Tier of Fear: VTD04 JPN Format

VGE-V-TD04With the Bushiroad Championship Series starting up and the launch of Booster Set 02: Strongest AL4! in Japan, I believe it is time to take a look at the Japanese meta post-VTD04 Ren Suzugamori. Since the majority of our Bushiroad Championship Series events are under this format, let us take a look how the format shaped up.

Format Details:

This format began on the 10th of August 2018 and the format will end on the 30th of August, 2018. The format consists of 11 clans released in the Japanese game. These clans and their gifts are as follows:

Force: Royal Paladin, Kagero, Shadow Paladin, Spike Brothers and Dimension Police.
Accel: Nova Grappler, Tachikaze and Aqua Force.
Protect: Oracle Think Tank, Megacolony and Granblue.



VGCS are unofficial, large-scale events run by card shops and various members of the Japanese Cardfight Vanguard Community. These events usually have considerably larger turnout than weekly shop events and act as valuable representations of the trends in the Japanese Cardfight Vanguard meta. So far in this format, we have had only one VGCS (due to the relatively short period that this format falls into).

193 VGCS – Number of Players: 57 (19 Teams of 3)

Top 4 Teams:

1st) Oracle Think Tank / Spike Brothers / Kagero
2nd) Oracle Think Tank / Spike Brothers / Kagero
3rd) Oracle Think Tank / Kagero / Aqua Force
4th) Oracle Think Tank / Dimension Police / Spike Brothers

Event Clan Distribution:


So as we can see, Oracle Think Tank and Kagero are the highest represented decks.
OTT has dominated the Standard Format from the beginning, this trend continues to prove just how strong the clan is. Out of the 19 teams, 15 of them had an OTT player and every single team in the top four had an OTT player. The first and third place teams had reported that their OTT players had won 6 out of 7 rounds. The second and fourth place teams also reported that their OTT players had won 4 out of the 7 rounds.

The statistical perspective for OTT:
A team representation percentage of 78.95%.
100% placement rate in the top 4.
A win rate of 71.43%

Kagero has been the anthesis of the OTT from the beginning of Standard. Thanks to Dragonic Waterfall’s ability to ignore the sentinels added via the Protect gift marker. Kagero also features the most efficient control skills in the game so far and a unit that can stand back up: Dragonic Overlord (which also doesn’t do anything against OTT). Out of the 19 teams, 15 of them had a Kagero player and every single team in the top three had a Kagero player.

The statistical perspective for Kagero:
A team representation percentage of 78.95%
75% placement rate in the top 4.

Spike Brothers also did really well, out of the 4 players using the clan, 3 of them made it to the top four. Dimension Police and Aqua Force had a representative each in the top 4 but had average usage. The Dimension Police player reported winning 4 of the 7 rounds while the Aqua Force player only managed to win 2 rounds out of 7.

Top 4 Teams Performance:

This VGCS clearly demonstrates the chokehold OTT has on the meta, how will the weekly events hold up?

Weekend Tournament Summary

The Japanese card shop scene is active and growing. The TCG industry is a mainstay of Japanese pop culture. We can use the weekly tournament reports from the local communities, to distinguish trends in strategy, card choice and the ever-evolving theory-crafting of the Japanese version of Cardfight Vanguard.

Week 1 to 3 (August 10th – August 28th) Clan Win Rate (Percentage):


Let us have a look; OTT has an ever-increasing win percentage with an upwards trend, an increase by 10 points in the second week and another 2 points by the third week (or an 80% increase over two weeks). Following up in second place is Kagero who also won a greater proportion of events over the three weeks. A 3 point jump in the second week shows the trend that if OTT increases, so do Kagero, to a lesser degree.

Dimension Police is also a solid Force contender, with a 6 point increase in the second week before losing steam with a decrease in the third week with a 7 point drop. Quite surprisingly Aqua Force had a huge jump in the second week with 9 points before ending on par with Dimension Police. Aqua Force is the only Accel clan to gain points in this format. Shadow Paladin also ended the format on a high note by disappearing in the second week but a strong finish in the third with 7 points.

Now for the losers, Granblue had a strong start in the first week but lost 9 points by the second and ended up slightly better off in week 3 with. Spike Brothers were vying with Dimension Police in the first week but lost 12 points in the second week and had no representation in the third week. Poor Tachikaze and Nova Grappler, both clans started out the format relatively low, lost a third of their representation in the second week and have no representation in the third week. The Accel Curse (the fact that each Accel clan have been completely outclassed when the newest Accel clan is released) takes no prisoners. So, let us get to the main event, what everyone has been waiting for.

Tier List (VT04 Format)

So based on the last three weeks and the VGCs, I will explain my tier list and justify my placement of clans in each tier. Below I have a table showing the clan win percentage and their average percentage, which I have used to rank each clan.


So in a game like Cardfight Vanguard, it feels a bit disingenuous to simply make a tier list based on average results and consider it gospel. The idea that Aqua Force, for example, cannot beat Dimension Police is false and I think this is why people often get angry and upset at any mention of a tier list. If your favourite clan isn’t doing well, this does not indicate that the clan is bad, per se. It is more of an indicator that your clan requires more support and lacks the tools to deal with clans with greater (or better quality) support. This list may also indicate the higher ranked clans may have less of a luck factor or may feature unfair mechanics that actively make it easier for them to win. This tier list is simply a representation of what clans have higher chances of winning events and does not accurately make judgments on which clans are “bad” or “good”.

I’d like to think of the whole idea of rankings is based on this logic using OTT as an example: OTT can win events easily which leads to increases in the usage of OTT players. The increase of OTT players leads to more chances of winning due to the increase in the number of OTT in the player proportion at events. This causes a decrease in the usage of other clans while increasing the usage of clans that can compete against OTT. In this scenario, clans that cannot win against OTT consistently will automatically drop to the lower tier and even if the clan can win games against OTT, it has to compete with other clans that may have better matchups against it.

But are the clans that are highly ranked actually overpowered? To find that out, let use some more maths! So for this analysis, I am going to use a standard deviation of the average win percentage from the table above to work out if the higher tier clans are too overpowered. So we start by calculating the mean of the averages (so add them up and divide by the quantity) we get: 99.55/11 = 9.05. Now we have to work out the standard deviation with is: the square root of the sum of each value minus the mean squared and divided over the number of values. I’ll skip all of the long tedious maths to get our standard deviation of 7.57. So what do these two values mean and how can I use them to draw a conclusion?

Well, the mean of the average win percentage (9.05) indicates that the average win percentage of all the clans put together is 9.05. If we look at the table again, we can see that the clans with an average value higher than 9.05 have a greater range in their usage percentages over the three weeks and events. This implies that these clans are doing well compared to the other clans in the game. You could argue that the clans that have higher averages are the “good” clans, which I  agree with but let us continue.

Now, the standard deviation is a representation of how members of a group differ from the average value. It signifies the amount of variation or dispersion in a set of data. That is what we want to see, how varied is our meta and what this indicates about the health of the clans. Our standard deviation was 7.57 which helps us create categories we can slot our data into to come to our conclusion. We do this by adding or subtracting the standard deviation from our mean, up to three times in both directions:

9.05 – 7.57 = 1.48 – 7.57 = -6.09 – 7.57 = -13.66
9.05 + 7.57 = 16.62 + 7.57 = 24.14 + 7.57 = 31.71.

So we have our standard deviation distribution categories of:

-13.99, -6.09. 1.48, 9.05, 16.62, 24.14, 13.71.

So what do we do with these categories? Well, we check which of the clan averages fall into between each set of values. I have then summarised them into tiers below.


S Tier: Oracle Think Tank
A Tier: Kagero
B Tier: Dimension Police, Aqua Force and Spike Brothers.
C Tier: Royal Paladin, Granblue, Megacolony and Shadow Paladin
D Tier: Tachikaze and Nova Grappler

So of course, OTT is extremely dominant and Kagero being their natural counter (as close as you can get anyway) is behind in the second tier. OTT has a six-point lead over Kagero and is the only clan in the S Tier. Kageros specialisation in cheap retires and powerful Grade 3 units helps the deck maintain a stranglehold on the other clans. Spike Brothers follow the duo from EB02, Aqua Force, in particular, increasing in popularity while Dimension Police asserts itself as a dominant Force clan in B Tier. C Tier is the middle group where both paladin clans have to fight off the other two protect clans, which may have the better match-up due to how the Protect gift nullifies massive columns. Finally, the D Tier features the Accel curse on display. It is a little saddening to see Shadow Paladins with only a Trial Deck dominate two accel clans with full support.

So that is the VTD04 meta, nothing much has changed at the cutting edge, I am excited to see what AL4 does to the game. I hope you join me next week when we look at what the first week of the AL4 meta gives us.

Tier of Fear features statistical analysis based on data collected from a variety of sources (Twitter, secondary news resources, various Cardfight Vanguard Discord channels, Reddit etc). With this first edition, I am not able to provide sources for all my data due to the sheer size and amount. I will in future editions make it a priority to list my sources in order to legitimise my claims and to help other players access the information freely to make their own conclusions.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s