Modern Regionals by the Numbers

Last weekend, Star City Games hosted a series of Regionals events around the country.  Event attendance fell somewhere between 200-400 contestants and each cut to a top 8.  Since Star City Games was kind enough to publish the top 8 deck lists in each of the 14 tournament locations, the data from this weekend represents a more accurate reading of the diversity of the Modern metagame.  Though WoTC publishes its own results from 5-0 finishes on MTGO on a bi-weekly basis, these results only reflect one of each style of deck that finishes in this way and does not represent the frequency of that type of deck during each event.  Being an avid baseball fan and a very competitive Magic player, I’m always eager to compile the SCG Regionals results in a spreadsheet to see how frequently the “best decks in the format” are actually receiving the expected results.  In this post I’ll share my findings on that front.  If math makes you queezy and spreadsheets trigger a gag reflex, you may want to walk away now.  If you’re generally interested in making an informed deck choice based on the numbers, stay tuned as the following information may surprise you, confirm your suspicions, or make you think, “why the hell am I building Owling Mine right now?” and opt for something a little better positioned.

Locations

SCG Regionals events took place in 14 locations throughout the country.   Click on each location for a link to the top 8 from that event.

That gives us 112 deck lists to consider.


Percentage of the Field

…in the top 8, that is.  While we don’t have specifics about how many of each deck participated in the event overall, what is perhaps more relevant to us is how many of each deck made it to the top 8 of its tournament.  Here are the most represented decks in each of the 14 events:

7. TIE: KCI and Colorless Eldrazi (2.67%, 3 decks)

6. TIE: Titan Shift and Blue Moon (3.57%, 4 decks)

5. TIE: G Tron and Burn (4.46%, 5 decks)

4. Grixis Death’s Shadow (5.36% 6 decks)

3. TIE: Mardu Pyromancer and UR Gifts Storm (6.25%, 7 decks)

2. TIE: Affinity and Jeskai Control (7.14%, 8 decks)

1. TIE: Humans and Hollow One (9.8%, 11 decks)

Compare that to the Modern metagame represented on MTG Goldfish which is largely populated by MTGO 5-0 results.

Accurate as of June 5, 2018

It was largely agreed upon that Humans and Hollow One were the two strongest decks in the couple of months.  As digital results have tended away from that being the case, at least for Hollow One, paper players may be less likely to change decks as quickly or as frequently as their online counterparts.  Percentage-wise, the differences in the eight most played decks are not significant and we are looking at a mere 2% of the meta-share difference between Humans and Hollow One in this case.

As far as the Regionals results go, for our most played decks to represent 10% of the top 8s, the difference is substantial.  Read that as “one in ten decks is Humans or Hollow One” or…”one in five decks is an aggressive creature deck.  With that in mind, its easy to see why Jeskai Control has benefited so much from this recent shift in meta.


Average Quality of Finish (Slugging)

Returning to the baseball analogy, it is difficult to find the perfect stat point to quantify a player’s value on offense or defense.  While a batting average can show how often a player is able to hit the ball, it doesn’t represent whether that hit is a single, double, or triple.  For that, baseball statisticians use a metric called “slugging” which factors the value of each of those hits in.  In order to accomplish a similar result with the Regionals data, I’ve assigned a point value to each finish to represent how deep a deck was able to go into the top 8.

1st place is worth 5 points

2nd place is worth 4 points

3rd and 4th place (functionally the same thing; semi-finals) is worth 3 points

5th and 6th place (quarter-finals) is worth 2 points

7th and 8th place (quarter-finals) is worth 1 point

Though Humans and Hollow One were the most represented deck, they may or may not have been the most successful.  In order to determine success based on the quality of finish for a deck, I’ve taken the total point value of all similar decks and divided that number by the total number of those decks to acquire what I will call the “average quality of finish” or “AQF”.  In other words, if there were 11 Humans decks that accumulated a total of 24 points, I would divide 24 by 11 for an AQF of 2.18.

An AQF of about 5 means that most copies of this deck won their tournament.  An AQF of about 1 means that most copies of this deck lost in the first round of their tournament.

Here are the top performers by that measurement:

14. (AQF 2.18): Humans (11 decks)

13. (AQF 2.3): Boggles (3 decks)

12. (AQF 2.4): G Tron (5 decks)

11. (AQF 2.45): BR Hollow One (11 decks)

10. TIE (AQF 2.5): Grixis Death’s Shadow (6 decks), WB Eldrazi (2 decks), and Ponza (2 decks)

9. (AQF 2.63): Jeskai Control (8 decks)

8. (AQF 2.67): KCI (3 decks)

7. (AQF 2.75): Blue Moon (4 decks)

6. (AQF 2.86): Mardu Pyromancer (7 decks)

5. TIE (AQF 3): Colorless Eldrazi (3 decks), UW Control (2 decks), Eldrazi Tron (2 decks), UR Prowess (1 deck), Naya Midrange (1 deck)

4. TIE (AQF 3.5): Counters Company (2 decks) and UW Miracles (2 decks)

3. (AQF 3.7): UR Gifts Storm (7 decks)

2. TIE (AQF 4): Elves (2 decks), Temur Scapeshift (1 deck), Kiki-Chord (1 deck), and Goryo’s Vengeance (1 deck)

1. TIE (AQF 5): Jund (1 deck) and Dredge (1 deck)

Now this list looks quite a bit different.  In fact, our two most represented decks in the last category are in the bottom half of decks that were actually able to “seal the deal” once they made it to the top 8.  Keep in mind that getting into the top 8 in the first place is quite difficult to do.  None of these decks “failed” by not winning the tournament.

While this list might pique your interest, there are some shortfalls that we need to address before we read to far into it.  The best performing decks by measure of average quality of finish were Jund and Dredge.  Both decks were the only of their kind in all fourteen events and, in both cases, they also won their events.  If we were to discard these types of outliers (anything with fewer than 5 copies total), our top 5 performers would look something like this:

5. (AQF 2.45): BR Hollow One

4. (AQF 2.5): Grixis Death’s Shadow

3. (AQF 2.63): Jeskai Control

2. (AQF 2.86): Mardu Pyromancer

1. (AQF 3.7): UR Gifts Storm


Archetype

Lastly, we’ll consider representation by deck archetype.  I know we aren’t all going to agree on the way that we should label each of these decks so, since I’m the one writing this article, I’ll tell you how I do it and you can mine your own data if you don’t like it.  😉

Aggro: Aim is to attack or burn the opponent as fast as possible and before they are able to establish a combo, lock, or control

Attrition: Generally, any deck that plays pointed discard spells and typically aims to win by attacking (could be considered a strategy within-in the umbrella of Mid-Range)

Big Mana: A deck that’s goal is play costly spells earlier than is possible by playing a single basic per turn

Combo: A deck that tries to cheat and doesn’t feel too bad about it

Control: A deck that plays re-active spells that can either counter or remove permanents that are essential to their opposition’s strategy

Creature Combo: A deck the relies on creatures for its combo, and thus, has a functional midrange back-up plan

Mid-Range: A deck that plays creatures and spells that provide more value than the average card and typically wins by attacking with creatures

The most played deck by archetypes are as follows:

Creature Combo: 7/112 (6.3%)

Mid-Range: 9/112 (8%)

Combo: 12/112 (10.7%)

Big Mana: 12/112 (10.7%)

Attrition: 14/112 (12.5%)

Control: 17/112 (15%)

Aggro: 41/112 (36.6%)


Based on that information, I’ll leave the rest up to you.  Do you approach an aggro-dominated meta with a Jeskai Control deck knowing that Jeskai Control decks were able to average at least a first round win in the top 8?  Or do you pick up Storm knowing that it had a very respectable showing of finals and semi-finals appearances?

If you haven’t had enough numbers for today, feel free to peruse my entire document at your leisure…

SHOW ME THE DATA!

Discussion