One more stretch in the competitive season is over. It’s time to kick back, relax, and enjoy the… what? Regionals is WHEN!?
Yup, Regionals here in North America will pick up where Cities left off on January 19th (this coming Saturday). I’ve got a UG article coming up, but it won’t be released in time for me to have a comment on Cities analysis or Regionals, so I thought I’d write this up now.
Different Strokes for Different Folks
There are many ways to look at results such as those on my PokéGym thread. You could count total wins, total T4’s, the ratio of T4’s converted to wins etc… You could even scour the pages for information on matchups, trying to pick apart every discernable scenario when two decks of interest played, tallying up the winner to get a picture of if the matchup is really 50/50 or not.
If you’ve been reading my articles for a while now, you’ll probably know I’m a Bio Major (done my degree now, just waiting on the presentation), and as such I love stats. I’m crap at actually hand-calculating them, but I love thinking about stats and analyzing stats to try and paint the most accurate picture about what I’m trying to say.
One of my nervous twitches is when I see someone looking at results and taking away things from them that I wouldn’t take away (and would argue are false). Sometimes this is a result of the method they’ve chosen to analyze the results, and other times it’s an artefact of the way that the data is organized itself.
So today I’m going to look through the Cities results in a few ways to both try and show what I’m talking about, and figure out the actual trend in City Championship results.
There was a really good UG article on this sort of stuff by Andrew Wamboldt a while back, talking about “Power Rankings.”
Now… he looked at 4 tournaments. The PokéGym City Champs thread has 154 Masters results over many more weeks, so I don’t think I’m going to go to the effort of evaluating all the T4’s over every week. But I will look through and analyze the evolution of winning decks as Cities developed, a sort of Power Ranking if you will.
First, let’s just see the results.
Decks Arranged by Total 1st Place Finishes
I’ve chosen to present them in descending order solely by 1st place finishes. In the event of a tie, it went to 2nd place, T4 etc… So what could we take away from this table?
First and most obvious, Blastoise/Keldeo seems like a pretty clear BDIF. Blastoise/Keldeo has distinguished itself from the pack in almost all categories. Many more 1st’s than any other deck, and many more total T4’s than any other deck.
The next most winning deck would be Landorus/Mewtwo sitting pretty with the next most wins and the next most T4’s. And rounding out the top 3 we have Darkrai/Hydreigon slightly ahead of the rest of the pack in wins, but far ahead in total T4’s.
So the play for Regionals is clearly to counter the BDIF Blastoise, while keeping good matchups with the next most winning decks, right? There’s no trick I can apply here to say “Well this deck may have WON the most, but the deck that T4’d the most was _____.” Blastoise/Keldeo, Landorus/Mewtwo, and Darkrai/Hydreigon all managed to distinguish themselves from the pack in both categories.
But we must remind ourselves that the way in which we gathered this data is by total values. This method of analysis is only truly valid if we assume an equal distribution of play amongst each of these decks.
And I think we can all agree that a deck like Klinklang didn’t see as much play as Darkrai/Hydreigon.
If both decks were as proficient as each other, then we might expect the same T4’s converted to wins ratio, despite a difference in total values. Let me throw another table in your face.
Decks Arranged by Top 4 Win Percentage
|Deck||1st||T4||T4 Win %|
This table has organized the same decks by their percentage of T4’s converted to wins. Obviously you take the percentages here with a grain of salt, since I’m not trying to argue that Mewtwo/Terrakion is better than the most winning decks. But once you get past the background noise of decks with low totals, you can see an interesting pattern emerge.
For one thing, a deck that completely slipped past the radar in the first analysis (even if I had analyzed top 4 or top 5) is Hammertime. Of the decks with an appreciable amount of Top 4’s, Hammertime claims the best win percentage, just barely eclipsing Landorus/Mewtwo and Blastoise/Keldeo.
The other thing we see here is that Landorus/Mewtwo and Blastoise/Keldeo both come in right near the top again, suggesting that we shouldn’t attribute their totals success to simply being overplayed.
After that we see a gradual degradation of win percentages with every deck being relatively successful. What’s surprising is how our 3rd place total T4’s deck, Darkrai/Hydreigon, comes in dead last amongst these decks in terms of total T4’s converted to wins.
And another deck I’ve somewhat ignored is Darkrai/Landorus, which has the best T4’s-Wins ratio of any deck with a powerful 60%, and a worth-considering 10 total T4’s. The kind of information we can take from a table like this is that Darkrai/Hydreigon is seeing a lot of play, and is a good deck, but doesn’t seem to have the stuff to make it deep in top cut.
And what’s more is that a deck like Darkrai/Landorus clearly isn’t a bad deck, and might be underplayed right now. Though I wouldn’t expect its win % to remain at 60% with increased play, I would nonetheless expect it to be one of the best decks in the rankings.
The reason it’s easy to underestimate Darkrai/Landorus is because the Darkrai playerbase is divided. When there are 3 categories all describing approximately the same deck (Hammertime, Darkrai/Stuff, Darkrai/Landorus), it’s easy to underestimate the archetype as a whole.
These are all “Big Basic” Darkrai decks that can be built off a very complete 40+ card common skeleton. The establishment of separate archetypes is somewhat defined by a dissimilarity in the general skeleton, but for some reason we look at Darkrai/Big Basics as separate decks, as if they played differently and had different strategies.
Imagine how ridiculous this would have been in past formats if we had tallied ZekEels by the choice of tech attackers. If we took 35 wins for ZekEels and subdivided it into “Pure ZekEels,” “ZekEels/Thundurus” and “ZekEels/Tornadus.” We’d end up with tables like the ones above, where ZekEels wouldn’t appear at the top of the chart.
What’s more, some variants would be much farther down, out of sight and out of mind. This is what has happened with Darkrai/Landorus.
People look at Darkrai/Landorus and view it as a completely different deck due to the fact that it is successful enough to have earned its own category in the 1st place listings. How messed is that? What’s more, many Hammertime or Darkrai/Stuff lists featured Landorus, but were grouped into their respective categories. The way in which the data has been presented encourages people to ignore Darkrai/Big Basics due to the fact that each category alone does not have an impressive amount of wins.
But treating these decks as one unit can paint a much better picture of their success. Allow me to demonstrate.
Condensed Decks Arranged by total T4’s
|Deck||1st||2nd||T4||Total||T4 Win %|
As the number of wins is pretty similar amongst many decks, I’ve arrange this table by total T4’s (as it mirrors total wins fairly well, but shows what decks did best better). I’ve also united Hammertime, Darkrai/Landorus and Darkrai/Stuff into one category.
Moreover, there were results in the 2nd place finishes and 3rd and 4th place finishes for Darkrai/Mewtwo as its own category, as it picked up a lot of placings on its own.
In organizing my thread, I did this because people want to look at the thread and see interesting techs and unique decisions. But in painting a bigger picture, this is a far more appropriate representation.
As a point of comparison, I’ve also united RayEels and MewtwoEels as “Eels,” which picked up 3 more T4’s from the “Eelbox” category. And credit where credit is due, Garbodor DRX/Probopass DRX won a Cities, and I’ve united Garbodor as a category. Garbodor/Probopass is also probably one of my favorite rogue concepts to come out of these Cities, so major props to the person who came up with it.
Lastly, I united Empoleon/Friends as a category, despite varying strategies. The strengths of the deck are built on the same backbone (just the same as Eels). Though I wouldn’t necessarily argue that these should be united as decks (nor that RayEels and MewtwoEels are essentially the same), I’ve done this table up as a point of comparison.
It’s an alternate way to view the data organized solely by the backbones of the decks (which keeps Darkrai/Hydreigon separate from Darkrai/Stuff).
And just look at the difference in conclusions we draw from it! Blastoise/Keldeo remains #1, but Darkrai/Big Basics is clearly a dominant force to be reckoned with. This is why the UG has spent so much time looking at Darkrai, despite seemingly average showings.
And as we saw in the previous table, Hammertime has the best win % of the Darkrai/Stuff decks, which helps explain why the UG writers are enamoured with Hammertime when they talk about general Darkrai/Stuff. And of course, Eels as a whole has done incredibly well, sitting right near the top with the big boys.
If I were to look at a metagame and choose my deck based on the most successful decks throughout Cities, I would look to this table. Though I admit that it may be fallacious to unite Eels as one category (as the skeletons really aren’t that similar), the strategy for beating both decks is nonetheless very similar. If you take out the Eels and have a good answer for Mewtwo, you’ll be in a good position.
What’s more is that both variants lose a significant portion of their games to Landorus or Mewtwo + DCE donk. There is no other archetype up there that has to worry about such an easy chance to be donked in this metagame.
So, is this the way we should be looking at Cities results in picking our decks for Regionals? Well, thanks for sticking with me through this whole analysis, because everything I’ve said up ’til now is somewhat bunkum.
Though all the aforementioned results and tabulations are accurate, and the conclusions are logical, the data itself is not an accurate representation of the strongest play for Regionals. This data ignores trends in the metagame that developed as the weeks went by.
The best way for me to demonstrate this is to show exactly when these decks picked up their successes. I’m not about to look at overall T4’s throughout the weeks of Cities, but I will look at when decks picked up their wins.
This is the pseudo-Power Ranking I alluded to at the beginning.
Decks Arranged by Total 1st Place Finishes
|Deck||Weeks 1-2||Weeks 3-4||Weeks 5-6||Weeks 7-8||Total|
Now, there are a couple things that are going to strike you as odd so I’ll address them immediately.
First, the total wins listed in this table don’t match the total wins of the previous tables. Why? Because there are 214 total posts to skim through, and the data isn’t well labeled to avoid confusion. I’ve probably included a few Seniors results in here that I might not have for the previous tables.
And perhaps in the previous table I included Seniors results that I avoided in this table. Perhaps a tournament was reported twice and I didn’t pick up on it. The list goes on, but overall it doesn’t change the point of what I’m trying to illustrate.
It’s pretty obvious what deck archetype(s) progressed as the metagame matured. Let’s start with the decks that picked up speed.
The most obvious is Blastoise/Keldeo. Blastoise/Keldeo is a sleeper BDIF. It was thrown into a format full of established archetypes (even Landorus-EX/Mewtwo EX had Bouffalant DRX/Mewtwo EX/Terrakion NVI as a predecessor) and told to catch up.
Once marathons hit in the last couple weeks of Cities, Blastoise/Keldeo had finally been refined to an optimal state. And the results speak for themselves. Everyone fumbling through the dark with the deck in the first few weeks laid the path for that success.
Another deck that might fly under the radar a bit is Klinklang. Before US Nats, Chandelure NVI/Accelgor DEX/Vileplume UD managed to pick up a win or two in only the final week of Battle Roads. And at Nationals and Worlds it demonstrated itself as a potential BDIF (although the time problem was apparent).
I would watch out for Klinklang in the same way. This is a deck that people with deckbuilding skill have only picked up recently. I’m willing to bet a faction of strong Klinklang players are going to show up at Regionals with a vengeance, and I predict a Klinklang or two to go deep into cuts.
Another similar sleeper would be Garbodor. Now, Klinklang was already a bit of an extrapolation, but Garbodor most certainly is. It’s not like it appeared out of nowhere, but I do think that Garbodor might make a stronger play at Regionals than it has been throughout Cities.
The metagame has had time to develop, and the coverage options that Garbodor needs are better established. Another fringe deck to watch for.
Lastly, Landorus/Mewtwo came back strong in the last couple weeks. After dominating in the early weeks, the deck fall off the map. What’s more interesting is that this isn’t even the timeframe where Blastoise/Keldeo rose up. Weeks 5-6 had the fewest results of all, but nonetheless it’s interesting to see such a decline in success.
Of course, it bounced back alongside the resurgance of Blastoise/Keldeo. And perhaps you could argue that Blastoise/Keldeo saw such great success because Landorus/Mewtwo became so prevalent.
Something that isn’t obvious in this data is the prevalence of Eels throughout Cities. In the first weeks RayEels contributed most of the wins. In the second weeks, it was more MewtwoEels. And in the final weeks, it was again more RayEels than MewtwoEels. If there’s a deck that needs its Eels, it’s RayEels. And if there’s a deck that kills Eels better than any other, it’s Landorus/Mewtwo.
RayEels saw a real resurgence in the last couple weeks, but those efforts could not be converted to wins. Landorus/Mewtwo squashes the deck so hard (this is what I gather from results, which I think speak stronger than individuals’ playtesting) that despite a large increase in play*, the deck couldn’t jump ahead. Even if it’s argued to have a good Blastoise/Keldeo matchup (which was everywhere).
*unsourced personal observation
Decks on the Decline
And what about the decks that declined? There’s one deck I see in that table that was hit hard by the maturing metagame. Darkrai/Stuff decks win totals declined as the weeks went by. This isn’t a picture-perfect way of looking at it, because Darkrai was actually the most winning deck of weeks 5-6.
But nonetheless, Darkrai/Stuff started out as one of the strongest deck choices at the beginning, and as the metagame developed other decks stole its thunder.
Does this mean it makes a bad deck for Regionals? Not at all! Darkrai was highly successful throughout Cities for a reason. It’s consistent, fast, and when teched properly, is difficult to counter.
But in terms of overall power, the deck has lost its place amongst the kings of the format. In BLW–DRX RayEels was the more “powerful” deck, capable of dealing out 1HKOs as it needed. And now we have Blastoise/Keldeo, also capable of dishing out 1HKOs when it needs to.
This leaves Darkrai/Stuff sitting on the outside, resigned to 2HKOs in a 1HKO format. It’s not hard to see why the general archetype has faded. The answer that the deck needs is a way to restore its 1HKOs (which is definitely possible).
Don’t let me discourage you from Darkrai/Stuff, but be conscious of the gap in power between it and other decks, and tech to compensate this gap.
The Play for Regionals
So what’s the play for Regionals? I’m not going to say. I think that really depends on your playstyle, and what you deem to be problematic in testing.
What I wanted to do with this article was to show how you can analyze results best. And what I hope I’ve done is given you a good way to look at the results we have, and come up with your own conclusions.
No matter how you look at it, Blastoise/Keldeo is the frontrunner for BDIF. But it is not the dominant BDIF that the first tables suggest. Darkrai/Stuff, Landorus/Mewtwo and Eels are nipping at its heels in total wins, and total T4’s.
Considering how many players are going to look at these results and see Blastoise sitting pretty, there is going to be as much Blastoise hate as possible. So your play for Regionals should really encompass the decks that players will be drawn to as their answer to Blastoise/Keldeo.
And at the same time, be conscious of those targeting the metagame as a whole. I can really see three-four best deck choices in that list that all could do very well this format. So we’ll just have to wait for Regionals to find out what decks exceed or fall short of expectations.
P.S. Check out my blog at: http://tcgwithhats.blogspot.ca/