oraymw's picture
By: oraymw, Matthew Watkins
Mar 07 2014 1:00pm
Login or register to post comments


Welcome to Ars Arcanum, the MTGO stats-based limited column. For this installment I have the privilege of bringing you the Born of the Gods draft overview. These are the exciting ones; the articles where I watch hundreds of matches on MTGO, put together data from those articles, and then analyze the numbers. Although these are exhausting to put together, they are my favorite articles, because they provide such a foundation for approaching a draft environment. We’ll cover a lot of different topics in this article, including the speed of the format, the popularity of different color combinations, the win rates of deck archetypes, and some points on how to maximize those archetypes. Moreover, this data should help us pinpoint a handful of key principles to help us understand the BNG/THS/THS draft format. During this article, I make frequent references to the similar study that I wrote about in my Theros Draft Overview, and being able to pull up that article would be pretty helpful to making sense of this information.

Before beginning, I usually like to take a moment to explain the philosophy of these articles. My goal with these articles is to look at questions about draft formats and use real numbers in order to start answering those questions. As important as it is to read about the gut feelings of great players, or to look at the ways that a variety of people are interpreting their own experience with a format, I have found that there is a lot of power in making use observation, data, and math in order to understand a format. This article isn’t born out of a few drafts; it comes from many hours of painstaking work in making and recording observations, and then crunching the data from those observations in ways that can help us find answers to our questions.

This approach has a lot of strengths. These studies are always very in-depth, and they provide some pretty useful information on a few parts of the format. Because I gather all of the data through personal observation, I’m able to do more sophisticated analysis of the data, and I gain a lot of insight based on seeing so many games. It also provides a solid foundation that is based on real data, rather than the kinds of wishy-washy advice that players tend to give based on their own more limited experiences. With that said, there are also many limitations to this method. By necessity, the sample sizes for these studies cannot be as massive as if I used a program to scrape data from MTGO. While I only use data that is statistically significant, there is also a lot of value in using big data, if people know how to mine the data correctly (although they usually don’t). These studies also only cover a few events from MTGO; specifically, they are taken from several 64-man events over a one week period. This causes two limitations. First, all different kinds of events introduce their own kinds of artificial effects on the draft, and second, the data covers a specific slice of time. Usually, the information you can get from these articles is much less accurate in as much as a week after the article is released. The main principles will often still hold, but the particulars of the data will shift, sometimes in big ways.

But enough talk. Let’s get to the data!


Ending Turn of Born of the Gods Draft Games

Ending Turn of BTT Draft Games as Compared with Average Draft Environment

Speed of BTT Draft Games as Compared with x3 Theros Drafts

When Born of the Gods first came out, a lot of players were predicting that the format would accelerate. It seemed like everything was in place for that to happen. In my Born of the Gods Spoiler Analysis, I pointed out that there were a lot of indicators that seemed to show us that the format would be speeding up, but that they were subtly deceptive for various different reasons. It turns out that, as I predicted, the format seems to have slowed down just a hair with the inclusion of Born of the Gods. We see that the majority of draft games in BTT end between turns 7 through 10, which is interestingly a little less than a turn slower than our average draft data. It definitely doesn’t clock in as a slow format, but BTT definitely gives all kinds of different deck archetypes a chance to shine. The major difference between BTT and an average draft format is that you spend a little bit of time in the early game building up threats. This makes the development stage of the game about half a turn longer, but the rest of the really doesn’t drag out very much, and games tend to end quickly once someone has built up a fairly powerful threat, or dealt with a large threat. The chart that compares BTT with an average format shows this clearly; we see that BTT has fewer games that end on 6 and 7, but more games that end on 9 and 10. But once we get to about turn 11, both BTT and the AVG chart kind of lock in at about the same numbers through the rest of the chart. The development stage of the game takes a teensy bit longer, but once a deck in BTT is ready to close out the game, it is very difficult to stop it. These aren’t the long, drawn-out, and attrition based matchups of a format like IPA or M14, but instead it is similar in many ways to something like Rise of the Eldrazi; a battlecruiser game where things are built up for a few turns, and then everything comes crashing to a thrilling conclusion.

The chart showing the comparison between BTT and THS drafts is particularly interesting. We see that BTT is a hair slower than THS, though almost imperceptibly so. In fact, BTT is only about .04 turns slower than THS on average. But we do see three significant differences between the two formats. First, BTT has fewer games that end on turn 7. Second, it has fewer games that end on turn 9. Third, it has a few more games that end on turns 10 through 13. There are a few reasons for these differences. One is that the format has fewer ordeals, which means that there are fewer lightning quick games that end because one player landed an early ordeal, and the other player couldn’t handle it. But the biggest contributing factor is an interesting structural difference. In Theros, the Bestow cards all followed a particular format. The common bestow creatures granted +2/+2, an ability, and the Bestow cost was 5 or 6 mana. The uncommon bestow creatures granted +3/+3, an ability, and the Bestow cost was 6 or 7 mana. These creatures warped the format in a tremendous way. When you could reach 5 or 6 mana, you could start bestowing your common creatures, which would often end the game within a turn or two if your opponent couldn’t deal with it, and we see that in the high number of games ending on turn 7. When you got to 6 or 7 mana, you could start bestowing the uncommon creatures, which would end the game in even faster fashion, and the turn when you would often reach that amount of mana was on turn 9.

But BNG’s Bestow creatures are much different. First, the common creatures don’t grant any abilities. Losing things like Nimbus Naiad makes a huge difference on the format, because you can’t just slap it on a big creature and crush through unimpeded. Nyxborn Wolf and Nyxborn Triton are still formidable cards, but on a ground creature, they can still be blocked. Spearpoint Oread, Cavern Lampad, and even Observant Alseid had similar kinds of tempo impact on the game, and BNG just doesn’t replicate that. The uncommon bestow creatures are even more different. While their abilities tend to be very powerful in BNG, they also grant significantly smaller bonuses, from +3/+3 all the way down to +1/+1. This means that the creatures don’t end the game in such spectacularly quick fashion as the Emissaries did in triple THS. The basic structure of the format didn’t change tremendously, but BNG is loses the three most swingy cycles from THS and replaces them with slightly more grindy types of cards. Since BNG is only one pack, the change isn’t overwhelming, but it is big enough to notice a slight tick towards a slower format.

The instincts of a lot of players when they hear that a format has slowed down a little is that they start picking expensive cards more highly. That would be a tremendous mistake in BTT. Firstly, we can see a problem just by looking at the BTT vs. THS comparison chart. The games are still ending pretty early, it’s just that the focus has shifted slightly to the right. The change in speed is very small, and it doesn’t really reflect a change in strategy, just a slight change in the cards that make up the key archetypes. Secondly, the nature of BTT is that you build up threats over the first few turns, and then send those threats clashing at each other in the mid game. If you don’t pick up cards that can affect the early game, then you can easily get behind in building up your own board, and it is pretty difficult to come from behind in BTT because of the battlecruiser nature of the format. This means that you desperately need your early game. Thirdly, one of the big reasons why the format has slowed down is because the availability of good creatures on the low end of the mana curve is smaller, while the availability of five drops has increased.

This leads to my first suggestion for drafting BTT; prioritize cheap creatures. There are a lot of five drops in the format, which means that you simply don’t have to prioritize them very much. Unless they are extremely powerful, you should often just take the cheaper card. Meanwhile, there are just not very many good two drops in Born of the Gods, which means that you have to value the ones that are there even more highly than you normally would. Swordwise Centaur, Nyxborn Eidolon, and even Deepwater Hypnotist are all cards that have been climbing in my evaluation because it has become much more difficult to fill out the early part of your curve.


Popularity of Two Color Archetypes in BTT Draft

Popularity of Colors in BTT Draft

In the above two charts, we see the popularity of decks in BTT. In the first chart, we see the popularity of two-color archetypes in BTT, and in the second chart, we see the popularity of individual colors. When we normally look at the popularity of decks, we see them broken into tiers of about five decks, and this set is no exception. The top tier includes Simic, Boros, Dimir, Azorius, and Gruul. This shows us 3 blue decks, two green decks, two white decks, two red decks, and one black deck. In the second chart, we see that white is the most popular color, though it is followed closely by blue and green, while red and black trail behind, with black being the least popular color in the format. It’s also worth noticing that three of the four black decks are among the four least drafted decks in the format, with only Izzet dropping lower.

This chart is also particularly fascinating when we look at the same charts from Theros. All of the top five two-color combinations are the same, with one exception. In Theros, Golgari was the third most drafted color combination, but in Born of the Gods, it has dropped all the way to the second least popular combination, and it has been replaced with Gruul, which was the most popular of the second tier decks in Theros. As for colors, there were a few shifts among all the colors, but the most notable changes were the heavy dropoff in popularity for black, and the relatively large upswing in popularity for red (as compared with where it was).

It’s not hard to figure out why this is happening. Red gained a lot of value in Born of the Gods, with several powerful commons, including the headliners of Fall of the Hammer and Bolt of Keranos. Meanwhile, Black lost a lot of value in Born of the Gods. It loses multi-format all-star Gray Merchant of Asphodel, but it also loses Baleful Eidolon, Disciple of Phenax, and even Returned Phalanx all of which were key role-players in black decks. In their place, it gets Asphyxiate, Nyxborn Eidolon, and Servant of Tymaret. In my BNG spoiler analysis, I predicted that the draft format would have a warping effect on some aspects of the format, because the colors that are strong in BNG would be drafted more heavily, while the weaker colors would be underrepresented. It is obvious that this has been the case with BTT. Black was probably the strongest color in the format before Born of the Gods, and it just doesn’t make sense for it to drop so precipitously by changing only 33% of the format. At the same time, Red was clearly the worst color in the format, and such a significant change is a little extreme with the addition of one pack worth of commons.

This is a fascinating development in the format. Players are changing their draft strategies tremendously based on the things that are happening in the first pack, even though they won’t see that pack again later in the draft. Getting passed a Fall of the Hammer and a Bolt of Keranos seems like a clear signal, but it’s important to remember that the depth of red drops off sharply in the second and third packs. Meanwhile, it’s interesting to keep in mind that black will often seem artificially shallow in the first pack.

In the next section, we’ll look at how these dynamics affect the win rates of these decks.


Win Rate of Two Color Archetypes in BTT Draft

Win Rate of Colors in BTT Draft

In these charts, we see the win rates of various deck archetypes. In the first chart, we see the win rates of the ten two-color combinations. In the second chart, we see the win rates of the various colors. I should make a note about this; these numbers are different than what you’ll find in the numbers of various people using data scrapers to collect data from MTGO. The biggest reason why is because the people running those programs use different heuristics for assigning colors to the data they collect. For example, if a deck runs a majority of red and white cards, but is splashing couple of blue cards, then those programs will record it as a WUR deck, essentially just looking to see if certain basic lands are played in the deck. Similarly, a WU deck that is splashing R will be recorded the same way, even though it is much different. In my studies, I will count that deck as a RW deck with a U splash, since the bulk of the deck is focused on a RW strategy. This changes the win rates of these decks significantly. Obviously a well-tuned aggro deck with no splashes is going to outperform one that is forced to splash a third color over a large sample of games. Because of this, I tend not to trust the archetype win rate data that is used by these kinds of data scrapers. As I’ve said many times, a big sample does not make up for rigorously collected data.

In these charts, we see some very interesting numbers. In this study, we see that Golgari, Orzhov, Simic, Azorius, Gruul, and Dimir all come in with win rates above 50%, though Boros is just a hair under 50%. Four of the five most popular decks in the format are showing up with pretty respectable win rates, which is something that we hardly ever see on Ars Arcanum. Surprisingly, it seems that on the whole, people are drafting this set pretty well, which is not something that we normally expect to see. This creates a pretty interesting situation where the win rates of the decks in this format are extraordinarily flat. The top deck has about 56%, which is just very low for the top deck in a format. The one exception to this is Izzet, which has a 37% win rate, which is one of the lowest win rates that I’ve seen in the time I’ve done this series.

These charts still give us a few very important insights into the format. First, we see that black edges out green and blue as the winningest color, with white coming in just a hair under 50% and red falling well short of the 50% mark. There are a couple of things going on here. One is that black is being underdrafted, which means it is picking up a few more wins than it would normally. It is entirely possible that green is the stronger color overall, though we won’t know for sure unless we see a study where black’s popularity increases. The other important thing is that both Black and Green are the winningest colors in the format, despite being arguably the weakest two colors in Born of the Gods. One of the things that I predicted for this format was that the Theros packs would be more important than the Born of the Gods packs, and that it would be more important for players to set up a good pair of Theros packs than it would be to maximize their value out of the BNG pack. It seems that this prediction was correct.

Along that same note, the archetypes that performed the best in this study were the ones that seemed to change the least with the release of Born of the Gods. All five of these decks picked up a few goodies in BNG, but for the most part, they draw the majority of their strength from the Theros packs.

We see this strategy confirmed, that the way to best maximize your win rate in BTT is to set yourself up to make the most out of your Theros packs. I think that there is one question that best defines this strategy, and I think it is currently the defining question of the format, which is “What do I do with red?” Red is obviously very strong in Born of the Gods. It gets several powerful uncommons, and a bunch of powerful commons. I would argue that it is the strongest color in the BNG pack, and a lot of top players agree. The problem then is that if people focus their drafting on reading the red signals in the first pack, then they will often be fighting over red in packs two and three, which lowers the win rate of the red drafters all around the table. The most obvious conclusion is that you should just avoid red in the first pack, but if everyone does that, then you will see Fall of the Hammer and Bolt of Keranos going much too late. If only one person drafts red, that person’s deck is likely to just be insane. But if two people that are next to each other both draft red, then both decks will probably end up being miserable. Because of that, red is the most difficult color to draft in the format. There are potentially high rewards to the color, but it is important not to draft it unless you are able to get a lot of strong picks out of BNG, and that’s just not something that you can count on reliably.

My suggestion to the average drafter is to avoid red in the first pack. The strategy that I’ve been taking is usually to build my decks with a base of either black or green in Born of the Gods, depending on which color seems to be more underdrafted, and then I will supplement those colors with either blue or white. What this usually means is that I have to pass a few strong cards in BNG in favor of less powerful cards that are in a color that is more likely to be open in the Theros packs. During the first pack, it can often feel a little bit miserable. However, I’ve been reaping the rewards when I reach the Theros packs. I’ve been virtually mono-black multiple times, picking up late Gray Merchants of Asphodel, and I’ve even picked up Abhorrent Overlord after pick 4 a few times.

The trick is that since neither black or green are exceptionally deep in Born of the Gods, it isn’t that difficult to cut off any signals from those colors. If you simply take all the Asphyxiates or Nyxborn Wolves, it will be very hard for the people to your left to justify taking a black or green card because the other options will usually be significantly stronger. This can often mean that you are cutting off black or green for 2 or even 3 seats to your left. When the second pack comes back the other direction, it will often pay off in a huge way.

The question then is what to do when you don’t have strong black or green cards in the pack. I definitely do not advocate just forcing those colors; it is important to remain flexible. If you try to cut off one of those colors in pack one, but it is being cut to your right, then you won’t get any rewards out of pack three, and your deck will still be miserable. What I’ve been doing instead is that when I draft blue, white, or red cards out of pack one, I focus on cards that have easy color requirements. For example, I’ll draft Retraction Helix, Nyxborn Triton, or Fall of the Hammer pretty highly. I take Fall of the Hammer much higher than Bolt of Keranos, for example, because it has a lighter color commitment. If the red signals start flowing, then I’ll be able to jump into red, but if red follows the normal pattern of being cut off, I still have the option of splashing Fall of the Hammer, whereas I wouldn’t have that option with Bolt of Keranos. Searing Blood is another card that I don’t take highly out of Born of the Gods, because I just can’t justify taking a card with RR in its mana cost. Likewise, I’ve been taking Vanguard of Brimaz lower than I normally would. I started out pretty high on Nullify but it’s moved much further down my pick order because I don’t want to have to rely on having UU cards in my deck during the first pack. On the flip side, I’m perfectly happy committing to a GG or BB spell, and I’ve been taking Swordwise Centaur and Asphyxiate much higher as a result.


Thanks for reading! In conclusion, I’d like to briefly sum up the main points from this article.

1.      BTT is slightly slower than Theros, but not enough to justify a change in strategy.

2.      Instead, the main way that people should adjust to the speed of the format is to prioritize two drops a little more highly than they normally would. This is because five drops are overloaded in BNG, while two drops are much harder to come by. While it is harder to build a good curve, it hasn’t become less important, which means it is valuable to take two drops higher.

3.      The most important thing to do in the BNG pack is set up a strong pair of Theros packs. Having the BNG pack opened first means that the pack is skewing people’s draft strategies to the colors that are strong in BNG, but that isn’t necessarily an optimal strategy.

4.      Black seems to be underdrafted currently, mostly because it is weak in BNG. Cutting off black can be a strategy that pays off with big dividends in the second and third packs.

5.      When drafting colors besides black or green in the first pack, it’s usually better to take the card with less strict color requirements, so that it is easier to move between colors in anticipation of the Theros packs.

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





Excellent post by Avatar at Sun, 03/09/2014 - 16:37
Avatar's picture

This posting has been a great help to my draft game...But, can you add in the shuffler factor...lol.