Boards›Gameplay

Does win rate indicate champion balance?

Familiar Faces·3/20/2017, 9:06:08 PM·4 votes·2,195 views

A common factor to when people perceive a champion to be over powered is win rate. By and large when a champion is nerfed, it's win rate tends to dip and visa versa when buffed. We can all agree on this. But I'd like to propose that win rate should not be the primary factor when considering whether a champion is over powered to an unhealthy extent. In patch 7.2 a certain champ had a win-rate of 40.6% which was really below average, bottom tier. That champ was picked 7.9% which is more than average. This same champion had a ban rate of 47.8%. Guess who! <-- a link to a champions.pickban.com page Yup.., The swordsman whom many of us would agree is by no means remotely under powered had a 40.6% win rate (one of the lowest win rates) and a ban rate of 47.8% (one of the highest ban rates). Think about that a minute.

Here's what win rates really inform. Win rate reflects how well PLAYERS of that champion exploit that champion. That's it. How overbearing is the exploit, how practiced is the exploit and how accessible is the exploit go into win rate. What of ban rates? Ban rates are suggestive of how difficult players feel that CHAMPION is to counter play vs.

Might not be the case for win rates but for every case a champion's ban rate was notably higher than average, that champion had a kit and itemisation that were considered widely unhealthy for the game. Whether it may be a key skillset with negligible counter play or over powered damage/ damage mitigation values.., a champion banned more often than usual is brute force testingly considered as being some degree of unhealthy to the enjoyment of a match.

TL;DR Win rates are ok indicators of when something is over powered but Ban rates do a better job at highlighting what is unanimously considered unhealthy.

12 Comments

ZephyrDrake3/20/2017, 9:27:19 PM4 votes

winrate by itself means little but it can be used as a sign that something might be wrong with the champion (whether it's being OP or UP)

Incognonymous3/20/2017, 9:23:17 PM3 votes

It's always part of the calculation, but perhaps not in a vacuum. I started playing in S2. From S2 through S5, it wasn't that unusual for tank junglers to have seemingly high winrates, but it wasn't because of the effects of an individual carrying, rather, an individual providing their team with the tools necessary to win. This obviously marks a hard line between teams who do and do not work together to obtain victory. In that respect, win rate is only useful when other factors are taken into consideration.

Glîtchy3/20/2017, 9:29:43 PM3 votes

To some degree yes, as there have been times where Zed was at like 70% winrate and Kassadin was at a 100% ban rate (obviously for good reason) and Talon right now since you just roll your face on the keyboard to kill something.

MissingKayo3/21/2017, 12:48:42 AM2 votes

Not by itself. Too many potential factors that contribute.

Avienal3/20/2017, 9:16:10 PM2 votes

[{quoted}](name=A Shadow,realm=NA,application-id=3ErqAdtq,discussion-id=QYbcsz6I,comment-id=,timestamp=2017-03-20T21:06:08.060+0000)

A common factor to when people perceive a champion to be over powered is win rate. By and large when a champion is nerfed, it's win rate tends to dip and visa versa when buffed. We can all agree on this. But I'd like to propose that win rate should not be the primary factor when considering whether a champion is over powered to an unhealthy extent. In patch 7.2 a certain champ had a win-rate of 40.6% which was really below average, bottom tier. That champ was picked 7.9% which is more than average. This same champion had a ban rate of 47.8%. Guess who! <-- a link to a champions.pickban.com page Yup.., The swordsman whom many of us would agree is by no means remotely under powered had a 40.6% win rate (one of the lowest win rates) and a ban rate of 47.8% (one of the highest ban rates). Think about that a minute.

Here's what win rates really inform. Win rate reflects how well PLAYERS of that champion exploit that champion. That's it. How overbearing is the exploit, how practiced is the exploit and how accessible is the exploit go into win rate. What of ban rates? Ban rates are suggestive of how difficult players feel that CHAMPION is to counter play vs.

Might not be the case for win rates but for every case a champion's ban rate was notably higher than average, that champion had a kit and itemisation that were considered widely unhealthy for the game. Whether it may be a key skillset with negligible counter play or over powered damage/ damage mitigation values.., a champion banned more often than usual is brute force testingly considered as being some degree of unhealthy to the enjoyment of a match.

TL;DR Win rates are ok indicators of when something is over powered but Ban rates do a better job at highlighting what is unanimously considered unhealthy.

Its really a mixed package that involves alot of lazy B.S.ing, people could easily have a champion have a low win rate because people spam the champion because 'omg diz champ so opz! Must Playz and Be a Supa Pro Elit3zest!', logic ends up kicking in on them and it just results in that champions win rate dropping because so many mooks end up blurring the proper 'overpowered lines'.

If Riot can't take the proper time to address champions who have been constant pub stomping nuances for several seasons, then its clear they are barely giving much of a care on breaking up lethality ADC builds either, because they gave black cleaver health and only scratched its AD, when really it needs to be properly evicted as a specific situation type of item and not become a go to grab for the majority of top lane, basically half the assassins/divers between Zed LeeSin Pantheon Graves an so on and all the scummy adcs known as Lucian and MissFortune who still feel completely unphased from the lethality drop on item 3142 , the scrape on lethality on masteries and even the lost of AD on item 3071 , they still exploit the hell out of not only the lethality that is still present but the move speed active on Ghostblade (Lucian R abuse be retared an all that), alongside the shred to easily get away with 6~20 armor shred on even squishie un'armored targets' after ripping off a good sum of the armor with lethality (if lethality is applied first, which would likely be more broken if it was applied after % shred i guess).

Only effective way i can see them shutting black cleaver off of Asssassins and Marksman is to make it a melee only item to start with an scrap a good deal of the raw stats since 20% CDR, the 'move speed on-hit effect' and the 5% armor shred per hit up to 6 times is just too impactful in this snowball meta.

[slayer-jinx-unamused]

Troy2426213/20/2017, 9:58:59 PM2 votes

Win rates are an important facet of determining balance, but there are confounding factors that need to be accounted for.

Emmy Cha0s3/20/2017, 10:30:32 PM2 votes

Champion winrate rarely has anything to do with balance. It's the sum of nth other factors that the average sheep intentionally overlooks. "How often does champion build X item? How often do they spec into Y mastery tree? How often do they take Z runes? How often are they picked against L character? How often are they picked with M character? What is the winrate with any of those? What is it without them?"

When you isolate the numbers and look at them individually, you start to notice a pattern. Let's say for example, support character S typically builds non-support item L. Her winrate with L may be around 58%, yet item L's average winrate across all champions could be 65%. So S is actually underperforming with L, comparatively. However without L, S's winrate falls all the way down to 43%.

Even if S's average winrate is something like 53%, that's solely because it's being artificially inflated by a blatantly overpowered item. Regardless of this fact, the bleating masses will still look solely at the sum 53% and ignore all of the contributing factors. It's truly sad and it makes me disheartened as somebody who enjoys math.

Of course, that's just looking at 1 major factor in a specific case. There may also be other contributors like E or Sh and when these are isolated, S's winrate may even fall to an abysmal 28%. Rito's statistics team are the ones who are supposed to look at all of these and hand the findings over to "balance," but as a stats employee once told me, "Whether they actually listen to those numbers is a different story." We all know they do not.

DunkinNoobs3/21/2017, 8:27:26 AM1 votes

"Winrate matters when it supports my argument and doesn't matter when it contradicts my argument." - Everyone ever.