Destiny Analytics Quarterly Report - Q1 2019


1. INTRODUCTION
Written by WADE HAYES

Much like a quarterly filing for a company, the Destiny Analytics Quarterly Report will try to provide you with all the information you need to know about tournament results, notable trends, and strategies to counter the meta. This will hopefully encourage you to be more creative and feel empowered to challenge the status quo.

NOTE: Although this article primarily looks at data from the late Across the Galaxy meta, plenty of the principal ideas found within the analysis can be used to reflect on the current early Convergence meta.

2. METHODOLOGY
SAMPLE SIZE
For my analysis, I selected a sample size of 30 tournaments. This number is generally agreed upon to provide a statistically significant sample size from which to draw meaningful conclusions. The underlying reason is that as the sample size grows, the more closely the data aligns with the true or expected value. For instance, when you roll a die 10 times, the average of the rolls may be 2.5, but after rolling the die another 20 times, it is statistically probable that the results will normalize and the value will be nearly in line with the expected average of 3.5 for the 30 total rolls.

Though I don’t define an expected value for my sample, I deem the sample size sufficient to draw meaningful conclusions regarding Star Wars: Destiny and the competitive environment.

3. LEGEND
As part of my statistical procedures, I assigned attributes to each deck in my sample. The following is a listing of the attributes of which one was assigned to each deck.stats1jpg

4. 'GAME THEORY'
As we think about the competitive environment, it is important to remember the interaction of each deck type. The diagram below displays a rough estimate of how decks (archetypically) interact, with aggro decks beating mill decks, mill decks beating vehicle decks, and vehicle decks beating aggro decks (classic rock-paper-scissor). STATS2jpg
One forgotten element of Destiny is the control deck type, which traditionally relies on mitigation, removal or some other control element to choke the opponent.STATS3jpg
This deck type has nearly ceased to exist, with very few if any decks being truly control decks. For an example of a control style deck, check out Intellectx’s Unkar/Wullf deck, which he took to a 5-1 finish in the Toronto Galactic Qualifier (the deck was originally used in Standard Format, but is no longer legal due to rotation).STATS5jpg
As these deck types interact in a "fixed order", examining which types of decks are most prevalent in the meta (at the moment) helps to ensure that you can adequately counter the meta. For instance, if an “Aggrodeck like eVader/Greedo or Mother Talzin/Phasma is dominating the meta, it might be worth considering playing a tanky “Vehicledeck variant, as these traditionally have provided the best counter to “Aggressivedecks.

5. RESULTS.
The following analyses , (6). Deck Type and (7). Deck Speed and (8). Deck Composition, include an "Overall Table" and a "By Finish Table". These are intended to give you an idea of the number of decks in the Top 8 by the designated attribute (6. Aggro, Control, Mill or Vehicle / 7. Fast, Typical or Slow / 8. Hero, Neutral or Villain) and for what percentage of the whole population that attribute accounts (the percentage of that type from the entire analysed data).

Additionally, the "By Finish Table" is included to show whether those decks outperformed or underperformed their representation.

6. DECK TYPE (OVERALL/BY FINISH):
OVERALL TABLE:stats6jpg
BY FINISH TABLE:
analytics1jpg
ANALYSIS
Somewhat unsurprisingly, given the popularity of certain characters (in different metas: Darth Vader/Phasma/Palpatine), Aggro decks dominated the late Across the Galaxy meta. That said, they have still outperformed their representation in the meta, meaning that if you made the Top 8 with an Aggro deck, you had a greater likelihood of winning than if you brought a different deck type. Another interesting note is that Aggro decks were also over represented just outside the Top 4, which is likely due to self-attrition, or Aggro decks facing other Aggro decks (basically cannibalising each other).

The success of Aggro decks came at the expense of Mill and Vehicle decks, as both of these decks were underrepresented in the first position, compared to their total representation in the Top 8 (both deck types look to be revived in the early Convergence meta, but this time at the expense of Aggro decks). Another interesting note is the lack of control decks. This will hopefully change as time goes on and the Convergence meta continues to develop, but I fear this deck type (Control) may be relegated to the history books.

7. DECK SPEED (OVERALL/BY FINISH):
OVERALL TABLE:
analytics2jpg
BY FINISH TABLE:
analytics3jpg
ANALYSIS
Interestingly, it seemed that action cheating was no longer as profitable, with fewer decks that were considered “Fast” winning at the top table (even before Force Speed, Hit and Run, etc. rotated out of Standard Format). In contrast, more methodical Aggro decks such as eVader/Greedo or even eSnoke/eTarkin rose to prominence in the first quarter of the year. These types of Aggro decks relied less on action cheating and more on careful turn sequencing. This strategy seemingly paid off, as a larger number of average speed decks won tournaments than were present on a percentage basis in the Top 8.
speed in destinyjpgAnother interesting take-away from this analysis is that it seems that you were more likely to finish outside the Top 4 bringing a deck that utilized substantial action cheating, which has traditionally not been the case, but is probably a welcome change for many Destiny players. Slow decks, though not generally outperforming their representation, did perform well and outperformed their representation in positions 2-4, which is very respectable and shows the consistency of that particular deck speed.

For those who wanted to win a tournament, based on this sample, it would seem your best chance would have been to play a methodical aggro deck, but if you wanted to counter that deck type and play something unique, it would have been an opportune time for a very slow and heavy hitting vehicle deck to take center stage (as the FAT Vehicle decks indeed did).

8. DECK COMPOSITION:
OVERALL
analytics4jpg
BY FINISH
analytics5jpg
ANALYSIS
The picture painted by this analysis is INCREDIBLY clear. If you wanted to win a major tournament in Quarter 1 of 2019: YOU SHOULD NOT, UNDER ANY CIRCUMSTANCE, BRING A HERO DECK! This was a sad reality of Destiny at that time, but your chances of winning a tournament with a hero deck was abysmally small. Though I don’t have data on this, I would assume that hero and villain decks were fairly equally represented in most major events, yet only 30% of the Top 8 were hero decks and that number dropped to 23% at the top place. From a statistical perspective, this is brutal. A SPARK OF HOPEjpg
Even if it is pretty thematic that hero decks are the underdog as they are in Star Wars itself, the fact that the top table were only 20% hero and 80% villain/neutral could indicate a pretty grim future for heroes. As I could hardly believe this statistic, I did a little more digging to determine how dismal the situation in fact was. Of the 7 hero decks that won a Regional event or a Galactic Qualifier, 4 of those winning hero decks were Mill decks with the remaining being FAT vehicle decks. This means that not a single Aggro hero deck won any of the 30 tournaments I analyzed.

That said, Convergence has brought with it many new characters and cards for heroes, thus there is still hope. Rebellions are built on hope as we all know! My recommendation though would still be to NOT bring an Aggro hero deck to a competitive event unless you are certain you have a strong chance against all Gauntlet contenders.

9. MYTH #1:
"The competitive environment is diverse, with many different decks reaching the top tables".

VARIETY IN DESTINY
Are there generally a variety of decks represented at top tables in Star Wars Destiny? The general consensus of the community seems to be yes; however, after the first quarter of the year, the answer is decidedly NO. Those that said yes would likely attribute the dearth of decks to overrepresentation of certain pairings, but this is also a myth that will I'll address below. diversity in destinyjpg
Out of my random sample of 30 tournaments selected for analysis, half were won by either eVader/Greedo or eSnoke/eTarkin. The other half was won by eYoda/eLeia2, eHan/eQi’ra, or eDooku/eTalzin. With the number of cards in Destiny, there are nearly infinite unique decks that could be built, thus I find it incredible that of all these combinations, there are only FIVE decks capable of winning a tournament in metas across the globe.

10. MYTH #2:
"Overrepresentation of certain decks drives tournament results".

REPRESENTATION DRIVING RESULTS
One commonly held theory in the Destiny community is that the overrepresentation of certain decks in the field determines the top cut or result of the tournament (simply put: If a given deck won loads of tournaments then it was because a lot of players played the deck). To test this theory, I performed an interesting calculation on what was arguably the most overrepresented deck in the first quarter of 2019: eVader/Greedo.Vader greedo listjpg
Ideally, I would have liked to have seen:
  • How many players in total played the deck,
  • How many proceeded to the top cut, and
  • Whether the deck won the tournament.
However, I do not have that data available. What I do have is information on the top cut and the winning deck.
overrepresentation in destinyjpg
To draw a meaningful conclusion from the data I do have, I looked at the total population of the top cut and determined what percentage of the decks were eVader/Greedo. I then determined what percentage of the winning decks were eVader/Greedo. analytics6jpg
If overrepresentation determined performance, we would expect to see the Top Cut percentage exceed the Winning percentage, as the deck would be “overrepresented”, which would in turn drive performance. In reality, the Winning percentage actually exceeds the Top Cut percentage, which suggests that the deck was performing well and winning not because there were a greater number, but because it was simply a 'better' deck.

Due to the statistical nature of the test, we can extrapolate the results to the entire tournament pool, which would then allow us to say that even if fewer eVader/Greedo decks were played, it is likely it would still be disproportionately represented in the top cut, and in the top cut would have a higher likelihood of winning.

Though this may not be the case for all decks at all points in time, at the present moment, overrepresentation does not drive performance (and it was at least NOT the case for eVader/Greedo).

11. CONCLUSION:
Though the meta is constantly changing, and the data represents an "old meta", I feel that the analysis performed here also provides meaningful insight into the current state of the game as well as provides direction for deck builders looking to beat the meta!

Personally, I felt the biggest surprise was:
  • The lack of diversity in the Top 8, as well as
  • The debunking of the myth that representation drives results.
This is definitely something to keep in mind going forward, as the Gauntlet has shrunk and the chances of winning bringing a 'unique' or 'non-mainstream' deck have tumbled.

Thank you for bearing with me through the massive amount of data that is compiled and analyzed here and please don’t hesitate to reach out if you have any additional questions or ideas for additional analyses to be performed in future articles!

PATREON SUPPORT:

As a Patron you support the YOUR Destiny website, podcast and YouTube Channel, and you help us to continue providing content for Star Wars Destiny. We are a dedicated Destiny website and release articles almost daily. As Patron you can enter our Patron only Discord Channel as well as getting access to all our "Chat 'N' Play" videos, while Advanced and Expert tier patrons also get access to exclusive training videos and test videos. Thank you for your consideration!