oraymw's picture
By: oraymw, Oraymw
Jun 27 2014 12:00pm
5
Login or register to post comments
9759 views


Welcome to Ars Arcanum, the MTGO stats-based limited column. In this installment, I’ll be doing a draft overview of Vintage Masters (VMA) limited. For this article, I’ve watched hundreds of matches of VMA draft on MTGO, gathered several data points from the games, crunched the numbers, and then analyzed the data. As always, I’ll take a look at the speed of the format, the popularity and win rates of color combinations, and the key principles of how to draft the format.

I’ve been very excited to do this article for VMA. I’ve been drafting Magic for a long time (I started in about Tempest), and VMA brings back a lot of my favorite cards to draft once again. It feels like VMA has been a resounding success on MTGO, so I don’t think I’m alone in this. I’ve been particularly excited since VMA is a premium set; the boosters cost nearly twice as much as regular boosters, and the value of the cards in the packs is also quite a bit higher. On top of this, VMA seems to be a particularly difficult format to draft since there are so many interactions in the set. Because of these two factors, I was really excited to put together a draft overview in order to help players get more value out of their VMA packs.

I like to begin by restating my philosophy for these articles for those that are coming here for the first time. The goal is to take questions about draft formats and start to apply real numbers to them. I have long felt that too much about MTG is based on gut feelings, and I have found that my win rate improves dramatically when I begin to apply real data to my game. There is a lot of power in using observation, data, math, and analysis in order to understand a format. I can’t answer every question, but the ones I do address, I address thoroughly. Some players have the benefit of large play groups that can put in extensive hours of team playtesting in order to understand a format. These articles try to put the same kind of tools into the average player’s hands. Most of all, these studies are based on rigor and many hours of effort.

There are many strengths to this approach. These studies are in depth and they provide a lot of detailed information about a few specific elements of the format. The analysis is based on personal observation of hundreds of matches, which means I’m able to provide a more sophisticated view on what the numbers mean. It provides a solid foundation on real data, rather than just gathering the gut feelings of a group of players. This approach also has several weaknesses. By necessity, the sample sizes for these studies cannot be as big as if I used a data-scraper to gather data from MTGO. While the data I use is always statistically significant, it does some things worse than big data when big data is gathered and analyzed correctly. Most importantly, this kind of study does not replace individual experience. I like to think of these articles as a survey of the terrain, it offers perspective, but things often look different when you actually put your foot onto the ground.

This article also came with its own particular challenge. Normally I gather my data from 64-man drafts on MTGO. For the first week or so of VMA, those drafts were firing somewhat regularly, though still at a reduced pace as compared with most set releases. But soon the 64-man drafts stopped firing at all, and I still hadn’t gotten a statistically significant data set. Fortunately, I discovered that I can watch replays for drafts in the release events room in the V4 MTGO client. I then booted up the beta, and finished off the second half of the sample with those numbers. However, 64-man drafts and 8-4 drafts have slightly different artificial effects. 64-man drafts tend to be more competitive and players don’t mess around quite as much, while 8-4 drafts allow players to stretch their wings a little bit more. Also, I normally gather all of my data over the period of about 4 days, but this time the sample came from about 11 days of events, which leads to a slightly different study. I can’t say for sure how that might affect the data, but it is worth mentioning that whenever someone does a statistical study, the context of their data is going to affect their analysis in some way.  With all of that out of the way, it’s time for the data!

Speed

Ending Turn of Games in VMA Draft

Ending Turn of Games in VMA Draft as Compared with GTC

Ending Turn of Games in VMA Draft as Compared with JBT

Since I started doing my Ars Arcanum articles, Gatecrash draft was the fastest draft format that I had analyzed. I started with AVR, but Gatecrash ended up being a hair faster than that format, though not quite as fast as triple Zendikar. However, that position has now been taken up by VMA draft. Just looking at these charts shows a significant difference between these two formats. The ending turn for VMA drafts spike at turns 7 and 8, though about 18% of VMA games end before turn 7. Before I did the VMA article, I had wondered if including wins on turns two and three were being superfluous, but in this format I was able to see a couple of games end on both of those games, as well as a few games ending on turn four.

In the GTC comparison chart, we see that VMA has more games ending on every turn up to turn seven, and that it hangs just under GTC for the rest of the graph. The average ending turn for games in GTC was 9.1, the average ending turn for games in VMA was 8.8. Even more importantly, 64% of GTC games ended by turn 9 or sooner, while 67% of VMA games ended by turn 9 or sooner. All that this means is that this format is really, really fast.

I included the chart for JBT in order to help people connect this format with the most recent format and get a little more perspective. When we looked at JBT, we saw that it was a fairly fast format, that it was very tempo based, and that you really didn’t want to waste your time on seven drops. Comparing VMA with JBT is simply stunning. On the chart, we see a huge difference between the two formats, and in the numbers ,we see that JBT had an ending turn average of 9.564 as compared with VMA’s 8.8. Furthermore, only 58.5% of JBT games ended by turn 9 or sooner as compared with VMA’s 67%.

Turn nine is a particularly important turn, because it is the turn at which we start to expect to deploy our six drops with regularity. In a format where 67% of the games end by turn 9, your will usually not be able to hit your six drops, and when you do, they will usually be on the board just in time to get overwhelmed by a swarm of attackers, and then you lose the game. A card like Desert Twister would be a reasonably high pick in a format like JBT, but in VMA you just can’t reasonably count on being able to cast it. Cards like Killer Whale or Jungle Wurm would be reasonable cards in other formats, but in VMA they are bottom of the pack chaff. Game speed is key to understanding a format, since it affects every decision you make in the draft. It affects what cards you play, and it affects what strategies you deploy in game. Vintage Masters is a format where you cannot afford to mess around. You need to have a game plan to deploy as early as possible, and you can’t afford to waste your picks on things that are only relevant in the late game. People that played Zendikar or Gatecrash should have a reasonable understanding of what this means; you simply cannot afford to durdle.

There are a few factors that make the format so fast. One is that there is a very fast red deck in the format with Goblins. The Goblin deck is able to deploy a high number of threats early in the game, and then back those threats up with Goblin General and clear out blockers with Chain Lightning. There are also a lot of powerful auras like Elephant Guide and Empyrial Armor that force you to have answers immediately or you will just lose the game. There are also a lot of threats with a high relative power to mana cost ratio, such as Wild Mongrel, Aquamoeba, or Arrogant Wurm. There are a lot of powerful tempo plays to back these aggressive decks up like Man-o’-War, Paralyze, Predatory Nightstalker, or Repel to make sure that you can keep getting in with those powerful common creatures. There are also a couple of powerful token generators with Beetleback Chief and Battle Screech at common that give the aggro decks a critical mass of creatures that keeps an opponent from ever getting back in the game. There are also several shadow creatures that give you cheap evasion, and make it difficult for attrition decks to stabilize. There is even an arguably viable storm deck that can win the game early on if it isn’t disrupted to heavily. Finally, the inclusion of very powerful rares, including the Power 9, leads to games that close very quickly. A game with Sol Ring or a Mox is going to be over very quickly, on top of all the common factors that are increasing the speed.

Whenever I tell people that a format is fast, they automatically assume that this means that the fastest deck in the format is going to be the best deck, but that is usually not the case. They key is to build a deck that correctly responds to the metagame; a deck that is well-equipped to handle a lot of fast decks is going to do very well in the format, and there are plenty of tools to control the game in VMA, but it is important for your control decks to include tools that are both fast and efficient. You can’t just throw in a bunch of Deep Analyses and say that you are a control deck, even though that card is very powerful. If you build a deck that wants to get to the late game, but that is unable to get to do that, you are going to find yourself very frustrated.

In the next section, we’ll look at the popularity of decks in the format, and see how that is affecting the format.

Popularity

Popularity of Two Color Combinations in VMA Draft

Popularity of Colors in VMA Draft

Relative Color Popularity Intensity of Two Color Combinations in VMA Draft

In these charts, we see the popularity of the different decks in the format. In the first chart, we see the ten two-color combinations and their popularity given as a percentage of the format as a whole, in the second chart, we see the percentage of decks that are running each color as one of their main colors, and in the third chart we see how intensely the colors are being drafted in each of the two-color combinations.

There are a few things that are very clear right off the bat from these charts. First, people really like blue. UB and UG are the most heavily drafted archetypes, with RW coming in third. Secondly, people are avoiding red. One of the three most popular decks is red, but the other three two-color combinations that contain red are the three least drafted decks. Part of this is because of the relative popularity of the Mono Red deck. I usually don’t include mono color decks on this chart because they don’t get a large enough representation for statistical significance, but in the case of VMA, the mono red deck has a lot of popularity, more even than UR. This does throw off the numbers a little, since the mono red deck leeches more red picks out of the draft than a two-color combination that includes red. The truth is, we don’t know whether the underdrafting of red is what makes it possible for people to draft mono red, or whether people drafting the mono red deck are stealing too many cards from the other red decks for them to be viable.

There are a few other curious things in these charts. Black doesn’t immediately appear to be all that popular, since it is only really in the UB deck and it has a reasonable representation in the WB deck. Both Golgari and Rakdos combined don’t even have the popularity of Dimir by itself. However, we see the effect of having just a few decks dominate the popularity of the format so dramatically, since black is still the second most heavily drafted color. People really want to draft UB in this format, and that data is skewing everything else in the study.

It’s also amazing the effect that having two incredibly popular blue decks has on the field. Izzet is drafted so rarely that it almost dangles off the end of the chart, and the Azorius deck is the fourth least popular white deck. All of this is because blue is being drafted so heavily into UB and UG. These two decks together make up about 1/3rd of the format overall, and about 4 in 9 decks will feature blue as one of their main colors.

However, there is an element here that is a little bit deceptive, which is the impact of UB Storm on the format. When the designers of VMA talked about the format, they mentioned that UB Storm would be a viable strategy if you managed to get the right combination of cards, but they also said that it wouldn’t come together all that often. Apparently MTGO players did not get the message. In this study, I added a Storm tag to all of the UB Storm decks, and a little more than half of the UB decks are dedicated Storm lists. People really want to play storm, and it has particularly been popularized by streamers over the past year, so it really shouldn’t come as much of a surprise that it is getting drafted so heavily. For reference, UB Storm by itself clocks in at 8% of the overall field, which means that it is the sixth most popular deck by itself, and it is more popular than WU, BG, RB, RG, Mono R, or UR. In fact, the UB Storm deck is drafted more than twice as heavily as the UR deck.

The popularity of the next two decks reflects the current opinion of the best decks of the format. UG comes in with a tie in popularity with UB, and RW comes in at a somewhat distant third place. The UG deck is built around madness, but it is also built around putting out powerful aggressive creatures and then backing up those creatures with tempo spells. The RW deck is something else entirely, but I’m going to save some of the discussion of that color combination for later in the article.

But first, let’s take a look at the win rates of these color archetypes.

 Colors and Archetypes

Win Rates of Two-Color Combinations in VMA Draft

Win Rates of Colors in VMA Draft

Here we start to see a few disturbing things. There are three things that should stick out immediately. First, RW is clearly dominating the field, with about a 10 point lead on the next closest deck. Secondly, the UB deck performed terribly at just under a 40% win rate. Third, the rest of the decks are all in pretty tight competition. There is a fourth very important, but perhaps less obvious thing, which is the disparity between UG and UB. Both decks were drafted at about the same popularity, but UG still managed to put up about a 51% win rate.

I’m going to talk about three things in this section. First, I’m going to talk about whether RW or UG is the best deck in the format. Second, I’ll talk about why UB is performing so terribly. Third, I’m going to talk about how to go about drafting the most effective deck in the format.

If we just look at the win rates, it appears that RW has UG beat by a lot. It has about a 12 point lead on UG, which is a huge difference when we are  talking about the win rates of draft decks. However, these numbers are a little bit deceptive, given what we know about the popularity of the decks in the format. UG was tied as the most popular deck in the format, but it also was at the top in a margin that we don’t normally see in regular draft environments. This deck is getting drafted very heavily, and that is surely affecting its win rate. With that said, the popularity of the two decks is still close enough together that I would normally easily call the race in favor of RW. The problem is that the other most popular deck in the format is UB. Between these two decks, blue cards are getting leeched out of the packs incredibly quickly. It really is impossible to determine which of these two decks would have the higher win rate when one of them is being overdrafted by such crazy amounts. Right now, I’m still inclined to give the RW deck the edge based on what I’ve seen after watching about 1000 games, and it certainly is a better deck to draft in the current metagame, but the UG Madness deck is so strong that if you see that it is open, you should jump on it immediately.

And now, we need to talk about UB. Earlier, I mentioned that there were basically two UB decks, the UB Storm deck and the UB Tempo deck. There also isn’t really a ton of overlap between the two decks. The Storm deck accounts for about half of the UB decks overall, which seems really high considering that the designers of the set didn’t think it was going to be a consistent strategy. Things get uglier when we look at the win rate for the two decks. The UB Storm deck came in with a win rate of 28%, while the UB Tempo deck came in with a win rate of 54%.

Here is the first important piece of advice that I can give you about VMA. Stop drafting Storm. The thing is, people love this deck, and they have an emotional attachment to it. They go into VMA looking to force Storm. When they see a Dark Ritual, a Cabal Ritual, a Brain Freeze, or a Tendrils of Agony, they get this look in their eyes that just screams “I want to spew value!!!” Do not give in to this. Forcing the Storm deck is the single best way to hemorrhage value in VMA drafts. This is perhaps the perfect example of my argument for efficacy vs. power. RW and UG are examples of effective decks; they do the thing they are supposed to do, and even though they may not be as flashy as the Storm deck, they will always be able to compete. The UB Storm deck is the perfect example of a deck that skews toward power; there are Storm decks that win on (technically) turn 2. I’ve even seen it once (aided by Time Walk). When you draft a really good Storm deck, you feel incredibly powerful because your opponent just can’t interact with your game plan. If they don’t kill you fast, you are just going to dump everything and win the game and they won’t be able to do anything about it. But remember, though a great Storm deck is going to steamroll people, a bad Storm deck is going to be completely terrible.

Here’s an example, the other day I was drafted with my friend David Stewart, and we had a decent UG madness deck that splashed white for Armadillo Cloak. Round one, we faced a Storm deck. David pulled up his first hand, and it contained no lands. Obvious mulligan. We got back the next hand and it had all lands, so we had to mulligan again. The next hand had no lands again, so we tossed it back. The four card hand also had no lands, so back it went. The three card hand was hilarious. We had two Obsessive Search and the plains we were using to splash our one Armadillo Cloak. The poetic justice of this hand was great, and we had a good laugh and consigned ourselves to lose, but David didn’t concede. We soon realized that we were playing against a Storm deck (since he played no creatures except a Nightscape Familiar.) After four turns, we finally got an Island and cycled away our Obsessive Searches. And here’s the thing, our opponent’s deck just a mediocre Storm deck. It’s not that it was terrible, it just wasn’t great. And you know what? David steamrolled him. On turn 12, the Storm deck had to try to go off because we’d gotten out one creature he couldn’t deal with, and he didn’t find any of his kill cards, and we won the game without any problem. In game 2, we got off to a great start, and he was forced to try to go off on turn 6. This time he had the Brain Freeze, but he was 6 cards short of milling us to death, and we won. The thing is, if your game plan can’t consistently win against a mulligan to three on the play that has only the splash color land, then your deck is not a good deck.

This is what I saw over and over again as I watched this format play out; bad storm decks facing off against halfway decent aggressive decks. I did not keep track of all the times that the storm player was forced to go off a turn early and ended up one or two spells short of the kill, but it was a lot; far more than the games the Storm players won. There are three things that are just killing the Storm decks. First, there just isn’t that much support for the deck in the cards. It is possible to draft a good UB Storm deck, but it is certainly not likely. Second, way too many people are forcing this deck. It is already thin on support, and splitting those cards up among too many people is the fastest way to kill of the deck. Third, the format is just too fast. Sure, the Storm deck can occasionally win on turns 2 through 4, but it is very bad at putting up early game defenses, and it regularly loses on turns 5 or 6. When you are facing aggro decks that can goldfish that quickly, you have to have a very good Storm deck to not just lose. Because the format is so fast, the Storm deck just doesn’t have enough time to build up the right hand, and it is forced to go off a turn too early, and then it just loses. If the format were a turn slower, the UB deck might be really good, or if people would just stop drafting it so heavily, but until then, just avoid the Storm deck. Unless you open up Yawgmoth’s Will, and then you had better draft it or I will never forgive you.

However, this does not mean that UB is out of the question. In fact, if we take out the storm decks entirely, we’ll see that UB actually performs very well, with a 54% win rate, which makes it the second most competitive deck in the format. This is a very different kind of deck; it is an aggressive UB tempo deck. The idea in the deck is to land a high powered creature in the early game, such as a Sarcomancy or Carnophage, and then use tempo spells to keep your opponent’s off balance, as well as disruption spells to keep your opponent from being able to respond. Cards like Man-o’-War and Predatory Nightstalker are very good in this deck since they allow you to add pressure to the board while also keeping your opponent from blocking. Another card that doesn’t get played nearly enough in this style of deck is Dark Ritual; people think that it is a storm card, but it is also very effective in the UB tempo deck by dumping out a few extra threats early on, and then backing them up with tempo and counterspells. A great example of this is playing Dark Ritual to put out a one drop and an Ophidian on turn one, and curving into Man-o’-war. I’m not going to go into too much detail on how to draft this deck, since I need to cover RW in more detail, but I will point you in the direction of Marshall Sutcliffe of Limited Resources fame, because he’s been an advocate for this deck since VMA was released, and he’s absolutely right that this deck is well suited for the metagame.

Finally, I’m going to tell you about how to draft RW.

Win Rate by Turn for RW in VMA

In this graph, you see the win rate for RW by turn, and it is the perfect way to begin a discussion about the archetype. The thing is, there are two different RW decks, and both of them are very powerful. As this chart shows, the RW deck is able to maintain a 50% win rate or better throughout the entire game; there isn’t a point in the game where it isn’t competitive, and I think that this is the reason why RW is the best deck in the format. It always has game. But, this graph demonstrates something important about RW, which is that it is basically two different decks. We see that there is a fast RW deck that competes really well in the early game, and we see that there is a slower, more controlling RW deck that is able to take over the late game.

These two archetypes both revolve around a pair of cards. The aggressive deck is a tokens deck, and it hinges around Beetleback Chief and Battle Screech. The idea with this deck is that you play out some early creatures, knock down your opponent’s life total, and then play one of these token generators in order to create a critical mass of creatures that swarm over your opponent for the win. Battle Screech is definitely the better of these two cards, since your tokens have evasion, but Beetleback Chief also does a great job of taking over the board. Meanwhile, there is an entire other RW deck in existence, that is based around two uncommons; Lightning Rift and Astral Slide, as well as a pile of cycling cards. This deck is the most controlling deck in the format, and it is actually quite well suited for the format. Lightning Rift is very good at picking off small aggressive creatures, while Astral Slide allows you to play a lot of creatures with ETB effects that can stabilize a game, and then reuse their effects, or alternatively you can Slide out their enchanted creatures, which allows you to completely dismantle your opponent’s game plan. The key with these decks is to build up a good early game defense so you don’t get overrun by the aggressive decks, and also play a lot of cyclers in order to dig to your late game cards. The biggest problem with these decks is that they don’t have many good win conditions, so it can sometimes be a struggle to close out a game. The one deck you fear with the Slide decks is the Storm deck (ironically) since cycling goes really far into your deck, and it is really hard to beat Brain Freeze if you don’t happen to have a Gilded Light in hand.

But here is the greatest part about the two decks; they are highly synergistic together. The two highest picks in both decks are Flametongue Kavu and Goblin Trenches. FTK is absolutely the best uncommon in the set, but it is even better in an Astral Slide deck, where every cycling card becomes a cantripping Flame Slash. In the RW aggro deck, it takes out a blocker while adding a high powered creature to the board, and the tempo you get from FTK is just amazing. Goblin Trenches is better in the aggro deck, where it turns into the best possible late game win condition, allowing you do turn all your lands into two 1/1 goblins, and then just attack for the win, but it is also useful in the RW Slide decks. Since your biggest problem is closing out the late game, Goblin Trenches provides the perfect solution. Obviously you’ll take Slide or Rift over the Trenches in packs 2 or 3 if you are already in those decks, but Goblin Trenches is a perfect card to pick up earlier. Additionally, Beetleback Chief and Battle Screech are also perfect for both versions of the deck. The Beetleback Chief stabilizes the game and interacts well with Astral Slide, which also gives you a great win condition, and Battle Screech does something similar, though with a higher power level but lower synergy.

The synergies between the two decks go even further. For example, the Slide deck really wants early game defense so that it can take over the late game. Cards like Benalish Trapper, Chain Lightning, Kindle, and Phantom Nomad are high picks in a typical RW aggro deck, but they are also very good at keeping you alive in the early game. On the other side, the RW deck really wants to just keep its velocity going in the late game. Cycling lands are better in the Cycling deck, but they allow the aggro deck to draw into more threats in the late game. A deck that naturally picks up 5 or 6 cyclers can even play a Lightning Rift for late game reach. Solar Blast is another example of a card that plays very well in both decks. And the two token generators are superb in both decks.

This is the deck that I’m looking for every time I go into VMA. If UG Madness is open, or if I need to go into UB, I’ll do that, but I’ve always got my eye out of Battle Screeches, FTKs, Goblin Trenches, Lightning Rifts, or Astral Slides in my early picks.

Conclusions

Thanks again for reading! Here are my conclusions and recommendations for the format.

  1. Vintage Masters is the fastest format that I have ever studied on Ars Arcanum. It is vital to base your deck around early game interaction.
  2. Every color has a lot of ways to punish slow draws, so it is important that the speed of the format affect every decision you make.
  3. Furthermore, every deck has ways to pull out a win in the late game, so you can’t assume that you will win once you’ve established board control.
  4. Stop drafting Storm. Don’t draft it unless people stop overdrafting it.
  5. UB Tempo is a very powerful deck that makes much better use of the cards in the two colors.
  6. RW decks are the best in the format, but it is effectively two separate decks, an aggro deck and a control deck.
  7. However, both of these decks manage to have a lot of cross synergies that allows you to build powerful hybrid versions of either deck.

As always, you can follow me on twitter @oraymw for updates about articles. I’ve also put up a Tumblr account at http://oraymw.tumblr.com/ where I post links to my articles. You can go there and subscribe to the RSS feed, and then you’ll be able to get updates whenever a new article goes live.

Finally, I encourage you to check out the podcast that I do with my buddy Zach Orts, which is called All in the Telling. In it, we look at stories from a professional standpoint in order to get a better understanding of why they are important to the human experience. But mostly, we just talk about what makes awesome stories awesome.

Ars Arcanum Archive

 

 

3 Comments

Nice by doc_brietz at Fri, 06/27/2014 - 17:03
doc_brietz's picture
5

Must have taken forever to write and pull all this up. Superb Article. Very well written and informative.

:D It takes a while, yes. by oraymw at Fri, 06/27/2014 - 17:06
oraymw's picture

:D It takes a while, yes.

As someone who lost a ton of by Doctor Anime at Fri, 06/27/2014 - 17:27
Doctor Anime's picture
5

As someone who lost a ton of money trying to force Storm over and over, thank you for silencing the voice in my head that kept telling me to just try one more time with it and maybe it'll work. 28% win rate, yeesh.

Yeah, RW Cycle is absurdly good. Astral Slide and Lightning Rift are insane. Battle Screech is the best common in the set. And having a deck based around cycling makes things very, very consistent.