Patch 2.8 In-Depth Statistical Meta Report

Hey folks, Dr. LoR here with an in-depth stats-based meta report, the first one that is published on RuneterraCCG. For those who haven’t followed my past work on Reddit or Twitter, let me first introduce myself.

I’ve been playing Runeterra semi-competitively since Open Beta, hitting Masters every season since the game’s official release, and I’ve also qualified for every Seasonal Tournament to date (going 4-1 in the first two and opting to sit out the third).

More importantly, I have been regularly producing stats-based in-depth meta reports (starting from patch 1.4), as well matchup tables (starting from last season). I have also written a series of posts for Reddit, where I attempt data-driven optimizations of popular decks.

In this article, my goal is to provide you with an up-to-date meta snapshot, using the data from with the last week of play.

Sources and Methodology

• The 600k matches of data for the analysis below come from Mobalytics Meta Stats. The timeframe is May 21-27.
• I am relying on Mobalytics’s archetype algorithm (still in beta), which uses champions and regions to define archetypes. This means that Shyvana Asol is separate from Shyvana Zoe Asol and Renekton Sejuani is different from Renekton Sivir Sejuani. Where appropriate, I manually combined Mobalytics archetypes with slightly different champion combinations into single archetypes for the purposes of this report. In particular, I chose to combine all the Matron Cithria archetypes (championless, or various combinations of Shyvana, Kalista, Elise, Lucian, Kindred, Garen, and Jarvan) since they all have similar winrates.
• I use Bayesian smoothing toward a win rate of 51%. It is a technique that allows to avoid statistical flukes. If you want to learn more about the concept, check out this explainer.

Most Popular Archetypes

In the chart below you can see the playrates for the 16 most popular decks, sorted by rank. When combined, these 16 decks account for about 75% of Diamond/Masters metagame, 66% of Platinum, and 56% of Gold.

As you can see, Azir Irelia and Thresh Nasus are still going strong, accounting for nearly one third of the meta.

Most Successful Archetypes

Below, you can see the winrates for the 16 highest-WR decks that have a minimum of 2500 matches in Plat+, again grouped by rank.

Riot has previously noted that 15% playrate and/or 55% winrate at Plat+ is their threshold for balance adjustments. Azir Irelia is the first archetype I have ever seen to exceed both thresholds at the same time.

Below is a list of all the decks that have more than 1000 matches in Plat+, sorted by winrate.

The table is split into super-columns by rank. Under each super-column, you’ll find a sample size (match), the playrate (pr), and the winrate that is adjusted using the Bayesian smoothing technique (bwr).

In addition, the Plat+ and Masters data include 95% confidence intervals for each winrate. What that means is we can be 95% sure that the ‘true’ winrate for each deck lands somewhere between numbers shown in the columns ‘95% Conf.’ and ‘Interval‘ (for example, we can be 95% sure that Azir Irelia has a winrate between 55.1% to 55.8% in Plat+).

Most Meta-Defining Archetypes

Let’s plot win rate versus play rate for Platinum+, to see which decks are most impactful in the current meta.

In the chart below, you can see ‘a dot’ and a ‘vertical line’ for each archetype. The vertical lines represent the 95% confidence intervals for win rates (see the explanation above). Even a sample size as big as 2500 matches still gives you a margin of error of +/- 2%, so it’s important to understand that if we really want to compare decks that differ by 1-2% winrate, we need to have a very big sample.

The X-axis (playrate) is placed on a logarithmic scale – otherwise, everything will be squashed to the left while Azir Irelia will be very far away on its own to the right.

• Some decks have high winrates but lower play rates than they probably should. Most notably, Azir Noxus is one of the best counters to Azir Irelia and sports a healthy 55% winrate – but sees less than 3% playrate… I guess people don’t like the deck’s play patterns? Discard Aggro and Draven Ezreal (both are good or at least close to even against Azir Irelia) also have winrates above 53%, with fairly low playrates.
• Dragons continues to be overplayed a bit. Its winrate has never risen above average. It’s definitely a powerful deck in the right hands, but the less-skilled players pull the numbers for the archetype down.
• TLC continues to gain playrate, especially at Diamond and Masters. It beats most of the other top decks – other than Azir Irelia and Overwhelm.
• Frozen Thralls featuring Lissandra and Taliyah is starting to gain traction. Curious to see if this deck can be strong enough to compete once Azir Irelia gets nerfs. The versions with Zilean and the build that features Lissandra-only have lower winrates than Taliyah build.
• People continue to play decks featuring Spectral Matron + Cithria, Lady of the Clouds, and they continue to win less than 50% of their games. Every week, there’s a streamer claiming that the deck really is good, but the fact that no combination of champs in this shell has over 50% winrate at any rank suggests to me that the deck is a dud.

Conclusion

The meta continues to be dominated by the same handful of decks, with little change over the last couple weeks. A few newer archetypes are starting to rise in popularity, but the dominance of Azir Irelia (and to a lesser extent, Thresh Nasus) will have to end before any new contenders can rise to true prominence.

Follow me on Twitter for more updates!