Analyzing LoL's new Pick and Ban system

by theScore Staff Jan 7 2017
Thumbnail image courtesy of Riot Games/lolesports / NA LCS Spring 2016 / Riot Games

Riot has introduced a new pick and ban system for the 2017 competitive season, shaking up staff and making good on their promise to do so in the past (a good omen for Riot in 2017, perhaps?)

Lots of critics have observed the current state of drafting in competitive play as stale, with only 57 champions chosen at 2016 Worlds of the 123 available. The hope is, from professional players, staff and viewers, that the new format will shake up champion and strategic diversity. However, some have also criticized the format, calling for a setup similar to the one seen in Dota 2.

There are lots of problems with the conception that adapting another big MOBA's pick-ban system is automatically the correct option, so I've looked into a simple comparative between other MOBAs in order to draw out a distinction that disqualifies the Dota 2 format as a necessary addition to the LoL competitive ecosystem.

A comparison of MOBA Contemporaries

One thing to note immediately is that a pick and ban system is meant to be another factor in competitive balance; one side will inherently have an advantage in the draft no matter what, purely based on game systems and the metagame that develops in competitive play. Understanding why Riot adopted a pick and ban phase that wasn’t Dota’s has everything to do with the unique aspects of these two MOBAs and their associated quirks.

Champions (or Heroes) aside, one of the bigger functional differences between Dota and LoL is the difference in the way the maps are laid out. Dota is more unique in its map’s asymmetry, and as such carves out specific advantages for one side over another. In the past, Dire had an advantage because of its proximity and easier ability to reach Roshan, but Radiant has had an advantage as well at certain points in time. Radiant has an easier time stacking their ancient camp (imagine if you could cause dozens of Krugs to spawn in one camp, then put them all on gold/exp. steroids), and has an easier time pulling creeps into the jungle to create experience discrepancies.

Conversely, the cartography of Smite is unique as well. But one thing remains consistent when compared with LoL: the maps are fairly symmetrical in their composition. This means that there are theoretically no inherent advantages for one side of the map because both teams operate with the same quadrants and the same architecture. However, other factors may determine the differences in map winrates.

For red side in League of Legends, the oft-cited reasons for its lower overall win rate involve the camera angle being more awkward for players to lane on the red side (an issue not seen in the third-person perspective that Smite uses), as well as the limited ways to enter and engage or disengage Baron Nashor in more balanced mid and late game situations. As well, LoL's high number of power and flex picks in recent metas tends to neutralize the counterpicking power that red side holds in the draft stage.

From maps to metagames

But it can’t be as simple as just asymmetry, can it? Dota's draft system and associated meta has often shifted the power balance in the other direction to compensate for the fact that Dire side gets two picks in a row during the first rotation. For example, at the Boston Major, Radiant still outperformed Dire 151-134 (52.9%) even though Dire has the advantage in draft. The new Dota map is unclear data-wise as to who has the inherent side advantage, but its asymmetry is still particular to that MOBA.

However, LoL is dealing with a different kind of asymmetry. One of the other big ways in which competitive play is balanced is the meta, and how many champions are considered priority picks at a given moment. A glut of power picks potentially gives a slight edge to whichever side has more opportunities to get what they want or need, and lane assignments can shift the ways in which we conceive of the priority roles or champions in a draft. This is all more theoretical than the asymmetry mentioned, but worth considering as the game actively changes.

The system that LoL has adopted is most comparable to Smite, in so far as it is the exact same pick and ban phase, probably for all the aforementioned tangible reasons. The symmetrical maps lend the corrections to side advantage to be minimal based on the draft format. As such we see something fundamentally different to the Dota draft: a reduction of phases. Dota 2 has six phases total, while LoL now has four.

This simplifies the process, and allows for less strategic diversity than Dota 2. This is because Dota's final phase, as shown below, has each team banning and then picking their last choices, completely revealing a good portion of their team composition and opening up a last chance at securing vital counterpicks or the finishing pieces of a team composition.

In Dota’s pick and ban system, Dire has a double pick very early in the draft that Radiant never has access to, allowing them to secure two very powerful picks immediately. This differs from the LoL system, which gives its blue side more double picks. In exchange, red side has a counterpick in each picking phase.

This is a good indicator of pick-ban balance to offset the aforementioned red side detriments, and it will be interesting to see if this swings win percentages too far in favor of red side. I'm inclined to say that it does, especially in the early stages of the implementation of this draft, as staff may put greater emphasis on comfort picks for players rather than crafting compositions that can execute a particular style.

In the current meta, the following picks are all regarded as fairly viable, among others:

Lane Champions
Top lane Camille Poppy Fiora Nautilus Maokai
Jungle Lee Sin Rengar Vi Hecarim Rek’Sai
Mid lane Orianna Cassiopeia LeBlanc Ryze Viktor
AD Carry Caitlyn Jhin Ezreal Varus Ziggs 
Support Zyra Karma janna Brand Nami

At this point, it seems likely that power picks are still comfortably a part of the first set of bans. Afterward, securing picks which don’t reveal crucial parts of what you need in your composition would be best. Picks such as Poppy, Lee Sin, Orianna, Ryze, Karma, and Zyra are well-rounded enough that they don’t reveal much with respect to the type of composition you are running, versus something like a Fiora which would reveal what is likely a splitpushing composition.


At the moment, the biggest weakness of the new format is the proposed time for each pick and ban, which has been reduced by half, from 60 seconds per pick/ban to 30. If the change in format was meant to induce strategic diversity, this change reduces it just as much. It will hopefully be changed before the season starts; otherwise, panic picks are more likely and coaching staff's plans can be thrown to the wayside very easily if a particularly off-meta or otherwise unusual pick shows up in a draft.

However, it is important to note that overall, the new system makes much more sense in LoL's current competitive ecosystem than implementing Dota 2's system would. Dota's system is meant to compensate for an asymmetrical map and a relative dearth of absolutely necessary powerpicks, neither of which LoL has to deal with.

Gabriel Zoltan-Johan is a News Editor at theScore esports and the head analyst for the University of Toronto League of Legends team. His (public) musings can be found on his Twitter.