Chiming in with the perspective of someone who has played on the pro tour more times than I have played games of commander: I would scoop in most cases if it hurt the player attacking me.
Lets use the example of 3 players.
Player A (You) has nothing in hand/play, is at 10 life. Your deck is full of removal spells.
Player B has a 10/10 lifelink and is at 2 life.
Player C has a mage-ring bully and has 30 life.
Turn order is player A, then player B, then player C. After passing the turn, I will warn player B that if he attacks me (killing me), I will concede to the attack, then he will lose to mage-ring bully, which must attack (lets assume B drew land). If on the other hand he attacks player C, there is still a very real chance that I as player A will come back and win.
So, clearly strategic concession raises my win % here. It also raises your win % in the case of the insurrection example - if the insurrection player knows you will concede if it resolves, then they will be less likely to case it there. Basically, I will always try and damage the player who finises me off in multiplayer as much as possible. I will cultivate a reputation of using everything I have against the player that brings me down (but only if Im actually about to lose).
Now, is this mindset practical in casual? Probably not - hence why I dont play casual magic much. However, if a bunch of people with competitive mindsets all decide to play a multiplayer game with cash on the line - then it seems this type of behavior makes sense.
@MrPhysics Could you share us the reasons for your sideboard choices and 1-ofs in the maindeck in your Pro Tour list? Thank you so much!
Since you asked nicely ;).
There were generally two reasons - I had tested so much that I knew I wanted certain exact ratios of cards with certain functions in my deck, and secondly, I really wanted as many sideboard slots as possible, so I decided to mainboard some sideboard cards.
I believe I had 4 1-ofs in the maindeck - probe, dismember, spell pierce, and become immense.
For a very long time, I actually had 2 dismember main and 2 side, since I really liked dismember against pod (full 4 copies post side).I still liked it vs twin and abzan and affinity, but it is obviously terrible vs burn,amulet, and scapeshift. Even vs slower versions of zoo I liked it. I also learned before the event that pantheon was on infect. Given my expected meta, I knew I wanted 2 dismember in my 75, but felt it was good enough vs the expected field that it was fine to maindeck one copy to save a sideboard slot.
Become immense I tested a lot as both a 1-of and as a 2-of. I felt the PT meta was going to be slightly more combo heavy than it was, were might of old krosa is usually better. Obviously BI is worse in multiples, so I went with the 'safe' option of a single copy.
The miser's spell pierce has always been good to me. Its definitely not something you want to draw multiple of, but it is devastating if it works. Also, casting spell pierce g1 is great since if you get them with it, they will play around it all match not knowing your a madman with exactly 1 spell pierce. Also, I knew I wanted at least 2 pierce (and possibly a dispel if I could fit it), so maindecking a copy saves me another sideboard slot. Also note that the above 3 cards are typically bad in multiples, Id almost always rather have 1 dismember and 1 spell pierce than 2 of either.
The 1-of probe is because I feel this deck works best with 20.3 lands, but I cant play 20.3 lands, so I play 20 lands and a probe. Probe is actually a really great card, but becomes less important once you have a ton of experience with the deck. Your lifetotal really matters, and Im already maindecking bad cards vs burn (dismember), so I really cant afford to play many probes. Finally, its also a card thats bad in multiples, just like all my other 1-ofs
Also, at the risk of sounding arrogant, I planned on top 8ing (I plan on winning every tournament I enter, thats just the competitive mindset I have) - I really felt like my modern prep was outstanding, and I got some very good draft advice the day or two before the PT. Imagine your my opponent and have my list - its much harder to play around all these random 1-ofs than if I just used 3 spell pierce for example.
On the sideboard:
3 nature's claim/1 hurkyl's recall: The recall was my 15th sideboard card, going into the PT, Id planned on having a 3rd relic, but changed my mind to wanting a dispel, then I tried to find fog, but the vendors didnt have it, so I settled on recall. Now, the only matchup where Id want all 4 claims is affinity, and if thats true, then I feel like hurkyl's recall is better than claim vs affinity (at least the first copy, Id probably rather have 2 claim than 2 recall, but id rather have 1 of each than 2 claim). 3 claims is needed vs burn, so I didnt seriously consider fewer copies. Its also nice vs affinity, and in general is very solid vs an open field, nature's claim is simply a very versatile card
1 dismember - as before, I already have 1 main, and id like a second vs the mirror, abzan, twin, etc. Dismember is another card thats good to have as an option - for example, I like a couple copies vs storm, as I find its very difficult for them to race me without electromancer.
1 spell pierce - pierce stops things like blood moon and living end and wrath (not verdict, but people often respect thrun). You really want at least 2 pieces of countermagic in infect, if not more
1 dryad arbor - amazing vs abzan, 'ok' vs non-combo. This card is just all around reasonable, although drawing it sucks. I probably shoudlve just maindecked it - then id get another sb slot!
2x relic - as mentioned previously, Id considered a third relic. You never know when you will face dredge or living end (it was key in both my wins vs living end!) or gifts -> rites. Also, Against abzan I liked it, and I also like it vs things like temur twin. Liking relic so much is another reason I didnt overload on become immense
1x spellskite - great vs boggles and burn - ok in the mirror, either great of mediocre vs twin (bad vs grudge)
1x wall of roots - I really wanted a card specifically for burn, and this was the best I could come up with. Spellskite, while great, can be hard to use since you have to shock for blue mana, and they usually have smash to smithereens post side. Generally speaking, Id rather turn 2 wall than spellskite vs burn, whereas spellskite it much better later, and for ths reason I went with a 3/1 split in the 75
2x twisted image : Once again, amazing in the mirror, and I knew pantheon was on infect (and I was legitimately afraid of their 3x twisted image). I also bring it in in a ton of matches since the upside is so high and the downside is so low, especially once you factor in wild defiance
2x viridian corrupter: Since I skimped on become immense, my list is a bit weaker vs chalice. I like having at least 4 answers to chalice in my deck, and once again - I like this card in the mirror, since it will often trade with a guy and a pump spell, and furthermore, can eat opposing skites. I also like corrupter in matches where I dont like skite, but have artifacts that arent critical to kill, but that I incidentally want to kill, for example, vs a deck like merfolk (aether vial) or hatebears (vial, splicer tokens). When corrupter is good - its often really good, and I like having this type of high impact card in my side.
Well, maybe he just got lucky and always had it all..
Can you tell me your sideboard for this matchup?
Well if you're playing America control and the guy has zero fetch lands, is playing against infect and electrolytes or bolts YOU then you know you're in a favorable matchup becuase it's like racing a 13 year old girl who is burning the clutch and grinding gears.
Uwr control is a bad matchup, vendillion, snap asters, bolts, Helix, cryptic command and mana leak are all a thorn.
If they don't board out remand, or sphinx then they don't know what they're doing.
And there are a lot of bad America control players on mtgo. I've played them before. Running the theros god and running less than optimal colonnades, in a PTQ nobody is going to helix you for 3 when you have 5 mana and 5 cards in hand playing infect unless it's for lethal
You repeatedly make bold claims in this thread (infect is a glass cannon, infect is weak to uwr control, etc), that many experienced infect players disagree with. You claim dailies are soft. Im curious as to exactly what experience you are basing your assertion that UWR is a bad matchup off of. Do you regularly play against Pro Tour/GP top 8 caliber players, and have significant testing experience against them?
So, youve seen people online make budget choices in a daily. Thus, you discount daily results? This is pretty flawed reasoning. The majority of online players are more skilled than the majority of paper players, because playing more makes you better at this game and online players have that opportunity. All this is coming from a paper player.
drinkard claims significant testing experience shows UWR is not a bad matchup. This does not sway you.
My personal experience. At 5-1 at my last PTQ, I played vs UWR control and won. I drew into top 8, then played UWR again (this time winning 2-0). At my local store, I regularly play against Kyle Boggemes who plays UWR control, and we agree the matchup is very close to even, possibly slightly favoring infect. Kyle has won a recent GP playing a control deck (albeit in standard) and made the finals of a Pro Tour, so hopefully that is sufficient proof he knows what he is talking about. I think my match record vs Kyle infect vs uwr is either even or im up a match lifetime.
So, Ill admit, if your opponent is very skilled, its tough. The thing is, Kyle has played me a bunch of times, and knows my deck. Even very good uwr pilots probably arent used to playing such a fringe matchup. However, even if your opponent is very skilled and knows the matchup, I think its a coinflip matchup at worst.
Overall, my record vs UWR is probably closer to 70% win. Obviously the matchup is not so lopsided, but I think an equally skilled infect player vs an equally skilled UWR control player should favor infect, given that UWR is the more popular deck, so the infect player is presumably more experienced in that matchup.
Note: all this applies for my particular list. Different infect lists may have very different UWR control matchups.
That is if I even win long Grindy matches. The chances of infect winning past turn 4 plummet to near 0 if your opponent does not misplay. Half the time the infect player is the one that misplays. I know I make play mistakes every game everyday..
Oh my gosh please stop posting in this thread.
My experience is that infect is perfectly capable of winning long grindy games against grindy decks (UWR control and Jund). I am 58-12 in matches with this deck at my LGS, and discounting intentional draws I am 13-4 in matches at the PTQ level. If my chances of winning dropped to 'near 0' after 4 turns, Im sure my win % would be far less than 50%. Going all-in early in the game is often a recipe for disaster in many matchups. My matchups against UWR control (considering kiki-resto UWR as a separate archetype) is 2-0 at the PTQ level, and against jund at PTQs its also 2-0. I only remember 1 game of those 4 matches were I won on or before turn 4.
I top4d a PTQ (180+ players) with my U/G infect list this Sunday
Matches were
Round 1: Green Devotion (black+red splashes). Win (2-1)
Round 2: Melira Pod. Win (2-1)
Round 3: Melira Pod. Win (2-1)
Round 4: RUG delver. Lose (1-2)
Round 5: UR Delver. Win (2-1)
Round 6: Affinity. Win (2-1)
Round 7: UWR control. Win (2-1)
Round 8: ID into top 8 (I had best breakers)
Top 8: UWR control. Win (2-0)
Top 4: U/R Twin. Lose (1-2)
None of my matches were particularly easy, as seen by the fact that every single match I played out in the swiss went to game 3. My pod opponents kept naturally drawing their 1 of melira, and I never drew my 2-of dismember, which made things difficult. I had to 'mindslaver' an SCG circuit top15 ranked player to win my third match in ridiculous fashion.
Top 4, I won game 1. Game 2 I nature's claimed a turn 3 blood moon, and then resolved a wild defiance turn 3 with inkmoth 2x, and breeding pool. My opponent spreading seas'd my only green source, and I didnt topdeck a second copy for the win. Its very difficult to lose with wild defiance and 0 pressure vs u/r from that situation (barring blood moon #2) so it was a rough loss. The third game I mulled to a very weak 6 and decided to risk it, and was easily crushed.
Hey guys. Just back from a 6 rounds Trial. I placed 13th on something like 60 players, piloting the mono-G cathedral list from Ultra-Magnus. Fun times, but i hoped for top-8 but got real bad tie breakers 'cause early losses & dropping people.
Match 1: against UR Twin. T2 Path, T3 Deceiver exarch, T4 Resto, T5 Twin. Errr. Both games went like this. 0-2.
Match 2: against BWR Goodstuff with Bob, Bitterblossom, Lightning helix, Lightning bolt, Ajani Vengeant, Path, Liliana. He had too much removal and chump blockers. Sorry. 0-2. Felt like my sideboard was zero ready for that.
Match 3: Bye.
Match 4: against Kiki-Pod. Went well. 2-1.
Match 5: against Tarmo-Twin. He combo'ed T4 the first game. On the second one, I kept a green mana up for a Vines of vastwood. What? Pestermite flashed in end of my turn, T4 combo again. Gee, man, sh*t. How's that even fair? 1-2.
Match 6: against Kiki-Pod. Excellent magic, grindy games. Cathedral and Nexus and Pendelhaven are surprisingly resilient together. But he ended up taking the win with a board that included: Glen-Elendra, Spellskite, Thalia AND Eideolon-of-just-one-spell-per-turn. 1-2.
Overall; I realize I didnt really know what to side for those matchup, and that my sideboard was a little off (ex.: 4 Tormod's crypt never used; never seen a blue instant so Guttural response also was dead, and so on.) TWIN felt HARDCORE. Do you consider it a bad match-up? They always seem to have some tech to delay or neutralize of exalted infectors.
I consider twin a challenging match-up thats about 50/50. Postboard I typically have 4 dismember, a couple nature's claim, a torpor orb and a spell pierce. This is a matchup where UG is probably a lot better than monogreen, but for a not so obvious reason - fetchlands. If you leave a fetchland untapped at EOT, and they go to tap it down, you can sac in response and keep all your 1 mana answers available.
If you have a twin heavy meta, I would suggest adding fetches to your monogreen list if you have them.
That BWR goodstuff deck is going to be very hard to beat for monogreen (or any ordinary G-based infect list). I would simply try my best to use vines/blessing to keep a rancored up myr alive and stuff it full of pump spells. If you run hierarch in your monogreen list it will help vs lili, and allow you to main or side a couple wild defiance, which can help vs all the burn.
Also, just curious, how do you get 13th out of 60 players with a 2-4 record?
Part I: Pro Tour Meta + Why Statistics can be misleading
Thesis: The PT metagame analysis and coverage on the official site is misleading when it comes to figuring out the best modern decks.
When the Pro Tour was over, I figured I'd look at the posted coverage and poke around some articles online to see which decks had performed the best. When I say 'the best' Im not talking about a deck that happened to spike, but if overall, players of that deck did well.
The first graph shows how many of each deck is in day 1, and how many are in day 2. This chart is obviously confounded by the 3 rounds of limited, and to be fair to Nate Price he points that out. The real problem is, that the chart does not include the total number of players to make day 2, and the total number of players. For reference, 252 players day2'd out of 393 players total, so you would expect an 'average' deck to send 64% of its players to day 2. With that in mind, you would think
splinter twin was below average and zoo was right about average by looking at that chart... and of course you would be wrong. There are several reasons for this, not the least of which is that the day 2 meta is different from the day 1 meta.
In terms of placing well in the tournament, its important to not only make day 2, but make day 2 with a good record. At a grand prix, you need 7-2 to day 2, and so a player at 8-1 isnt that far ahead of you, so pretty much everyone that day 2's has a shot at top8. This is not the case at the Pro Tour, where a large number of players day 2 with a 4-4 record and essentially no shot at top 8 (you might get 7th or 8th if you are 12-4 but odds are if you start at 4-4 your breakers are
terrible).
I assume that is why Nate Price made a graph showing deck win percentage - given that they had already made day 2. When you first read the percentages, they are confusing, it seems like all the decks are above average! Even the worst deck, tron, is 48% to win! Its important to remember, of course these percentages are over 50%, he is only talking about the percentages of the winning people! Now, lets say a deck is completely average. It wins 50% of its games. What win percentage should you expect, if you look at the people who finish 4-4 or better with that deck after 8 games? It turns out the answer is 60.7%. However, even with that knowledge, we can really only say a deck is 'good' if it has above this win percentage AND it has ~64%+ players make day 2 with it. If only one of those criteria is true, its a bit tricky to see if the deck is winning more than half its games.
So, what decks were good? Well, the UWR control/midrange strategies were lumped together, but they had good day 2 conversion and good match win %, so they overperformed. Clearly storm is also a winner, as is living end. Zoo is an example of a clear loser when you consider that average is 60% not 50%, when you only look at the win percentage of the 'winners'.Now, we dont have the data for what everyone was playing, but wizards did post the data for everyone that was 6-4 or better in modern. Now, they didnt post exactly how good each deck did, i.e. points, but you can manually look that up if you want. So, how can we decide if a deck was good, if we dont have match win % and only have partial deck information? Well, one way is to look at the total number of pilots who start with a deck, and the total number of pilots that ended up doing well with that deck. Technically that information is available on wizards site. What you can do is, use the search function, and it will tell you all the instances of a word.
Splinter Twin 14
Birthing Pod 13
Cranial Plating 5
Pyromancer Ascension 7
Living End 7
Wild Nacatl 12
Gladecover Scout 9
Mana Leak 18 (and 10 of those are UWR control/midrange)
Scapeshift 4
Rift Bolt 4
So, we see that every deck besides UWR can be reasonably identified by a single card. Now lets look at percentages (i.e. normalize the number of winning lists vs the number of total people running that deck to approximate win percentage in a crude way, we find.
Ok... So is that it? Well, not quite. Due to the small sample size, these results alone could be a bit misleading, what about if you consider only decks 7 wins or better?
So, some decks seriously outperformed (comparatively speaking) when compared to when 18 point decks were included (affinity, UWR, Twin), while others seriously underperformed (Pod, Living End, Scapeshift). This is possibly due to there being a different metagame in day 2 than day 1 (so some decks won a bunch day 1, then stopped winning at all day 2, see Hetrick on Living End) but I am not certain. A large part of this could also be due simply to varience in a small sample size.
So, who is our winner? Its easy to say storm, but you have to keep in mind that its dangerous to trust a small sample size, as our assumption of equal play skill breaks down and it starts to depend more on individual pilots (and Kai and Finkel are certainly above average pilots, for example). Even so, a second look still says storm
1) Storm
2) UWR decks
3) Twin
4) Boggles* (very metagame deck, weak to too many top decks)
Now, this is not considering fringe decks, like blue moon or ad naseum, that did not have many pilots. As you can see, I slanted the results towards the 21+ point decks, as those decks are more likely to actually win the money. The loser amoung the 21+ point decks is definitely pod, which only put a fairly miserable 4/43 pilots into 21+ points in the modern portion (the fact the deck made it to the finals means it will still be played a lot though).
A final note on misleading statistics relates to teams. Blake Rasmussen wrote an article, where he concludes the 'best superteam' was CGB classic, edging out CFB pantheon. From a match % perspective, this is wrong. CFBP managed to match CFB in wins, despite only 12/15 of their players making day2 compared to 15/15 from CFB. Counting 'wins' as your metric of success clearly rewards teams that simply put the most players into the second day. Imagine two teams with 'average' players. Team A
has 2 players with 4-4 records at the end of day 1, whereas team B has one player at 2-6 and another at 6-2. They play out day 2, and now team A has two players with 8-8 records, while team B has one player at 2-6 and one player at 10-6. Counting wins only says Team A is far superior (16 wins vs 12 wins), which seems very unfair.
Part II: Futute Meta thoughts
The meta moving forward:
One thing about predicted future metas is that people are irrational, they wont do an in depth analysis of the tournament to predict what deck they should run, and many people cant afford to switch decks. They will, likely, respect online results though, and once those absorb the pro tour tech, many will likely realize that MTGO is pretty good at showing the top decks in eternal formats.One thing is that I think zoo is a fine deck, despite its poor showing at the Pro Tour. What happened was, people over-prepared for zoo, leading to it being bad, and the anti-zoo decks like boggles and living end doing great. Then, once the zoo players were mostly killed, the actual good decks started showing more dominance (UWR/Storm/Twin) but those decks arent necessarily amazing vs zoo, they are better against anti-zoo decks like living end. This is why, the zoo decks that actually made it to a winning record, many converted that to a 7+ win finish. However, I think people enjoy zoo, and its not bad enough to make people not play it, so I think it will continue to show up in large numbers. Boggles will get weaker, as players will pick up the tech that beats it (porphyr (spelling?) nodes, runed halo, maindeck spellskites in twin, etc). Twin is good, and will stay popular. It comes in many shapes and sizes and is very difficult to beat. I do that that relic of progenitus really showed its power vs tarmo/snapcaster style twin, and I expect relic to increase its prevalence both main and side, as its strong against storm, tarmotwin, living end, etc. The winning Pro Tour decklist seems very solid to me, and I think people will adopt its tech.I expect that despite not many players playing it at the Pro Tour, the Black/Green Obliterator deck will pick up supporters. Reid Duke and Matt Costa are very popular players, and actually playing phyrexian obliterator will be appealing to many. Being able to get aggressive against twin by having slaughter pact, and having the discard package against an increasingly combo meta seems like a great place to be. I predict this style deck will try and fill the hole in the meta brought about by the death of jund, and it will become the new jund (but wont be anything close to the old jund).I think the fringe combo decks (ad naseum, amulet, etc) will stay fringe, and the meta decks like blue moon to die away as people get used to playing against them.
People will likely run an extra basic or two for a while out of respect.
I could see a meta developing that is relatively hateful towards Pod decks (rise of anger of the gods and storm). Once players find a reasonable jund/style deck
though, I predict it will be popular and help keep the combo in check. Otherwise, the top decks will stay the same, and it will be business as usual.
1
Lets use the example of 3 players.
Player A (You) has nothing in hand/play, is at 10 life. Your deck is full of removal spells.
Player B has a 10/10 lifelink and is at 2 life.
Player C has a mage-ring bully and has 30 life.
Turn order is player A, then player B, then player C. After passing the turn, I will warn player B that if he attacks me (killing me), I will concede to the attack, then he will lose to mage-ring bully, which must attack (lets assume B drew land). If on the other hand he attacks player C, there is still a very real chance that I as player A will come back and win.
So, clearly strategic concession raises my win % here. It also raises your win % in the case of the insurrection example - if the insurrection player knows you will concede if it resolves, then they will be less likely to case it there. Basically, I will always try and damage the player who finises me off in multiplayer as much as possible. I will cultivate a reputation of using everything I have against the player that brings me down (but only if Im actually about to lose).
Now, is this mindset practical in casual? Probably not - hence why I dont play casual magic much. However, if a bunch of people with competitive mindsets all decide to play a multiplayer game with cash on the line - then it seems this type of behavior makes sense.
TL;DR depends on play group IMHO.
4
Since you asked nicely ;).
There were generally two reasons - I had tested so much that I knew I wanted certain exact ratios of cards with certain functions in my deck, and secondly, I really wanted as many sideboard slots as possible, so I decided to mainboard some sideboard cards.
I believe I had 4 1-ofs in the maindeck - probe, dismember, spell pierce, and become immense.
For a very long time, I actually had 2 dismember main and 2 side, since I really liked dismember against pod (full 4 copies post side).I still liked it vs twin and abzan and affinity, but it is obviously terrible vs burn,amulet, and scapeshift. Even vs slower versions of zoo I liked it. I also learned before the event that pantheon was on infect. Given my expected meta, I knew I wanted 2 dismember in my 75, but felt it was good enough vs the expected field that it was fine to maindeck one copy to save a sideboard slot.
Become immense I tested a lot as both a 1-of and as a 2-of. I felt the PT meta was going to be slightly more combo heavy than it was, were might of old krosa is usually better. Obviously BI is worse in multiples, so I went with the 'safe' option of a single copy.
The miser's spell pierce has always been good to me. Its definitely not something you want to draw multiple of, but it is devastating if it works. Also, casting spell pierce g1 is great since if you get them with it, they will play around it all match not knowing your a madman with exactly 1 spell pierce. Also, I knew I wanted at least 2 pierce (and possibly a dispel if I could fit it), so maindecking a copy saves me another sideboard slot. Also note that the above 3 cards are typically bad in multiples, Id almost always rather have 1 dismember and 1 spell pierce than 2 of either.
The 1-of probe is because I feel this deck works best with 20.3 lands, but I cant play 20.3 lands, so I play 20 lands and a probe. Probe is actually a really great card, but becomes less important once you have a ton of experience with the deck. Your lifetotal really matters, and Im already maindecking bad cards vs burn (dismember), so I really cant afford to play many probes. Finally, its also a card thats bad in multiples, just like all my other 1-ofs
Also, at the risk of sounding arrogant, I planned on top 8ing (I plan on winning every tournament I enter, thats just the competitive mindset I have) - I really felt like my modern prep was outstanding, and I got some very good draft advice the day or two before the PT. Imagine your my opponent and have my list - its much harder to play around all these random 1-ofs than if I just used 3 spell pierce for example.
On the sideboard:
3 nature's claim/1 hurkyl's recall: The recall was my 15th sideboard card, going into the PT, Id planned on having a 3rd relic, but changed my mind to wanting a dispel, then I tried to find fog, but the vendors didnt have it, so I settled on recall. Now, the only matchup where Id want all 4 claims is affinity, and if thats true, then I feel like hurkyl's recall is better than claim vs affinity (at least the first copy, Id probably rather have 2 claim than 2 recall, but id rather have 1 of each than 2 claim). 3 claims is needed vs burn, so I didnt seriously consider fewer copies. Its also nice vs affinity, and in general is very solid vs an open field, nature's claim is simply a very versatile card
1 dismember - as before, I already have 1 main, and id like a second vs the mirror, abzan, twin, etc. Dismember is another card thats good to have as an option - for example, I like a couple copies vs storm, as I find its very difficult for them to race me without electromancer.
1 spell pierce - pierce stops things like blood moon and living end and wrath (not verdict, but people often respect thrun). You really want at least 2 pieces of countermagic in infect, if not more
1 dryad arbor - amazing vs abzan, 'ok' vs non-combo. This card is just all around reasonable, although drawing it sucks. I probably shoudlve just maindecked it - then id get another sb slot!
2x relic - as mentioned previously, Id considered a third relic. You never know when you will face dredge or living end (it was key in both my wins vs living end!) or gifts -> rites. Also, Against abzan I liked it, and I also like it vs things like temur twin. Liking relic so much is another reason I didnt overload on become immense
1x spellskite - great vs boggles and burn - ok in the mirror, either great of mediocre vs twin (bad vs grudge)
1x wall of roots - I really wanted a card specifically for burn, and this was the best I could come up with. Spellskite, while great, can be hard to use since you have to shock for blue mana, and they usually have smash to smithereens post side. Generally speaking, Id rather turn 2 wall than spellskite vs burn, whereas spellskite it much better later, and for ths reason I went with a 3/1 split in the 75
2x twisted image : Once again, amazing in the mirror, and I knew pantheon was on infect (and I was legitimately afraid of their 3x twisted image). I also bring it in in a ton of matches since the upside is so high and the downside is so low, especially once you factor in wild defiance
2x viridian corrupter: Since I skimped on become immense, my list is a bit weaker vs chalice. I like having at least 4 answers to chalice in my deck, and once again - I like this card in the mirror, since it will often trade with a guy and a pump spell, and furthermore, can eat opposing skites. I also like corrupter in matches where I dont like skite, but have artifacts that arent critical to kill, but that I incidentally want to kill, for example, vs a deck like merfolk (aether vial) or hatebears (vial, splicer tokens). When corrupter is good - its often really good, and I like having this type of high impact card in my side.
Hope that helps
1
You repeatedly make bold claims in this thread (infect is a glass cannon, infect is weak to uwr control, etc), that many experienced infect players disagree with. You claim dailies are soft. Im curious as to exactly what experience you are basing your assertion that UWR is a bad matchup off of. Do you regularly play against Pro Tour/GP top 8 caliber players, and have significant testing experience against them?
So, youve seen people online make budget choices in a daily. Thus, you discount daily results? This is pretty flawed reasoning. The majority of online players are more skilled than the majority of paper players, because playing more makes you better at this game and online players have that opportunity. All this is coming from a paper player.
drinkard claims significant testing experience shows UWR is not a bad matchup. This does not sway you.
My personal experience. At 5-1 at my last PTQ, I played vs UWR control and won. I drew into top 8, then played UWR again (this time winning 2-0). At my local store, I regularly play against Kyle Boggemes who plays UWR control, and we agree the matchup is very close to even, possibly slightly favoring infect. Kyle has won a recent GP playing a control deck (albeit in standard) and made the finals of a Pro Tour, so hopefully that is sufficient proof he knows what he is talking about. I think my match record vs Kyle infect vs uwr is either even or im up a match lifetime.
So, Ill admit, if your opponent is very skilled, its tough. The thing is, Kyle has played me a bunch of times, and knows my deck. Even very good uwr pilots probably arent used to playing such a fringe matchup. However, even if your opponent is very skilled and knows the matchup, I think its a coinflip matchup at worst.
Overall, my record vs UWR is probably closer to 70% win. Obviously the matchup is not so lopsided, but I think an equally skilled infect player vs an equally skilled UWR control player should favor infect, given that UWR is the more popular deck, so the infect player is presumably more experienced in that matchup.
Note: all this applies for my particular list. Different infect lists may have very different UWR control matchups.
1
My experience is that infect is perfectly capable of winning long grindy games against grindy decks (UWR control and Jund). I am 58-12 in matches with this deck at my LGS, and discounting intentional draws I am 13-4 in matches at the PTQ level. If my chances of winning dropped to 'near 0' after 4 turns, Im sure my win % would be far less than 50%. Going all-in early in the game is often a recipe for disaster in many matchups. My matchups against UWR control (considering kiki-resto UWR as a separate archetype) is 2-0 at the PTQ level, and against jund at PTQs its also 2-0. I only remember 1 game of those 4 matches were I won on or before turn 4.
1
Matches were
Round 1: Green Devotion (black+red splashes). Win (2-1)
Round 2: Melira Pod. Win (2-1)
Round 3: Melira Pod. Win (2-1)
Round 4: RUG delver. Lose (1-2)
Round 5: UR Delver. Win (2-1)
Round 6: Affinity. Win (2-1)
Round 7: UWR control. Win (2-1)
Round 8: ID into top 8 (I had best breakers)
Top 8: UWR control. Win (2-0)
Top 4: U/R Twin. Lose (1-2)
None of my matches were particularly easy, as seen by the fact that every single match I played out in the swiss went to game 3. My pod opponents kept naturally drawing their 1 of melira, and I never drew my 2-of dismember, which made things difficult. I had to 'mindslaver' an SCG circuit top15 ranked player to win my third match in ridiculous fashion.
Top 4, I won game 1. Game 2 I nature's claimed a turn 3 blood moon, and then resolved a wild defiance turn 3 with inkmoth 2x, and breeding pool. My opponent spreading seas'd my only green source, and I didnt topdeck a second copy for the win. Its very difficult to lose with wild defiance and 0 pressure vs u/r from that situation (barring blood moon #2) so it was a rough loss. The third game I mulled to a very weak 6 and decided to risk it, and was easily crushed.
1
I consider twin a challenging match-up thats about 50/50. Postboard I typically have 4 dismember, a couple nature's claim, a torpor orb and a spell pierce. This is a matchup where UG is probably a lot better than monogreen, but for a not so obvious reason - fetchlands. If you leave a fetchland untapped at EOT, and they go to tap it down, you can sac in response and keep all your 1 mana answers available.
If you have a twin heavy meta, I would suggest adding fetches to your monogreen list if you have them.
That BWR goodstuff deck is going to be very hard to beat for monogreen (or any ordinary G-based infect list). I would simply try my best to use vines/blessing to keep a rancored up myr alive and stuff it full of pump spells. If you run hierarch in your monogreen list it will help vs lili, and allow you to main or side a couple wild defiance, which can help vs all the burn.
Also, just curious, how do you get 13th out of 60 players with a 2-4 record?
1
Part I: Pro Tour Meta + Why Statistics can be misleading
Thesis: The PT metagame analysis and coverage on the official site is misleading when it comes to figuring out the best modern decks.
When the Pro Tour was over, I figured I'd look at the posted coverage and poke around some articles online to see which decks had performed the best. When I say 'the best' Im not talking about a deck that happened to spike, but if overall, players of that deck did well.
To me, the definition of best is simple: how 'good' a deck is is determined by its match win percentage.
For reference, my comments are in regard to charts in this link: http://www.wizards.com/magic/magazine/Article.aspx?x=mtg/daily/eventcoverage/ptbng14/metagame_day2
The first graph shows how many of each deck is in day 1, and how many are in day 2. This chart is obviously confounded by the 3 rounds of limited, and to be fair to Nate Price he points that out. The real problem is, that the chart does not include the total number of players to make day 2, and the total number of players. For reference, 252 players day2'd out of 393 players total, so you would expect an 'average' deck to send 64% of its players to day 2. With that in mind, you would think
splinter twin was below average and zoo was right about average by looking at that chart... and of course you would be wrong. There are several reasons for this, not the least of which is that the day 2 meta is different from the day 1 meta.
In terms of placing well in the tournament, its important to not only make day 2, but make day 2 with a good record. At a grand prix, you need 7-2 to day 2, and so a player at 8-1 isnt that far ahead of you, so pretty much everyone that day 2's has a shot at top8. This is not the case at the Pro Tour, where a large number of players day 2 with a 4-4 record and essentially no shot at top 8 (you might get 7th or 8th if you are 12-4 but odds are if you start at 4-4 your breakers are
terrible).
I assume that is why Nate Price made a graph showing deck win percentage - given that they had already made day 2. When you first read the percentages, they are confusing, it seems like all the decks are above average! Even the worst deck, tron, is 48% to win! Its important to remember, of course these percentages are over 50%, he is only talking about the percentages of the winning people! Now, lets say a deck is completely average. It wins 50% of its games. What win percentage should you expect, if you look at the people who finish 4-4 or better with that deck after 8 games? It turns out the answer is 60.7%. However, even with that knowledge, we can really only say a deck is 'good' if it has above this win percentage AND it has ~64%+ players make day 2 with it. If only one of those criteria is true, its a bit tricky to see if the deck is winning more than half its games.
So, what decks were good? Well, the UWR control/midrange strategies were lumped together, but they had good day 2 conversion and good match win %, so they overperformed. Clearly storm is also a winner, as is living end. Zoo is an example of a clear loser when you consider that average is 60% not 50%, when you only look at the win percentage of the 'winners'.Now, we dont have the data for what everyone was playing, but wizards did post the data for everyone that was 6-4 or better in modern. Now, they didnt post exactly how good each deck did, i.e. points, but you can manually look that up if you want. So, how can we decide if a deck was good, if we dont have match win % and only have partial deck information? Well, one way is to look at the total number of pilots who start with a deck, and the total number of pilots that ended up doing well with that deck. Technically that information is available on wizards site. What you can do is, use the search function, and it will tell you all the instances of a word.
Splinter Twin 14
Birthing Pod 13
Cranial Plating 5
Pyromancer Ascension 7
Living End 7
Wild Nacatl 12
Gladecover Scout 9
Mana Leak 18 (and 10 of those are UWR control/midrange)
Scapeshift 4
Rift Bolt 4
So, we see that every deck besides UWR can be reasonably identified by a single card. Now lets look at percentages (i.e. normalize the number of winning lists vs the number of total people running that deck to approximate win percentage in a crude way, we find.
Twin: 14/45 0.311
Pod: 13/43 0.302
Affinity: 5/22 0.227
Storm: 7/12 0.583
Living End 7/14 0.500
Zoo 12/64 0.188
Boggles 9/24 0.375
UWR 10/29 0.345
Scapeshift 4/13 0.307
Burn 4/15 0.267
Expected: 111/393 0.282
Ok... So is that it? Well, not quite. Due to the small sample size, these results alone could be a bit misleading, what about if you consider only decks 7 wins or better?
Twin: 10/45 0.222
Pod: 4/43 0.093
Affinity: 4/22 0.18
Storm: 4/12 0.333
Living End 2/14 0.143
Zoo 8/64 0.125
Boggles 5/24 0.208
UWR 8/29 0.276
Scapeshift 1/13 0.077
Burn 2/15 0.133
Expected 61/393 0.155
So, some decks seriously outperformed (comparatively speaking) when compared to when 18 point decks were included (affinity, UWR, Twin), while others seriously underperformed (Pod, Living End, Scapeshift). This is possibly due to there being a different metagame in day 2 than day 1 (so some decks won a bunch day 1, then stopped winning at all day 2, see Hetrick on Living End) but I am not certain. A large part of this could also be due simply to varience in a small sample size.
So, who is our winner? Its easy to say storm, but you have to keep in mind that its dangerous to trust a small sample size, as our assumption of equal play skill breaks down and it starts to depend more on individual pilots (and Kai and Finkel are certainly above average pilots, for example). Even so, a second look still says storm
1) Storm
2) UWR decks
3) Twin
4) Boggles* (very metagame deck, weak to too many top decks)
Now, this is not considering fringe decks, like blue moon or ad naseum, that did not have many pilots. As you can see, I slanted the results towards the 21+ point decks, as those decks are more likely to actually win the money. The loser amoung the 21+ point decks is definitely pod, which only put a fairly miserable 4/43 pilots into 21+ points in the modern portion (the fact the deck made it to the finals means it will still be played a lot though).
A final note on misleading statistics relates to teams. Blake Rasmussen wrote an article, where he concludes the 'best superteam' was CGB classic, edging out CFB pantheon. From a match % perspective, this is wrong. CFBP managed to match CFB in wins, despite only 12/15 of their players making day2 compared to 15/15 from CFB. Counting 'wins' as your metric of success clearly rewards teams that simply put the most players into the second day. Imagine two teams with 'average' players. Team A
has 2 players with 4-4 records at the end of day 1, whereas team B has one player at 2-6 and another at 6-2. They play out day 2, and now team A has two players with 8-8 records, while team B has one player at 2-6 and one player at 10-6. Counting wins only says Team A is far superior (16 wins vs 12 wins), which seems very unfair.
Part II: Futute Meta thoughts
The meta moving forward:
One thing about predicted future metas is that people are irrational, they wont do an in depth analysis of the tournament to predict what deck they should run, and many people cant afford to switch decks. They will, likely, respect online results though, and once those absorb the pro tour tech, many will likely realize that MTGO is pretty good at showing the top decks in eternal formats.One thing is that I think zoo is a fine deck, despite its poor showing at the Pro Tour. What happened was, people over-prepared for zoo, leading to it being bad, and the anti-zoo decks like boggles and living end doing great. Then, once the zoo players were mostly killed, the actual good decks started showing more dominance (UWR/Storm/Twin) but those decks arent necessarily amazing vs zoo, they are better against anti-zoo decks like living end. This is why, the zoo decks that actually made it to a winning record, many converted that to a 7+ win finish. However, I think people enjoy zoo, and its not bad enough to make people not play it, so I think it will continue to show up in large numbers. Boggles will get weaker, as players will pick up the tech that beats it (porphyr (spelling?) nodes, runed halo, maindeck spellskites in twin, etc). Twin is good, and will stay popular. It comes in many shapes and sizes and is very difficult to beat. I do that that relic of progenitus really showed its power vs tarmo/snapcaster style twin, and I expect relic to increase its prevalence both main and side, as its strong against storm, tarmotwin, living end, etc. The winning Pro Tour decklist seems very solid to me, and I think people will adopt its tech.I expect that despite not many players playing it at the Pro Tour, the Black/Green Obliterator deck will pick up supporters. Reid Duke and Matt Costa are very popular players, and actually playing phyrexian obliterator will be appealing to many. Being able to get aggressive against twin by having slaughter pact, and having the discard package against an increasingly combo meta seems like a great place to be. I predict this style deck will try and fill the hole in the meta brought about by the death of jund, and it will become the new jund (but wont be anything close to the old jund).I think the fringe combo decks (ad naseum, amulet, etc) will stay fringe, and the meta decks like blue moon to die away as people get used to playing against them.
People will likely run an extra basic or two for a while out of respect.
I could see a meta developing that is relatively hateful towards Pod decks (rise of anger of the gods and storm). Once players find a reasonable jund/style deck
though, I predict it will be popular and help keep the combo in check. Otherwise, the top decks will stay the same, and it will be business as usual.