So Many Insane Plays – Measuring Metagame Diversity and Balance

I geeked out with some new data tools.

I hope this inspires either more data analysis with this measure or a search for more new measures.

last edited by Smmenen

@Smmenen This is really cool. Thank you. This is a beautiful way of showing how restrictions and bannings tend to fix a format for awhile, before it is "solved" again and they have to ban something else. I think in the next few months, once RNA is released we will see if Vintage is really in a sweet spot or if it is just waiting to be solved. Another thing I am interested in is how the new player in the format, Survival will impact people's ability to "solve" the format, and if the MOCS and potential for Pro Tour will bring new grinders and players who will either help accelerate format diversification by bringing other cool new decks in, or may "solve" it once again, potentially leading to yet more restriction.

Someone in the facebook group misread or misunderstood my article as asserting that if a format is both diverse and balanced, then it can't be unfun. Nothing could be further from the truth. To avoid any confusion, I am very clear here that formats which are diverse and balanced can still be unfun.

In my view, balance and diversity are necessary conditions to a fun format, but, jointly, they aren't sufficient conditions. They don't automatically make a format fun, and I stated that explicitly in the article.

To be clear, even though I believe -P--> -Q, it does not mean that I believe that P --> Q, where P is the joint condition of diversity and balance, and Q is fun.

But to elaborate on this a bit: The components that make a format "fun" are not elusive or mysterious. They are well known.

The key to fun is a concept called meaningful choice. For Magic to be fun, people need to feel that they have meaningful choices:

  1. they have meaningful deck choices, and not circumscribed to certain colors or strategies.

  2. they have meaningful card choices, meaning that the design decisions they make, matter

  3. that the choices they make within a game influence the outcome of the game, from technical play to sideboarding.

And I define meaningful choice as:

"1) It means that there are valid options. (that is, there is not really one only valid choice, and the others are clearly invalid, illegal or infeasible or options

  1. it means that the decision to choose one option or another has an influence on the outcome of the game.
  2. It means that logic, reason, or some sort of analytical reasoning process can be applied to aid in those decisions.

All three elements have to be present for choices to be 'meaningful'.

Imbalanced and non-diverse formats are unfun to most people because they tend to violate the first form of meaningful choice.

That doesn't mean that everyone playing those formats are not having fun. Obviously, the players playing the dominant deck have a different experience than players who hate the dominant deck.

It's perfectly reasonable to acknowledge that Vintage is both balanced and diverse, but nonetheless view it as "fun." But, I assert in my article, whether a player finds game play is interactive or not depends massively on what deck you play and the matchups it faces. The Shops v. PO matchup is one of the least interactive in the format. If you are a Shops player, it's likely you will find this matchup unfun, and thereby enjoy the format less.

So, whether a format is "unfun" in the sense of being non-interactive or not can't be assessed as a valid generalization from a single deck or player perspective; it depends on the aggregate views of all of the players in the format. The only way to quantify that is by doing surveys on a Likart scale or something like it. But that doesn't follow that the data in my article isn't useful or important.

The Gini-Simpson formula can't tell us if games are interactive; but it can quantify whether a format is diverse and balanced, and if it scores badly on that front, it can certainly suggest that a format is unfun on those grounds. That makes it useful, at a minimum.

As I said, it's not the end-all-be-all, but it quantifies, in a single metric, two elements that matter quite a bit to assessing the quality of a format.

last edited by Smmenen

@smmenen Good read! Very cool with a tool that can measure diversity. Seeing is believing.

My only critique of your article would be that it wasn't immediately obvious that you flipped the index. You wrote

Recall that a higher value indicates a more balanced and more diverse metagame, and a lower number indicates a less balanced and less diverse metagame, with .99 being a nearly perfectly balanced and diverse metagame with a plethora of viable decks, and a value closer to 0 being a metagame dominated basically by a single deck.

But for that sentence to make sense, you would have to have read the endnote as you make no mention above of why this changed.

  • 4
  • 529