Skip to content

Runeterra Expeditions: Chickens, Eggs and Leaderboards

In a Mastering Runeterra interview, RubinZoo mentioned that there were no plans to add a Ranked mode for Expeditions that he knew of. “But it’s a chicken and egg thing, though,” he said, since an Expedition ladder would foster the competition that the mode currently lacks. Could the LoR community help solve this chicken-egg problem?

 

 

Back in mid-August, Riot’s RubinZoo sat with Jason and MajiinBae in the 27th Episode of the Mastering Runeterra podcast.

 

Around 41 minutes into the interview MajiinBae says, joining palms in mock prayer: “Poor Expedition gamers… they are probably like, yo, could we have some more Expedition stuff?”

 

“I would love to play more Expeditions,” Jason says. He mentions how he loves Limited formats, but Expeditions have nothing for him: he’s a competitive player who wants to qualify for worlds, win money, earn prestige and show he’s the best. A textbook Spike, looking to kick ass and take names, and who crushed Drafts back when he was an MtG pro. Yet LoR has, somehow, failed to grab Jason’s Limited attention.

 

When asked during the same podcast, RubinZoo confirmed that Riot could definitely create a Ranked ladder for Expeditions. “But as far as I’m aware,” he said, “there are no plans to do so.”

 

“I’m not sure there’s enough of a player base to support it,” said MajiinBae. “Expedition players tend not to be competitive.”

 

“It’s a chicken and egg thing, though,” said RubinZoo. Perhaps, if there were a Ranked Expedition ladder, it would foster competition. “But I don’t exactly see, right now—I’m not a producer, so I don’t make this type of calls—I don’t exactly see a huge benefit to making Expeditions a ranked mode. But we’ve considered it in the past, and if something has been considered in the past, we will always be willing to potentially revisit it.”

 

A month later, in mid-September, the reddit crowd mourned the departure of FTPBUST, an Expeditions-only content creator with more than 400 Trials on YouTube, and who (even after not playing for over a month) still sits at rank #5 for most 7-wins in the leaderboard generated by the LoR Guardian deck tracker

 

“Expeditions is one of the best limited formats I’ve ever played,” said FTPBUST in his last stream for LoR, “but they never did anything else. I plan to switch to Magic [Arena] until something happens to Expeditions. If there’s ever a shake-up, like adding Ranked, or switching the rewards to be something interesting, or a concerted effort to keep the format updated and fresh, I’m more than willing to come back.”

 

The reddit thread about FTPBUST’s departure brims with comments that echo the streamer’s point of view. “Expeditions is the only way I play Runeterra, and I prefer its variety over Constructed,” wrote arthurmauk, another Top 10 in LoR Guardian’s leaderboard. “What I’d love to see is an official ranked ladder system for Expeditions. That would bring the competitive players in. There are huge Limited communities in Magic and Hearthstone, so the fact that there isn’t one in Runeterra suggests Riot isn’t supporting it enough.” And, in parallel with RubinZoo’s chicken-egg problem, arthurmauk adds: “Though Riot thinks there isn’t enough demand, if they implemented it there may be.”

 

In the Mastering Runeterra interview, RubinZoo mentions one factor that could nudge Riot to change their stance: players’ ardent support for the game mode. Could the LoR community solve this chicken-egg problem?

 

Data Diggers

 

Although friends and family call him Alessandro Balconi, LoR players acquainted with his weekly Master Meta Report know him as Balco.

 

As a data analyst for a mobile videogame company, collecting and presenting data is Balco’s bread and butter. “I started collecting LoR data around April 2021,” he told me, “Mainly because there was no free website to check the data I was looking for, which were matchups tables. By that time Runeterra.ar was already quite famous but didn’t have all the data it has now: it was mostly a player match history website back then. Legna was also starting to collect data around that period. Mobalytics was the only website with the kind of data I was after, but it was locked behind a paywall.”

 

“I used to love the Vicious Syndicate meta reports, back when I played Hearthstone. The LoR Meta Report has a similar structure—it’s mostly stuff that I’m interested in, and stuff that I think other people might like.” 

 

Some days ago I approached Balco, along with other eminent data wizards, wondering if some of them could perhaps perform their data-gathering magics and obtain stats about play rates and win rates for Expedition cards…

 

 

A screenshot of a computer

Description automatically generated with medium confidence

… and Balco not only obtained some very interesting stats for Expedition Champions, but he was also able to extract data about Expedition players.

 

After tracking 24,661 Expedition games played by Constructed Master players, Balco was able to sort them by games played and by win rate, broken down by shard (Americas, Europe, and Asia).

 

 

Chart, bar chart

Description automatically generated

 

 

Chart

Description automatically generated

 

 

 

Chart, bar chart

Description automatically generated

 

 

 

Chart, bar chart

Description automatically generated

 

 

 

Chart

Description automatically generated

 

 

 

Table

Description automatically generated

 

As a die-hard Expeditioneer myself, I remember skipping a heartbeat when seeing these graphs.

 

Had Balco stumbled upon the Holy Grail? Had he, in a few hours, answered untold amounts of prayers and produced an official Expeditions Leaderboard?

 

Not quite, sadly.

 

Balco only tracks data from players that have reached Masters in the Constructed Ranked ladder. That is to say, the above stats do not take all players into account—players in Iron, Gold or Diamond, no matter how much (or how well) they play Expeditions, are not included in Balco’s sample, since Riot’s API does not allow Balco to do so.

 

“And what, pray tell,” I was forced to ask, ashamed of my encyclopedic ignorance of all things coding, “is that API you speak of?”

 

“An API allows two servers to communicate with each other and exchange data,” he explained to me. “That’s how Riot makes their data available to third-party developers, for example websites like op.gg or u.gg for League of Legends, or the websites that Legna, Vivo [owner of Runeterra.ar] or myself do for LoR.”

 

“Whenever we request data, we are asking Riot’s server to make some effort in sending us that data,” Balco said. “And to prevent overloading the servers, there are limits to how much information we can collect. Everyone can start collecting match data using Riot’s API, but unless you register your website or app and Riot approves it, the amount of data you can collect is really small. By default, a LoR developer can only collect 100 matches every hour—Legna, myself and probably Vivo can collect some more, but only because Riot specifically allowed us to do so. That’s why I only collect Master data. With the current limits, I can collect basically all matches from Master players, but not much more than that.”

 

Vivo, owner of Runeterra.ar, agrees: As he’s seen replying in the Twitter thread (in Spanish), “I have the same possibilities as Balco”

 

 

 

 

If a single APIarist isn’t enough to collect the data, could perhaps this be tackled as a community endeavor?

 

What if a large enough swarm of developers, as a team effort, tracked a larger number of players? Surely at some point they would, by sheer strength in numbers, strong-arm this stubborn API?

 

“I don’t think so,” Balco said. “It is indeed possible for different people to work together and collect more data. But the problem is that they would need to know which players they should be collecting data from. The data-gathering process starts with the API requiring a player’s name+tag. For Masters it’s simple: We have a Master leaderboard. But collecting data from all players is impossible since we don’t know the names of all players that play the game.”

 

And, even if Balco or some other master APIarist somehow managed to trick this wretched piece of gatekeeping code to cough up all the names of every player, they would face yet another hurdle: The opaque, elusive, never undisclosed and little understood…

 

 

Graphical user interface, text, application, email

Description automatically generated

 

Expeditions’ MMR

 

The Problem with MMR

 

Expedition players are matched via an MMR of which the community knows very little about, other than it exists—Balco has confirmed that this MMR is unavailable through the API.

 

“People who play tons of Constructed play less Expeditions, and vice versa,” said Conansson, who ranks #18 in LoR Guardian’s leaderboard and produces instructional Expedition vids on his YouTube channel. “I myself managed to get to Diamond without a lot of effort, but generally prefer to play different decks over mastering one and then grinding with it, so I stay in Bronze or Silver most of the time. Balco’s Leaderboard is a good starting point for sure, but we don’t know enough about how the MMR works to make anything more than educated guesses.”

 

“Unless Riot reveals the hidden MMR, the leaderboards we have now are sort of pointless,” says ShadesRealm, an Expeditions-only YouTuber and Twitch streamer who plays around three Trials per day and has uploaded more than one hundred 7-wins runs on his channel. “If a leaderboard ranks you by win rate, like Balco’s do, then you only need to get lucky a few times to get to the top. Doubly so if you are brand new to Expeditions and therefore play against low MMR players. If you rank by most games played, or by most 7-wins, then what you are really measuring is amount of free time to play LoR.” ShadesRealm’s stream sessions last four-five hours, in which he plays around twenty matches. “If I were to do that every day, I’d be at six hundred matches per month, which almost doubles the #1 spot on Balco’s leaderboard.

 

“Now imagine if someone actually played nothing but LoR for twelve hours a day and how high these numbers can go. Playing a lot doesn’t necessarily mean you are good at the game, just that you have too much free time. The only way to tell is by revealing the hidden MMR.”

 

“Given the tools at hand, it’s impossible for anyone to put together an MMR-based ranking,” said FTPBUST, the content creator who migrated to Magic Arena, where he now ranks among the Top 100 Limited players. “I considered doing it, but the API isn’t really built out for it. And in my opinion, any other measure—like win rate, or amount of 7-wins—is pointless unless we have access to our MMR. RubinZoo mentioned on one of the podcasts that Riot could set up an Expeditions ladder, but that the community support for ranked Expedition doesn’t warrant the work. Knowing that a ranked system likely isn’t going to be a reality was a fairly large hit to the competitive community.”

 

“If you had to guess,” I asked FTPBUST, “why is Riot so remiss about giving details about the future of Expedition, ranking systems, leaderboards and such?”

 

“The rewards for releasing information to the public are very low,” he said, “and the risks are high. People will latch onto features you mention even in passing and be toxic about them for years when they don’t come to life. Tweets and podcast remarks will get quoted out of context. Crudely put, a Dev talking about the future is walking on eggshells.”

 

 

The Other Leaderboard

 

 

Rankings are, as witnessed, a hot Expedition button. But player’s rankings was not the only interesting bit of data that Balco dug from the API mines. As it happens, he was also able to produce a Champions tier list.

 

 

Image

 

 

The numbers are for the exact same data sample as the previous graphs: It takes into account only Expedition matches played by Constructed Master players during the last thirty days. Readers interested in the data can check it in this link.

 

“It’s possible to analyze the play rate and win rate for every single card,” says Balco. “If we could collect matches from all players, we could also analyze which Champions and cards appear more frequently in decks that get to seven wins.”

 

What about not just appearance rate, but win rates based on when a card is drawn? Or stats about how long each card is kept in hand, on average?

 

“No, not through the API. This is something that only a deck tracker could collect [like the one LoR Guardian or Mobalytics use].” The API is also blind to everything that happens outside matches: Balco cannot track what happens during the Drafting phase, like which cards and Archetypes the player chooses, nor can he know if a player resigns the Trial without playing a single match (ie refusing to play the Gauntlet if they drafted a mediocre deck, and instead electing to quit the run and re-draft).

 

If data is interesting, doubly so are two sets of (conflicting!) data. As it happens, Champion win and play rates as shown by Balco’s and LoR Guardian’s samples don’t always agree. What could explain the difference?

 

Nami is no doubt a Constructed powerhouse,” says ShadesRealm when I asked about one such disparity. While Balco and LoR Guardian roughly agree about Nami’s win rate (Balco: 43%; LoRG: 49%), they differ in play rate (Balco: 4.7%; LoRG: 0.09%), implying Constructed Master players play the Queen of the Tides a lot more than her win rate would recommend.  “Nami already starts at a handicap because her Archetypes come with some really bad cards, like Ebb and Tidal Wave. And even assuming Master players can avoid these trap cards, without Elusive units it doesn’t matter how big Nami buffs her friends if they are just getting chump blocked by a 2/1 one-drop. In Expeditions, Nami’s basic problem is that she lacks a finisher, and that may not be obvious until you draft her a lot.

 

Irelia is another Champ that caught my eye, and my goodness do Masters players need to learn how to draft her. I’m guessing they may be going too hard on self-recalling cards like Lead and Follow. In truth, Irelia is really simple to draft: You see Greenglade Duo, you take the duo. Multiple duos just end games quickly.”

 

What about Xerath, who like Nami shows less-than-stellar win rates in both data samples, but huge differences (7,7% versus 0,15%) in play rate? “I’ve slowly come to think of him as one of the better Control champs, with a solid late game if you can pick up some Safety Inspector or the pretty Epic The Arsenal. My theory is that too many Masters aim for high-synergy combinations like Waste Walker, Obelisk of Power and Unleashed Energy. The stars align sometimes, sure enough, but more often than not you just wind up with a hand chock full of duds. The correct way to draft, I think, is actually to not draft the many landmarks. You want to be drafting Rock Hopper, Bomber Twins, and Unraveled Earth as your landmark generators.”

 

We should be wary of drawing too many conclusions, arthurmauk cautions. “Certain biases or preferences [in Balco’s data] can be rationalized as Constructed players expecting more from Expeditions, but I am not sure if that’s necessarily true. Drafting is often down to playstyle: between raw power and synergy, I lean more towards the raw power end. I want each of my cards to pull their weight independently, so I see some high-synergy cards to be traps. But I’ve seen, and been beaten by, many high-synergy decks. High-synergy can work, I just think it’s riskier and may lead to a lower win rate overall. I prefer a more risk averse approach.”

 

“In general,” says Conansson, “I’d say that champs who want to have a specific deck built around them, and don’t work in your generic play-on-curve Midrange shell, are Expeditions trap.”

 

I mention to him that, according to Legna, Riven seems to always have an underplayed deck: A deck that has a very good win rate in Masters, but that players folks just don’t seem willing to play it. Legna attributes this effect to Riven being extremely solid, but perhaps a bit boring or not flashy enough.

 

“Exactly,” Conansson says. “That’s why Riven has such a high win rate in Expeditions: She has no deckbuilding cost at all, you can just slam her into any deck, draft a good curve and some spells and blam, seven wins here I come.”

 

 

A Chicken with One Basket

 

 

Going back to the hot button of a competitive and accurate Ranking system for Players… if MMR and API confabulate to make it impossible for the community to solve the Leaderboard conundrum, how much of a hurdle would it be for Riot to do so? In-game changes are surely costly, but how about an out-of-game leaderboard in Riot’s website, like Hearthstone does for their draft mode, Arena?

 

“I have no idea how Riot structures their data,” Balco told me, “So I cannot say how long it would take them to put a leaderboard together. It would definitely be nice if they released, by the end of the Season, who the best Expedition players were—I would guess that is something they could do without too much effort, but then again it’s really hard to tell from the outside.”

 

Although at first Balco was surprised by the amount of Expedition games he tracked—he would have guessed it to be even smaller than what his numbers show—the fact is that Constructed remains vastly more popular than Expeditions, at least among Master players. And, as RubinZoo noted in the interview, Riot’s chicken has all its proverbial eggs in the Constructed basket.

 

“But, on the other hand,” Balco said, “if Riot were to add some incentive to playing Expeditions…”

 

As somebody deeply invested in drafting and limited modes, I can only say Amen.

 

May the baskets multiply, and may our Expedition players be heard.