Will (17 blue cards in a 60 card deck is equivalent to 3.96 in a 14 card deck) so you'll have to have at least 2 more. We've got 2 cards beyond that to play with. And you have little to no redundancy, so if something doesn't resolve or if it's removed, you're fucked.

The other way to show the fallacy of your theory is to discuss deckbuilding with a removal of the 4-of rule. If there were no limits on how many copies you could run of a given card, would you go to a single unique spell? Probably not. You need more than one effect. Tarmogoyf might be how you win the game, but 22 lands and 38 Tarmogoyfs will get the everloving snot kicked out of it by half the decks in Legacy. Your opponent plays a single Wild Nacatl and proceeds to beat you to death with it. We had a similar discussion in IBA's "Tour de Force" thread. The idea was that a 0 mana hard counter with no drawback would be the best card in the format, hands down. But you couldn't run a 60 Tour de Force deck. At a minimum, you'd have to have at least 1 win condition. Say a Mishra's Factory. And the way to beat that deck would be to run at least 1 Wasteland. So the way to beat that is to run say Crucible of Worlds and more Factories. And the way to beat that is to run Krosan Grips. And eventually, you follow this chain of "in order to beat this..." and you find out that you're not actually running all that many Tour de Forces. Maybe you can fit 8-10. Maybe not even that many. Any way you slice it, the format has grown a bit more complex from 20 Lotus, 20 Channel and 20 Fireball.

There are plenty of cards in plenty of decks that are phenomenally powerful and necessary for the function of the deck that you don't even run 4 copies of. When's the last time you saw a Natural Order deck run 4 Natural Orders? It's the focus of the whole goddamn deck and they play 3. Everyone plays 3. Because it's all about ratios and percentages. You don't want multiple NO's clogging up your hand. And you don't need to resolve more than 1. It's easily the most powerful card in the deck. No other card gives you 10 unblockable, untargetable power worth of creature for 4 mana. A pair of Tarmos may give you 10 power, but they'll be blockable and Plowable. And yet, despite being the most poweful card in the deck, you don't even run the maximum legally allowable.

And as a further exercise, let's discuss Brainstorm. Brainstorm has been called, by more talking heads than I can count, the best card in Legacy. But Brainstorm's power has a directly proportional relationship with the number of unique cards in a deck. For instance, if your deck is nothing but Relentless Rats, some U/B lands and Brainstorm, you're not getting much out of it. As the number of unique cards in the deck rises, the value of Brainstorm increases. Brainstorm is not the only card that acts this way. Any sort of filter effect becomes stronger the more unique cards the deck has. Creatures like Knight of the Reliquary gain a lot of EV from running tutorable lands even though those lands are suboptimal in a non-KotR situation (because they can't reliably be found without KotR and can't be run as 4-ofs). Think Karakas, Maze of Ith or Volrath's Stronghold. Maybe I decide that Volrath's Stronghold is the weakest card in my Rock deck and I cut it. But in doing so, I make my 4 KotR's worse. And maybe that loss is greater than what I'd lose cutting a Tarmogoyf. But cutting a Tarmogoyf makes my Volrath's Stronghold weaker. In an era with so many filtering and tutoring effects, cutting cards does not necessarily make a deck stronger. And you can't just cut the filtering effects, assuming that the deck will be stronger for simply playing more "good" spells and not running filler filter spells. Take Nassif's 2008 Counterbalance deck and pull Brainstorm and Ponder from it. See how playable it is.

[Julian23]

Thanks Mikey for pointing out what I might have been to lazy to present. Especially that 38 goyf 22 lands example really showcases how you can't just argue that Card 60 will/should always be better than card #61.

[Kuma]

@Kuma:

I disagree with your first assumption. Therefor I disagree with your conclusion.

Care to elaborate?

@Kuma isn't your first point dependent on what your matched against? Again, I think this comes down to meta-game knowledge. If you knew exactly what you'd be playing every time, then yes, you could rank your cards from 1-75 (don't forget your other 15). However, some of those might be relatively weak against some portion of the field, which would change their ranking.

Well yes, the weakest card in your deck is obviously match dependent. You should have enough metagame knowledge to know what the overall weakest card in your deck is, or you have another problem entirely.

When I say rank the cards in your main deck from 1-60, I mean in terms of overall usefulness against your likely opponents. If you're running a 61 card toolbox deck, there's probably one card in your main deck that is good in fewer and less common matchups than the rest. That card needs to be in your sideboard or out of the 75.

i'm going to stick to my last post and claim that this really comes down to understanding the meta-game. If you know the meta-game like the back of your hand, you should be able to narrow it down to 60.

Exactly, and if you don't know the metagame you have a bigger problem than the 61st card. Why not learn the metagame and solve both problems?

They can, but then you are taking into account the % of decks played in the metagame.

Obviously.

Let's say, just for the sake of argument, that you can predict 100% of your metagame and this wouldn't require NOAA Super Computers and complex models to be done.

Then you'd have to calculate the % of the time each card wins against each given possible matchup in the field in order to determine which is better, using a weighted mean.

This, without taking into consideration the interaction of that card with the rest of your deck, and the interaction of that card with varied build from different decks. Only then you'd have, precisely, the power level of each card on your deck, and could order then from 1 to 60.

etc...

I'm not saying that we can be absolutely certain what the worst overall card in our deck is, but if you know your metagame and have played with the deck significantly, you should have a pretty good idea of what that card is.

If you can't trust your judgment to know what the weakest card in your deck is, how can you trust your judgment to know that the 61st card you've added is correct?

And just to reinforce, I'm... just saying that given a few situations, you can as well play 61 cards. Unless of course you have NASA super computers and 300 IQ.

The 61st card reduces the chances of drawing your other 60 cards. That is a mathematical fact. If that 61st card is really better than the other 60 cards, it's not the 61st card, it's the 1st.

Let's follow the chain to its logical conclusion. If card 61 is better than card 62, and card 60 is better than 61 and 59 is better than 60, we would be forced to conclude that the best deck would be a single card, as card 1 is obviously superior to card 2. Obviously, this is impossible due to the mechanics of M:tG (you'd deck, you wouldn't actually be able to cast this card, etc.).

Except that since we can't take our decks below 60 cards, the fact that the idea breaks down below eight cards isn't relevant. We aren't discussing theoretical 59 card or smaller decks, we're arguing 60 vs. 61.

Even if we remove decking from the equation, a 13-14 card version of a current deck wouldn't be competitive (if it were legal).

I think combo would be pretty ridiculous if you could run 13-14 cards. The odds of you having your combo plus protection would be much higher than is acceptable for Legacy power levels.

A few years back, my friends and I played 15-card singleton with everything legal as a thought exercise. The best two decks in that format were Stax and Storm. Stax was good, because it didn't have to worry about consistency at all and won pretty much every game it went first. Storm was good because you put your whole deck in your hand and just won. I can comfortably say that at least that Stax deck would be ridiculous even in Vintage.

The other way to show the fallacy of your theory is to discuss deckbuilding with a removal of the 4-of rule. If there were no limits on how many copies you could run of a given card, would you go to a single unique spell? Probably not. You need more than one effect.

Oh, I don't know. I'll bet Burn would love to be 22 Mountains and 38 Lightning Bolts. It would be better if there were some Fireblasts in there, sure, but it would be ridiculous with just Bolt.

Tarmogoyf might be how you win the game, but 22 lands and 38 Tarmogoyfs will get the everloving snot kicked out of it by half the decks in Legacy. Your opponent plays a single Wild Nacatl and proceeds to beat you to death with it. We had a similar discussion in IBA's "Tour de Force" thread. The idea was that a 0 mana hard counter with no drawback would be the best card in the format, hands down. But you couldn't run a 60 Tour de Force deck. At a minimum, you'd have to have at least 1 win condition. Say a Mishra's Factory. And the way to beat that deck would be to run at least 1 Wasteland. So the way to beat that is to run say Crucible of Worlds and more Factories. And the way to beat that is to run Krosan Grips. And eventually, you follow this chain of "in order to beat this..." and you find out that you're not actually running all that many Tour de Forces. Maybe you can fit 8-10. Maybe not even that many. Any way you slice it, the format has grown a bit more complex from 20 Lotus, 20 Channel and 20 Fireball.

Okay, I'll concede the point that if there were no four-card limit in Magic, decks would not usually have lands plus multiple copies of one card. That said, the only way this is relevant to the discussion of 60 vs. 61 is if 60 wasn't enough to have everything you need to have a functioning deck, and 61 was. I don't think a single deck in Magic fits that criteria.

I'm the guy saying 61 is never correct. All you guys need is one counterexample of a Legacy deck where 61 is obviously better than 60.

You can't find one? Then, in the absence of solid evidence, shouldn't we go with what we know to be mathematically true, that running 61 cards hurts our chances of drawing our most powerful cards, no matter what they are?

There are plenty of cards in plenty of decks that are phenomenally powerful and necessary for the function of the deck that you don't even run 4 copies of. When's the last time you saw a Natural Order deck run 4 Natural Orders? It's the focus of the whole goddamn deck and they play 3. Everyone plays 3.

You might want to check out the NO RUG thread.

Because it's all about ratios and percentages.

Maybe 61 cards gets you the perfect set of percentages. If someone, in defense of some 61 card list could give me the perfect percentages, or at least what they think the perfect percentages are, and showed me how it's impossible to get those percentages at 60 cards, I'd concede the point.

But that's never what happens when somebody suggests a 61 card deck. It's always vague arguments about how you can't prove 60 is better than 61. Or that they don't know what to cut. Or how they aren't sure what the metagame will be so they want to cover their bases. Or the worst of them all, that it just doesn't matter, despite that being demonstrably, mathematically false. Every percent is crucial in tournament Magic. If you don't believe that, you probably don't have the mindset to be a truly great Magic player.

And as a further exercise, let's discuss Brainstorm. Brainstorm has been called, by more talking heads than I can count, the best card in Legacy. But Brainstorm's power has a directly proportional relationship with the number of unique cards in a deck.

To a point, yes. I'd argue that Brainstorm is better in Legacy than EDH, because it's more likely to get you the card/cards you need.

Creatures like Knight of the Reliquary gain a lot of EV from running tutorable lands even though those lands are suboptimal in a non-KotR situation (because they can't reliably be found without KotR and can't be run as 4-ofs). Think Karakas, Maze of Ith or Volrath's Stronghold.

Once again, this is only true to a point. Running too many singletons will lead to bad draws when you get useless singletons in your hand instead of relevant cards.

Even if Knight of the Reliquary gains EV from running more singleton tutor targets, not only are you less likely to draw a Knight of the Reliquary for each card over 60, you're more likely to draw a useless silver bullet than a useful card. You said yourself that you need a variety of cards in a deck to win. How does running cards that are marginal in a large number of matchups help you get relevant cards?

Maybe I decide that Volrath's Stronghold is the weakest card in my Rock deck and I cut it. But in doing so, I make my 4 KotR's worse. And maybe that loss is greater than what I'd lose cutting a Tarmogoyf. But cutting a Tarmogoyf makes my Volrath's Stronghold weaker. In an era with so many filtering and tutoring effects, cutting cards does not necessarily make a deck stronger. And you can't just cut the filtering effects, assuming that the deck will be stronger for simply playing more "good" spells and not running filler filter spells. Take Nassif's 2008 Counterbalance deck and pull Brainstorm and Ponder from it. See how playable it is.

If you've played your Rock deck enough, you should have a good idea of what the worst overall card in your deck is, despite its synergy with other cards in the deck.

I repeat, if you can't trust your judgment to know what the weakest card in your deck is, how can you trust your judgment to know that running the 61st card is correct?

[.dk]

Well yes, the weakest card in your deck is obviously match dependent. You should have enough metagame knowledge to know what the overall weakest card in your deck is, or you have another problem entirely.

When I say rank the cards in your main deck from 1-60, I mean in terms of overall usefulness against your likely opponents. If you're running a 61 card toolbox deck, there's probably one card in your main deck that is good in fewer and less common matchups than the rest. That card needs to be in your sideboard or out of the 75.



Exactly, and if you don't know the metagame you have a bigger problem than the 61st card. Why not learn the metagame and solve both problems?

I think we're agreeing, but strangely coming to different conclusions. I guess I'm saying that the more perfect your knowledge is, the easier it is to make an optimal 60. Since one can never have perfect knowledge, it is conceivable that bringing 76 with you may be the right choice sometimes for reasons already mentioned (marginally reduce the chance of drawing most of your types of cards, and greatly increase the chance of drawing 1 other one).

[Gui]

If you can't trust your judgment to know what the weakest card in your deck is, how can you trust your judgment to know that running the 61st card is correct?

Exactly, but, mathematically, I'm taking less risks adding the 61st than cutting what I believe to be the worth 60th, exactly because the effect of adding the 61st increases almost the same as if it was the 60th, but the reduction of each card in the deck is marginal, technically non-sensible, 1 out of 150 games difference.

I think we're agreeing, but strangely coming to different conclusions. I guess I'm saying that the more perfect your knowledge is, the easier it is to make an optimal 60. Since one can never have perfect knowledge, it is conceivable that bringing 76 with you may be the right choice sometimes for reasons already mentioned (marginally reduce the chance of drawing most of your types of cards, and greatly increase the chance of drawing 1 other one).

I second this opinion.