Best Bans by Tier in Each Region (Patch 6.8): A Statistical Analysis - Taric Edition

Aqua Dragon·Apr 26, 2016·74 votes·7,343 views

The way people choose bans is based heavily in frustration, popular opinion, and potential power. But these often lead to inaccurate choices to maximize the chances of winning.

So Jonnyy9, warwickofwallst, and I decided to make www.bestbans.com to calculate a ban list for each tier solely based on winrate, pickrate, and banrate which is partially added to pickrate.

Nothing else.

As a rule of thumb, optimal ban strategy i.e. the strategy that maximizes your probability of winning is to ban champions that are both high win rate and high play rate. In reality, ban rate is not only a function of win rate and play rate, but also of perception of power, transparency of power, frustration, and risk-aversion. A few months ago when LeBlanc held the title of most banned, she was a suboptimal ban outside of Master/Challenger. Janna, statistically one of the best bans in the game, was banned in less than 1% of games."

> ~Riot Jules

The pictures are data from North America, Ranked. Data for any region may be found at the Best Bans website, and adjusted at the top right corner.

####These are what should be banned, not what are being banned.

####Ban rate is taken into account.

####Always look at your team's pre-picks when deciding a ban; some compositions handle particular champions better.

> Bronze

> Silver

> Gold

> Platinum

> Diamond

Full table of champions on www.bestbans.com

These pictures are only for NA. Data for most regions found on the site.

#Observations

Ekko was super dominant the previous patch. Ekko is still super dominant, being within the top 3 in four different tiers. But don't be fooled; the nerf hit hard. Ekko fell from having an Influence rating of 400 down to a much more reasonable (if overtuned) 150 Influence. Their adjusted pickrate was as high as 99% in Bronze, and now it's only 60%.

Taric meanwhile looks like they're the star hitter for this patch analysis. Taric takes a two two spot in three tiers, though Taric's overall influence begins to fall as elo increases. One problem may be in simply understanding what's happening; their particle effects are so shiny even I have trouble telling the ult is going out, let alone someone who may not scour every patch note. Taric has some incoming nerfs on the PBE so this won't last too long.

Soraka is among the top 4 champions of four tiers, making them continue their multi-patch streak of being one of the most consistent champions in the game. Soraka has PBE nerfs as well, doubly so because Grievous Wounds will be changed in the next patch which disappoints me since I hate GW from a design point. But disregard my grievances; it will likely tank Soraka's influence.

Fizz isn't quite in the top bans, but we're starting to see them appear among the top 12 in most tiers. With the Mage Rework right around the corner (MALZAHAR MAH BOI), all of whom are still extremely weak to close-range assassination, we may see Fizz take the top spot in the following patch.

Kindred isn't picked too much, but their winrate is extremely high, hovering around 55% when picked in Diamond. Interestingly, Kindred is a champion whose consistency scales very linearly with elo, or at least that was the case until recently. If the PBE Kindred nerfs go through, we may see their overall influence in every tier fall.

Zed meanwhile suffers the brunt of the banning when Fizz is likely more deserving. Zed's influence has only been falling and falling, being the least consistent champion in Bronze with a measly 46% winrate, and barely reaching 40 influence in Diamond. Despite this, their banrate is an enormous 50%. When Fizz inevitably rises, people will switch bans pretty fast.

Master Yi is still banned fairly often, but they don't even appear in the top 12 in any tier except once in Diamond. It takes people a while to catch up it seems.

Blitzcrank's ban rate continues to hover around 12% but has otherwise stopped rising. This might be because Soraka is banned in 57% of games anyway, which Blitzcrank rose up to handle. When the nerfs on Sorak hit, Blitz will likely go to a more dormant banrate again (sadly).

Malphite, Graves, Udyr, Maokai, Darius, and Nidalee all have banrates above 10%. Some of these were terrors in earlier patches, but some are notably interesting like Darius, who is unexpected.

The list is what should be banned, not what is being banned.

Because the list is based on averages, the list is most useful when you know very little about either team or know little about compositions. Influence is defined as: How many times you will lose to a champion per 10,000 games compared to the average.

As with any bans, you only get the true benefit by making sure your team isn't intending to play the banned champion. Otherwise, you are denying the enemy team AND your team the chance of playing a consistent power.

The list does not show what is good, strong, or overpowered. It is an evaluation of which champions are the most consistent. To be consistent is different from being strong or good. It means a champion has both a good winrate and a good pickrate so that, on average, you will lose to the champion more often than you win against them.

Many champions are considered strong because their potential power is very high (Azir) but if people can't tap into that strength, then even a strong champ is inconsistent.

Similarly, even champions considered manageable or decent (Blitzcrank) can have extreme consistency that makes them worthwhile to ban.

#Why use these suggestions?

It bears repeating that the list isn't intended to replace specific banning, but is rather intended for use in an information vacuum. You should largely follow the list when you have little understanding of compositions or don't know what champions either team is running, a fairly common occurrence.

####Legitimate reasons for circumventing these suggested bans:

The enemy can pick a champion that counters a composition your team has planned
You know for sure yourself or your team is playing a champion that counters a suggested ban Zed isn't so scary when you know Malzahar is on your team
There's a high chance the enemy team has a champion specialist who will be much weaker if their primary champion is banned

####Reasons to circumvent bans that are not legitimate:

Because a champion is "overpowered." Bans should not be based on a champion's strength, but on their win consistency. Even if Tahm Kench could theoretically win 100% of the time with perfect play, that situation is so rare that it doesn't change that Kench wins only 46% of the time on average. Winrate reflects consistency, not strength.
Because your teammates will be annoyed. Let’s assume that your teammates get ticked off every time Tahm Kench is picked by the enemies. Even with this “buff” caused by annoyed teammates, Kench still only wins 46% of the time. Let your teammates be annoyed; avoiding the ban is still likely the most statistically advantageous chance of success. Reconsider only if it not banning a champion is very likely to put a teammate on extreme tilt.
Because a champion is annoying to fight. As annoying as certain champions are, if you're trying to maximize your winrate then it's still not a smart idea to ban them simply for being obnoxious. Most obnoxious champions have crippling weaknesses that cause their winrates and/or pickrates to be fairly low. Only if the frustration a champion causes is significant enough to impact your winrate should it even be a consideration.
Because you want to ban champions from your own team. If your teammate pre-picks a champion, you can always look up your own teammates and see if their history on the champion defies the average. If your teammate doesn't pre-pick, then you deny the enemy team the same chance of picking the banned champion which will work in your favor regardless. ...also, the champions you think you should be banning from your own team Yasuo, Vayne, Zed, new champions aren't the right choices anyway.

#Methodology

All information is compiled over a four day average from op.gg. In specific regions, Lolking provides more accurate pickrate data per tier, and data is used from Lolking in those instances. The data is calculated every day around midnight GMT-7.

Influence is defined as: How many times you will lose to a champion per 1000 games compared to the average.

The Influence calculation is done as follows:

10,000 * (WR - 50%) * (PR / (100% - BR)

Thanks for reading! I hope you will find this useful.