Pokemon TCG Standard Format Data Analysis: February 17 – March 17, 2018

There have been eleven Standard format tournaments over the past month.  I’ve done a thorough compilation and analysis on many of the decklists from those tournaments.  The source of the data comes from Limitlesstcg.com, where they do an amazing job of tracking down and publishing decklists from all major Pokemon TCG tournaments.  However, they don’t have every decklist, so I want to make that clear from the very beginning: while this analysis represents the vast majority of top decklists from these tournaments, it does not include every top decklist from every tournament.  Of the 248 potentially available decklists, we have a total of 171 lists included in this analysis (69% of the total lists).

There’s another limitation of this analysis: I didn’t put any weight on the placements for the various tournaments.  For this analysis, first place at Collinsville carries the same weight as first place in La Paz, even though Collinsville had 33 times the number of contestants.  Therefore, the eighth place finish for some of the smaller tournaments is included in this analysis, while in reality had there been as many competitors at that event, it might not have even made top 64.  The smaller tournaments could potentially skew the results of this analysis, and I may need to do a follow up review of just the larger tournaments.  However, for right now, I’ve decided to include all available data.

Therefore, this should be a good representation of the most successful decks in the current Standard format, but it might not be perfect.  I can’t put a plus or minus percentage on it, but there are several big picture points that you can take away from this study that I think will help us better understand what the meta truly is and how we can refine our decks to make them the best possible.  And just to be clear – this runs through March 17th and does not include the Portland Regional that took place this weekend … and will undoubtedly alter the statistics presented in this work.

So to begin this analysis, let’s start from the top.  As I said, there are 171 decklists in this research, and there were a total of 189 unique cards played in all of these decklists:

From all of the 64 decklists available that made 8th place or better, this is a list of all 149 unique cards played in all of those decks:

Here are all of the cards that didn’t make top 8:

And then this is basically the same as above, just sorted by type:

Alright, now we start getting into the meat of the analysis.  Here’s a list of the total number of cards played, how many decks they were played in, and the average number of cards per deck:

And again the same data only for the top 8 finishers:

And here’s a list of how many decks ran a particular card:

See, I’m not the only one who doesn’t run N!  There are at least six other people who see the light!

And the same data for top eight finishers:

And yes it’s not lost on me that of the 64 decklists that we have available that finished in the top 8, N is the only card that was run on every list.

Here now are all of the cards by type, and what their differences were between overall and top eight finishers.  That’s the far right column, it shows how much more or less a type of card was run between the entire population and the top eight.
I’ve highlighted the major differences here: Stadiums were the biggest increase, and the differences between Basic and Special Energy were extremely interesting to note.

It gets even hairier here: this is a list of all Pokemon used, and it also includes the difference between overall and top 8.

This measures what share of the overall population a card had overall versus its share of the population in the top 8.  Obviously, every card will be less in the top 8.  The far right column simply tells how much a Pokemon increased or decreased its share going into the top 8.
So hopefully I can explain this here so it will make sense:

There were more Tapu Lele GX’s overall and in the top eight than any other card.  All 171 decklists combined to run 396 Tapu Lele GX, and all 64 of the top eight finishers ran 133 Tapu Lele GX.  Overall, that 396 is 14.229% of 2,783 total cards.  The 133 of the top eight was 12.764% of 1042 cards.

(12.764% / 14.229%) – 1 = -10.30%

10.30% is the relative percentage decrease between the amount of Leles in the top 8 and the amount of Leles overall.

This means that the top 8 finishers in the aggregate ran about ten percent less Tapu Lele GX’s than the overall population.

Likewise Zoroark GX (10.557% / 11.319%) – 1 = -6.73%.  Zoroark GX was still the most popular feature Pokemon, but it ran into trouble getting into the top 8.  Proportionally, there were 6.73% less Zoroark GX’s in the top 8 than in the overall population.

On the other hand, Buzzwole GX (5.662% / 4.456%) -1 = 27.06%.  I would agree with this as I’ve gone 9 W 2 L with Buzzwole Garbodor BKP this weekend.

Here’s a summary of the biggest discrepancies one way or another:

Volcanion and Glaceon GX had the best increase, meaning a lot of those decks (proportionally) made it  into the top 8.  Garbodor BKP was surprisingly one of the biggest disappointments, losing about 40% of its share of the population going into the top 8.

Here’s the same thing, only with Supporters:
And the major differences again:

So I didn’t include a number of the lower card counts here, and I identified three others that are probably too small a sample size to really make a good judgment with this limited amount of data.

Here are Items:
And again the biggest changes in population share:

Here are the numbers for Tools – not sure how much to take away from this as the quantity of data here is pretty small:
And finally the last data to present is on Basic and Special Energy:

Everything above is fact.  Objective, unadulterated numbers.  Below is what I’m taking away from all of this analysis.  Feel free to disagree in the comments.

  • The Unique Card List Overall by Type makes for a really nice quick reference when building a deck. Very easy to have a list of Supporters, Items, etc. to use to make sure you didn’t forget something in your list.
  • Very interesting that top 8 finishers ran more basic and less special energy.
  • I mentioned Garb is down arrow, Volc and Glaceon up arrow.  Also Regirock EX apparently propels you into the top 8 more.
  • Worth noting: Zoroark GX decreased but Baby Zoroark increased by more than 50%
  • And run Octillery if you want to make it into the top 8 too.
  • Next to Garbotoxin, Promo Koko was also the big loser.
  • Very interesting is diff between Sycamore and Cynthia. This would lead you to believe that Sycamore is better than Cynthia.
  • E Hammer was interesting to see – run more of it, but I already identified that more basic energy than SPE made it into the top 8.  Seems to be contradictory, but the numbers are what the numbers are.
  • Not telling you anything you don’t know here – abuse Max Elixir to make it into the top 8.  I’m actually trying Order Pad just to go grab it faster or get a Puzzle if I have one in hand so I can get two Elixirs out of discard.  I had a game this weekend where I played six Elixirs … and hit every one of them (in fairness I was running like 16 energy).
  • Less Evosoda, less max potion – combine that with Elixirs and we might be making Stage 1’s as obsolete as Stage 2’s.
  • Tools – small numbers but FFB might be better than Choice Band

And the biggest takeaway of all – USE BIG DATA!  Everyone is doing it: sports teams, businesses, law enforcement.  If you’re still using “I think” and “I feel” to determine your deck choices, you are not being the best Pokemon player you can be.

Thanks much to Otaku for giving me the push I needed to get started with this.  If you would like access to the raw data, please leave a comment below and I’ll be more than happy to share the Google Sheets link with you.