[Premium Article] So Many Insane Plays -- The New Phyrexia Set Review & Checklis

[Smmenen]

So, my set review is a bit different from the other set reviews I've seen.  It's got alot of commentary I hope you'll find entertaining and enlightening.   I approach the set from a slightly different angle.

And, as a always, for your trading and collection needs, I've updated and revised the Complete Vintage checklist!   I hope you enjoy this article!

The set review is available to all Quiet Speculation insiders (http://www.quietspeculation.com/2011/05/new-phyrexia-vintage-review-vintage-checklist/

OR is available as a downloadable PDF here (http://www.quietspeculation.com/ebook-library/ ) (Just scroll halfway down the page to download) (they keep all the PDFs on the same page )

[Womba]

Do you accept cash...?






If you'd like to make a specific complaint or comment about the actual content of this article, feel free to do so in this thread. 

If you just want to whine about not wanting to pay for a product, you could do so elsewhere, perhaps on Quiet Speculation, or perhaps file a complain with Consumer Affairs.

Alternatively you could just not purchase it.

-Andy

[jtwilkins]

I purchased the article and I was happy with it. The article was a little lite because there (as always) isn't a whole lot to review when it comes to vintage. I think the vintage checklists put it over the top for me in terms of value. I really miss the Legacy set lists Steve did on SSG though. My suggestion would be to do a full Vintage + Legacy set review with both checklists and charge an even $5.00 for it. I think the increase in sales to legacy players would make up for not doubling the price and it would be a extra win for the vintage community because we would have more people exposed to vintage articles.
I really enjoy the example deck lists in the article and there cross over to ideas in legacy deck building.


---Begin Off Direct Topic--
I was one of the nay sayers when Smmenen started to charge for his reviews. My knee jerk reaction was not due to charging for his content (or really personally charging) but I was worried that the impressive research and important opinion of this vintage writer was going to be much much less likely read by non-hardcore vintage players. I just want vintage to stop dying. Maybe my suggestion for Vintage + Legacy will work out. Another cool idea some times used by music artists is to set a buy out limit for the item. Then set a minium purchase price... $5.00 with no maximum. Once the authors monetary goal is met the document because creative commons.

[Smmenen]

I purchased the article and I was happy with it. The article was a little lite because there (as always) isn't a whole lot to review when it comes to vintage. I think the vintage checklists put it over the top for me in terms of value. I really miss the Legacy set lists Steve did on SSG though. My suggestion would be to do a full Vintage + Legacy set review with both checklists and charge an even $5.00 for it. I think the increase in sales to legacy players would make up for not doubling the price and it would be a extra win for the vintage community because we would have more people exposed to vintage articles.
I really enjoy the example deck lists in the article and there cross over to ideas in legacy deck building.

I really appreciate these comments/feedback.  

I've been experimenting with the set review format over the last year.   I've found some recurring things that work (for example, doing a recap, but not just a recap, a recap that actually looks at tournament performance of predicted cards).   Last time, I reviewed every card, and while I enjoyed the exercise, I'm not sure it made for a more entertaining or useful product.   Reviewing every card has the upshot of helping us understand exactly what the parameters of Vintage play are.  I spent hours and hours in Gatherer, looking up every close relative.  

And I agree this set is not as deep as the previous two sets in the block, but I tried to compensate for that by doing something unique.  Instead of doing a commentary or analysis of the set in the conclusion, I found it, I think, useful to present one at the outset.  I think it served both as a way of framing the analysis that followed, but also I think, provided an important backdrop.   That's why, in some respects, I'm particularly proud of this set review.  

I'm glad you find the checklist valuable.  I actually put alot of time into updating it.  I go through every intervening reported Top 8 to make sure that I'm capturing every card that's actually played.   In addition, I review virtually every card in the checklist every time to make sure that the cards are still showing up in Vintage top 8s.   For example, I was going to cut Juggernaut, until I saw it appearing in Top 8s.  I thought for sure it would disappear with the printing of Precursor Golem.  

---Begin Off Direct Topic--
I was one of the nay sayers when Smmenen started to charge for his reviews. My knee jerk reaction was not due to charging for his content (or really personally charging) but I was worried that the impressive research and important opinion of this vintage writer was going to be much much less likely read by non-hardcore vintage players. I just want vintage to stop dying. Maybe my suggestion for Vintage + Legacy will work out. Another cool idea some times used by music artists is to set a buy out limit for the item. Then set a minium purchase price... $5.00 with no maximum. Once the authors monetary goal is met the document because creative commons.

Before I moved to QS, I was writing for SCG premium, so my articles have always been premium, but I understand your concern.   I acknowledge that my potential audience is not as large as it was when I was writing for SCG, but these things are serving an audience at QS that uses this information for slightly different purposes .

I would LOVE to update a legacy checklist, but keep in mind that when I created the Legacy checklist, Wizards sent me the GP Madrid decklists AND SCG was sending me the SCG $5K decklists.   I have access to neither one of these any more, and unlike morphling.de for Vintage, there isn't a major database that accumulates Legacy top 8s.   I suppose I could just rely on SCG $5K data.   I may well do that this summer, and write an article that is basically just updating the Legacy checklist.  I've had a few people request such an article.   I'll see of the folks at QS would be interested in me doing one.

In the meantime, I have two big QS articles I'm working on for this summer.   The first is going to be a long matchup analysis article, and another is going to be the 50 Decks of Vintage + Q2 metagame analysis.  

To be honest, I don't know how I pumped out an article a week like i did was I was writing for SCG.  I'd write like 10 + page articles every week! I just don't know how I had the time.  I used alot of that spare time to write the Gush book, but now that the second edition has been published, I'm looking forward to using that time to more Vintage related content.  

What I'd really like to do more of is more matchup analysis.   This is, I think, one of the richest areas for article formats -- it's a great way to meld pedagogical and entertainment goals.  It's also one of my favorites, and people really enjoyed them.   Maybe some time in the future I can film games with teammates and do a detailed deconstruction of them.   That's been a long term goal of mine, and I hope some day to accomplish it.  When Vintage goes online fully, it will be easy to do.  

[H]

I used alot of that spare time to write the Gush book, but now that the second edition has been published, I'm looking forward to using that time to more Vintage related content.

Speaking of which, i recall you having said all of those who bought the first would get a link the the second edition.  I never did, even after i emailed someone at QS.  Do you know what is the deal with that?

[Smmenen]

I used alot of that spare time to write the Gush book, but now that the second edition has been published, I'm looking forward to using that time to more Vintage related content.

Speaking of which, i recall you having said all of those who bought the first would get a link the the second edition.  I never did, even after i emailed someone at QS.  Do you know what is the deal with that?

Everyone who purchased a copy of the first edition of the ebook should have received a download link to the second edition.

Please message Hi-Val if you haven't: http://www.themanadrain.com/index.php?action=profile;u=502

[Solaran_X]

I definitely look forward to a Legacy checklist again eventually, and will definitely love the 50 Decks of Vintage. Just please go into depth about the differences between seemingly similar decks (such as Stax and Espresso Stax). Me and Steve may have had differences in the past, but he knows Vintage and I got to give him credit for how he puts his knowledge out there for the rest of us.

I did notice a few minor things in Steve's Vintage checklist, both this time and the previous one. No inclusion of Phyrexian Tower and Engineered Plague - both of which are sideboarded in Dark Times. And you include the other aspects of Dark Times. Also no inclusion of Elvish Visionary, Quirion Ranger, and Gleeful Sabotage - all are used in Elves. And you include the other aspects of Elves, including Regal Force and Summoner's Pact which, to my knowledge, see no play outside of Elves.

[Smmenen]

I definitely look forward to a Legacy checklist again eventually, and will definitely love the 50 Decks of Vintage. Just please go into depth about the differences between seemingly similar decks (such as Stax and Espresso Stax). Me and Steve may have had differences in the past, but he knows Vintage and I got to give him credit for how he puts his knowledge out there for the rest of us.

I did notice a few minor things in Steve's Vintage checklist, both this time and the previous one. No inclusion of Phyrexian Tower and Engineered Plague - both of which are sideboarded in Dark Times. And you include the other aspects of Dark Times. Also no inclusion of Elvish Visionary, Quirion Ranger, and Gleeful Sabotage - all are used in Elves. And you include the other aspects of Elves, including Regal Force and Summoner's Pact which, to my knowledge, see no play outside of Elves.

Thanks for the feedback.   I will certainly go into depth in any article on the 50 decks of Vintage.  Expect it to weigh in at 100 pages.   

An updated  Legacy checklist seems like low hanging fruit, if I use the GP results and this year's SCG 5K Top 16 decklsits.  I'll have to negotiate it with QS, but expect that in the not too distant future/ a few weeks. 

[evouga]

Stephen, I was particularly intrigued by the discussion at the beginning on reducing variance, and the difference between Street Wraith and blue library manipulation spells. The following quote jumped out at me:

Now, you might wonder: doesn’t that suggest that we should run decks of more than 60 cards but use
more library manipulation to compensate? Perhaps, but I don’t think so.

Could you elaborate a little bit more on this point? I would think that if a 60-card deck with an extra Ponder/Preordain is more optimal than a 59-card deck without, the same would hold true for 61 and 60. There is a tipping point where the decrease in variance due to card filtering no longer exceeds the increase in variance due to having a larger library -- but I have no reason to believe (and if it's true it would be a remarkable coincidence) that that magic number sits exactly at 60.

The point is moot if a deck of 60 is already running the full Brainstorm, Ponder, 4x Preordain (although the suite of second-string card manipulation spells for U, e.g. Sleight of Hand, should probably be at least briefly considered.) Otherwise, your argument seems to suggest expanding your library over 60 to fit them is likely correct. What are your thoughts?

[evouga]

Also, a minor rules quibble:

Notably, [Dismember] can be played for one mana under a Trinisphere.

I'm pretty confident that's incorrect. If you choose to pay  and 4 life, Trinisphere will raise that cost to and 4 life.

[Joblin Velder]

Also, a minor rules quibble:

Notably, [Dismember] can be played for one mana under a Trinisphere.

I'm pretty confident that's incorrect. If you choose to pay  and 4 life, Trinisphere will raise that cost to and 4 life.

This is correct.

[Smmenen]

Stephen, I was particularly intrigued by the discussion at the beginning on reducing variance, and the difference between Street Wraith and blue library manipulation spells. The following quote jumped out at me:

Now, you might wonder: doesn’t that suggest that we should run decks of more than 60 cards but use
more library manipulation to compensate? Perhaps, but I don’t think so.

Could you elaborate a little bit more on this point? I would think that if a 60-card deck with an extra Ponder/Preordain is more optimal than a 59-card deck without, the same would hold true for 61 and 60. There is a tipping point where the decrease in variance due to card filtering no longer exceeds the increase in variance due to having a larger library -- but I have no reason to believe (and if it's true it would be a remarkable coincidence) that that magic number sits exactly at 60.

Perhaps I did not address this question as well as I could have.   Maybe I don't have a good answer.   This is sort of a timeless question, and conventional wisdom circumscribes the discussion.  

My answer goes back to the distinction I was making in the article when I said:

Let me begin by making a critical distinction.  I would like to distinguish between cards drawn in the opening hand and cards drawn during the course of the game. Both sets of cards are drawn from the library, but they are drawn under very different circumstances.

The reason to run only 60 cards is to maximize your chances of seeing the very best cards (Ancestral Recall, Black Lotus, etc) and cards you want to draw or see early on in your opening hand. Also, the more cards you include in your deck, the more variance you will get in the kinds of draws you see in your opening hand, the more likely the mana ratios are to be undesirable, etc.  

This fact has to then be balanced against the in-game variance reducing effects of adding blue library manipulation spells (ala Preordain, Brainstorm, Ponder) as a 61st card (or beyond).  

One might argue that running Preordain over a Smmenen's Broken Bauble (or Gitaxian Probe) suggests the answer already: that it's better to have a 1 mana spell that sees 3 cards than a 0 mana spell that sees just one card, and therefore, pure deck reduction is not as important as cards that actually reduce variance by seeing multiple cards.  It would then seem to follow that one should include a library manipulation spell as a 61st card.   After all, if you run Preordain as a 61st card, once the game begins, you are reducing in-game variance more than by simply running 60 cards, which is logically equivalent to a 61 card deck with a zero-life costing Street Wraith in that slot, isn't it?  

I think the answer comes from the simple fact that in game variance MAY BE reduced by running more than 60 cards, but opening hand variance is not.  Both goals are important, and they are achieved in different ways.   By running just 60 cards, you minimize variance in opening hands.  But by running more than 60 cards, particularly if the additional cards are efficient forms of library manipulation, you reduce in-game variance, as you suggest.  I don't know of a simple way of balancing these two goals, but I would say that the first is very important, and that's why we typically run just 60 card vintage decks.  

Is that persuasive?  

EDIT: If, as I assert in the article, that a central goal of Vintage is to reduce variance (or to become more deterministic), then the pursuit of variance reducing strategies may appear contradictory, but satisfied in the ways i just outlined.  

[evouga]

I think the answer comes from the simple fact that in game variance MAY BE reduced by running more than 60 cards, but opening hand variance is not.

Yes, I think that's the central tension.

Here's a quick experiment I ran: suppose your goal is to find a 4-of on your first turn. How does the probability of success vary as you increase the size of your deck, and/or add library manipulation cards?

Obviously this analysis requires a heap of assumptions. Here are those I chose:
1. Your library contains 16 blue sources.
2. You don't mulligan.
3. If you don't draw your 4-of in your opening 7, you will play only one blue source (if you drew one) and then your best manipulation spell. (I didn't take into account cases where you drew 2x blue source and 2x U spell, or chain one U spell into a second.)
4. You are playing first (no sphere effects to fight through.)
5. Your opponent won't try to Force your U spell (dubious for Ancestral Recall.)


Here's the plot of the result. The horizontal axis is the size of your library, the vertical axis the probability of finding your 4-of.
The plot lines, from bottom to top:
Purple: No library manipulation spells.
Red: Deck includes 1x Ancestral, 1x Brainstorm, 1x Ponder
Green: Deck includes the above, plus 4x Preordain
Blue: Deck includes the above, plus a Sleight of Hand.

What conclusions can we draw here?
1. Adding Ancestral, Brainstorm, and Ponder improves your probability of finding the 4-of despite increasing the size of your deck by 3 cards. Simply adding them to a 60-card deck increases your success probability by ~2% (compare to 4% if you instead cut 3 pieces of chaff to make room.)
2. Adding 4x Preordain is also always correct, though the difference here is smaller. A 67-card deck with the full U suite has a ~2.5% higher probability of success than a 60-card deck with none (compare to ~8% if you instead cut 7 cards.)
3. Sleight of Hand is past the tipping point -- adding it does more harm than good.

Of course, maximizing your chance of getting a 4-of on T1 is a very very simplistic objective. Whether or not the 2.5% is still worth it after taking into account increased vulnerability to spheres, FoW, Chalice@1, etc. is also debatable. But I think it's safe to say that a) your argument about the relative worth of library manipulation spells vs. Busted Bauble is sound, and b) playing a blue-based deck without 4x Preordain is likely a mistake, even if adding them would require a deck to play slightly more than 60 cards.

EDIT: For those interested the Mathematica notebook containing the full combinatorics is available here.

[Metman]

I don't think you factored in that increasing a non free spell in the likes of Probe or Preordain requires you to play more land, which in turn forces you to increase the card count in the deck.  It isn't quite as simple as stuffing in four blue cards for sake of manipulation. 

[Stormanimagus]

@evouga -> is 0.48 representing 48% I assume it is but I just wanted to check you on your scale. Also, since Gitaxian Probe is potentially free that changes the amount of lands you are forced to run in a 60 card deck and thus may change these ratios a bit.

While I do think these sorts of thought experiments are beneficial, I also think that #4 is so rare because of the amount of shops out there that you cannot assume that even half the time right now. Play situations are what matter in magic. Not theoretical scenarios.

-Storm

[evouga]

@Metman: I did factor in increase in mana screw in the short term: I assumed that the number of blue sources was held constant at 16 and took into account that a U spell could not be played unless you'd also drawn one of those mana sources. You're right that longer-term mana development was not considered.

@Stormanimagus: Yes, 0.48 is 48%.

I completely agree that mathematical analysis will never fully replace playtesting. Nevertheless, I believe that analysis of theoretical situations has much to offer Magic and that it has not been used nearly to its full potential so far. Magic is much closer to Poker, where odds of important events can be exactly determined and decisions objectively evaluated, than a hopelessly strategic (vs tactical) game like Go. I don't think a computer will be able to pilot a control deck to a tournament win any time in the near future, but if I had the time and motivation I'm confident I could create a computer Dredge pilot that would outplay pros using the same deck.

Differences of a few % are extremely hard to detect when playtesting. By definition they would requires hundreds of test games to notice, and "soft" factors like your mood, wakefulness, etc. easily have a greater than 2% effect on your win percentage. Rigorous mathematical analysis can give you a small but meaningful edge over players relying on near-optimal rules of thumb.

[MaximumCDawg]

Here's the plot of the result. The horizontal axis is the size of your library, the vertical axis the probability of finding your 4-of.
The plot lines, from bottom to top:
Purple: No library manipulation spells.
Red: Deck includes 1x Ancestral, 1x Brainstorm, 1x Ponder
Green: Deck includes the above, plus 4x Preordain
Blue: Deck includes the above, plus a Sleight of Hand.

      

This is fascinating, but I think I might be reading your graph wrong.  You say that...

2. Adding 4x Preordain is also always correct, though the difference here is smaller. A 67-card deck with the full U suite has a ~2.5% higher probability of success than a 60-card deck with none (compare to ~8% if you instead cut 7 cards.)

   

... but it looks to me like adding 4x Preordain over and above 1x Ancestral 1x Brainstorm 1x Ponder actually is a net loss; from 44% for a 63 card deck down to 43% for a 67 card deck.  So why is running 4x Preordain always correct?

[evouga]

This is fascinating, but I think I might be reading your graph wrong.  You say that... ... but it looks to me like adding 4x Preordain over and above 1x Ancestral 1x Brainstorm 1x Ponder actually is a net loss; from 44% for a 63 card deck down to 43% for a 67 card deck.  So why is running 4x Preordain always correct?

44% is for a 60-card deck with Ancestral, Brainstorm, and Ponder. The percentage for a 63-card deck with the same is a smidge over 42% (follow the vertical line over 63 until where it intersects the red curve.)

[Solaran_X]

Also, a minor rules quibble:

Notably, [Dismember] can be played for one mana under a Trinisphere.

I'm pretty confident that's incorrect. If you choose to pay   and 4 life, Trinisphere will raise that cost to and 4 life.

I am not totally sure about that. This isn't like Force of Will or Daze where you are paying an alternate casting cost for the entire card, where the mana cost is actually 0 in the eyes of Trinisphere. With Dismember, you're paying 1pBpB for the spell, with actual card text (rules text) saying pB can be paid for with either 2 life or B. Either way, pB is met for the actual casting of the card at the printed mana cost, which is 3CMC.

[evouga]

Also, a minor rules quibble:

Notably, [Dismember] can be played for one mana under a Trinisphere.

I'm pretty confident that's incorrect. If you choose to pay   and 4 life, Trinisphere will raise that cost to and 4 life.

I am not totally sure about that. This isn't like Force of Will or Daze where you are paying an alternate casting cost for the entire card, where the mana cost is actually 0 in the eyes of Trinisphere. With Dismember, you're paying 1pBpB for the spell, with actual card text (rules text) saying pB can be paid for with either 2 life or B. Either way, pB is met for the actual casting of the card at the printed mana cost, which is 3CMC.

The comp rules are actually somewhat unclear on this interaction at the moment, but it's been officially ruled to raise the cost to 3 generic mana even if you pay Phyrexian mana with life. See for instance this thread http://community.wizards.com/go/thread/view/75842/27692021/601.2b_needs_to_reference_phyrexian_mana_symbols_aka_Trinisphere_vs_phyrexian_mana

[Solaran_X]

Also, a minor rules quibble:

Notably, [Dismember] can be played for one mana under a Trinisphere.

I'm pretty confident that's incorrect. If you choose to pay   and 4 life, Trinisphere will raise that cost to and 4 life.

I am not totally sure about that. This isn't like Force of Will or Daze where you are paying an alternate casting cost for the entire card, where the mana cost is actually 0 in the eyes of Trinisphere. With Dismember, you're paying 1pBpB for the spell, with actual card text (rules text) saying pB can be paid for with either 2 life or B. Either way, pB is met for the actual casting of the card at the printed mana cost, which is 3CMC.

The comp rules are actually somewhat unclear on this interaction at the moment, but it's been officially ruled to raise the cost to 3 generic mana even if you pay Phyrexian mana with life. See for instance this thread http://community.wizards.com/go/thread/view/75842/27692021/601.2b_needs_to_reference_phyrexian_mana_symbols_aka_Trinisphere_vs_phyrexian_mana

I wouldn't call that officially ruled. WotC_MattT said a change had been made for the next update, seeming in reference to the ubiquity of the wording of the comprehensive rules in reference to Phyrexian Mana. He didn't actually say how the spell interacts with Trinisphere. It remains my take that Trinisphere still sees 3CMC spent to cast Dismember, since the pB requirement can be paid for with either 2 life or B. pB is still being paid for. We'll see for sure soon.

[evouga]

There's also the thread linked to, http://forums.mtgsalvation.com/showthread.php?t=319893, where a NetRep confirms the Dismember costs 3.

[Smmenen]

There's also the thread linked to, http://forums.mtgsalvation.com/showthread.php?t=319893, where a NetRep confirms the Dismember costs 3.

This is a great example, if one was ever needed, where two rules interact in ambiguous ways, where two mutually exclusive reasonable interpretations are possible, and a Rules person is needed to resolve them, by choosing one over the other.   

[Elric]

Could you elaborate a little bit more on this point? I would think that if a 60-card deck with an extra Ponder/Preordain is more optimal than a 59-card deck without, the same would hold true for 61 and 60. There is a tipping point where the decrease in variance due to card filtering no longer exceeds the increase in variance due to having a larger library -- but I have no reason to believe (and if it's true it would be a remarkable coincidence) that that magic number sits exactly at 60.

Having not read this article, I'd say that if adding an extra Ponder/Preordain to a 59 card deck while turning into a 60 card deck improves it, the correct thing to do is to add that Ponder/Preordain in the place of a different card, so your deck remains 59 cards.  Similarly, if your 60 card deck can be improved by making it 61 with an extra Ponder/Preordain, it can be improved more by removing another card at the same time.

Tutors and the ability to fit in specialized bullets is about the only reason I can think of why this might not hold (no deck will have enough cantrips that exact mana ratios are that important), but that seems implausible.