Yeah, but urf isn't it either, because some champions are so ridiculously far above the rest that one of them can single-handedly carry the game from level 1.
Not really a complete fuck-around mode when you have that threat hanging over you (a strong incentive to not pick something because it seems amusing).
honest to god, I don't think this is a fuck-around mode at all. It looks to me like it was an internal balance tool released into the public.
URF is/was really good I felt for working on reactions and that's something that I'm going to kind of miss. Obviously there are plenty of moments like that in regular games-knock that Lee Sin out of his follow up Q, flash that Malphite ult, spellshield that hook, etc. But because it's just constantly happening it was kind of fun being in some kind of bullet hell where I just completely stop thinking about what to do and let it all just happen naturally.
Not even ashamed to admit it. I loved going into bot games as helicopter hecarim and ganking every lane with a 3 level advantage ~5 minutes into the match.
Not even ashamed to admit it. I loved going into bot games as helicopter hecarim and ganking every lane with a 3 level advantage ~5 minutes into the match.
helicopter hec was hillarious.
raid boss gp, triforce, shiv, all ap was *amazing*.
I'm a fan of Feral Flare. It buffs junglers that needed better item options, but still has the same weaknesses previous shittier iterations had.
I'll still be using spirit stone items most of the time, because they're more flexible and more immediately powerful and encourage a farm / gank / farm strategy instead of farm / farm / farm / oh hey is dragon up strategy.
If you powerfarm with it, it turns into a very strong item
but if you go 8/1/7 as the jungler you just buy very strong items
My big worry about Feral Flare is that it's going to be absolutely broken for lower skill levels. Junglers already get relatively well farmed there because nobody can CS properly, games last a long time because it's rare for teams to close out a game (meaning FF can get well stacked), and kills per team are also generally very high (again, meaning more stacks for FF).
It's apparently got a 70% win rate in Bronze, so, yeah. Strong item down there.
My big worry about Feral Flare is that it's going to be absolutely broken for lower skill levels. Junglers already get relatively well farmed there because nobody can CS properly, games last a long time because it's rare for teams to close out a game (meaning FF can get well stacked), and kills per team are also generally very high (again, meaning more stacks for FF).
It's apparently got a 70% win rate in Bronze, so, yeah. Strong item down there.
How do you calculate the win rate of an item? 70% of winning teams build it? The losing team could have built it too. Is it that 70% of games where only one team bought it resulted in a win for the team buying it?
My big worry about Feral Flare is that it's going to be absolutely broken for lower skill levels. Junglers already get relatively well farmed there because nobody can CS properly, games last a long time because it's rare for teams to close out a game (meaning FF can get well stacked), and kills per team are also generally very high (again, meaning more stacks for FF).
It's apparently got a 70% win rate in Bronze, so, yeah. Strong item down there.
How do you calculate the win rate of an item? 70% of winning teams build it? The losing team could have built it too. Is it that 70% of games where only one team bought it resulted in a win for the team buying it?
They track *everything*, so yeh, what are the odds you win if you build an ff?
My big worry about Feral Flare is that it's going to be absolutely broken for lower skill levels. Junglers already get relatively well farmed there because nobody can CS properly, games last a long time because it's rare for teams to close out a game (meaning FF can get well stacked), and kills per team are also generally very high (again, meaning more stacks for FF).
It's apparently got a 70% win rate in Bronze, so, yeah. Strong item down there.
How do you calculate the win rate of an item? 70% of winning teams build it? The losing team could have built it too. Is it that 70% of games where only one team bought it resulted in a win for the team buying it?
As far as I know, most of the sites doing stats for LoL use a "this is how often a team that has a player that has built this item wins" rule; so yes, the other team in that game could have built the item, too.
My big worry about Feral Flare is that it's going to be absolutely broken for lower skill levels. Junglers already get relatively well farmed there because nobody can CS properly, games last a long time because it's rare for teams to close out a game (meaning FF can get well stacked), and kills per team are also generally very high (again, meaning more stacks for FF).
It's apparently got a 70% win rate in Bronze, so, yeah. Strong item down there.
How do you calculate the win rate of an item? 70% of winning teams build it? The losing team could have built it too. Is it that 70% of games where only one team bought it resulted in a win for the team buying it?
As far as I know, most of the sites doing stats for LoL use a "this is how often a team that has a player that has built this item wins" rule; so yes, the other team in that game could have built the item, too.
But then doesn't that just make the item popular? Dorans blade probably has a 95+% win rate.
70% is a statistical landslide either way it's calculated
I wonder how common it is relative to the spirit items though from division to division....... effort
It's not. If every team always builds an item it has a 100% win rate, but that doesn't make it overpowered. Normal boots fit into that category, aside of course from salty teemo
My big worry about Feral Flare is that it's going to be absolutely broken for lower skill levels. Junglers already get relatively well farmed there because nobody can CS properly, games last a long time because it's rare for teams to close out a game (meaning FF can get well stacked), and kills per team are also generally very high (again, meaning more stacks for FF).
It's apparently got a 70% win rate in Bronze, so, yeah. Strong item down there.
How do you calculate the win rate of an item? 70% of winning teams build it? The losing team could have built it too. Is it that 70% of games where only one team bought it resulted in a win for the team buying it?
As far as I know, most of the sites doing stats for LoL use a "this is how often a team that has a player that has built this item wins" rule; so yes, the other team in that game could have built the item, too.
But then doesn't that just make the item popular? Dorans blade probably has a 95+% win rate.
That's what I was thinking too. Though I haven't checked the stats on that. Would be interesting to see a breakdown of items with associated winrates in the top 10 or so.
Red Potion probably has 99.9% winrate.
We could look at winrate of Feral Flare versus winrate of spirit stone to get an accurate picture.
My big worry about Feral Flare is that it's going to be absolutely broken for lower skill levels. Junglers already get relatively well farmed there because nobody can CS properly, games last a long time because it's rare for teams to close out a game (meaning FF can get well stacked), and kills per team are also generally very high (again, meaning more stacks for FF).
It's apparently got a 70% win rate in Bronze, so, yeah. Strong item down there.
How do you calculate the win rate of an item? 70% of winning teams build it? The losing team could have built it too. Is it that 70% of games where only one team bought it resulted in a win for the team buying it?
As far as I know, most of the sites doing stats for LoL use a "this is how often a team that has a player that has built this item wins" rule; so yes, the other team in that game could have built the item, too.
But then doesn't that just make the item popular? Dorans blade probably has a 95+% win rate.
No, if it was popular it would trend towards 50% (because half of the teams that build it lose). Warding totem has almost exactly a 50% win rate, for instance.
My big worry about Feral Flare is that it's going to be absolutely broken for lower skill levels. Junglers already get relatively well farmed there because nobody can CS properly, games last a long time because it's rare for teams to close out a game (meaning FF can get well stacked), and kills per team are also generally very high (again, meaning more stacks for FF).
It's apparently got a 70% win rate in Bronze, so, yeah. Strong item down there.
How do you calculate the win rate of an item? 70% of winning teams build it? The losing team could have built it too. Is it that 70% of games where only one team bought it resulted in a win for the team buying it?
As far as I know, most of the sites doing stats for LoL use a "this is how often a team that has a player that has built this item wins" rule; so yes, the other team in that game could have built the item, too.
But then doesn't that just make the item popular? Dorans blade probably has a 95+% win rate.
No, if it was popular it would trend towards 50% (because half of the teams that build it lose). Warding totem has almost exactly a 50% win rate, for instance.
That just makes me even more curious about how the stat is calculated. Is it one point if you win with it, one point if you lose with it, and then w/(w+l)?
Skin Codes that have already been used in NA. They may work on the EU/OCE/Turkey servers though. They will unlock the champion if you dont already have it as well. Give a shout if you use them there please
Riot Ward Skin
RWQPHQ4NNHRV4Y
Pax Hecarim
WWE4H3YDXWDM9G
Riot Blitzcrank
WWJ6W2AJCGYFQK
Artic Ops Varus
WWQ9UYG63QDURR
I may have another Arctic Ops Varus which will work on NA servers if my friend doesnt respond soon. The codes all go dead once patch 4.6 goes live.
My big worry about Feral Flare is that it's going to be absolutely broken for lower skill levels. Junglers already get relatively well farmed there because nobody can CS properly, games last a long time because it's rare for teams to close out a game (meaning FF can get well stacked), and kills per team are also generally very high (again, meaning more stacks for FF).
It's apparently got a 70% win rate in Bronze, so, yeah. Strong item down there.
How do you calculate the win rate of an item? 70% of winning teams build it? The losing team could have built it too. Is it that 70% of games where only one team bought it resulted in a win for the team buying it?
As far as I know, most of the sites doing stats for LoL use a "this is how often a team that has a player that has built this item wins" rule; so yes, the other team in that game could have built the item, too.
But then doesn't that just make the item popular? Dorans blade probably has a 95+% win rate.
That's what I was thinking too. Though I haven't checked the stats on that. Would be interesting to see a breakdown of items with associated winrates in the top 10 or so.
Red Potion probably has 99.9% winrate.
We could look at winrate of Feral Flare versus winrate of spirit stone to get an accurate picture.
yeh, obviously some items need to be discounted, like boots, red and blue pots, etc, etc.
but items then there are items that arent built by more than two people in a game, ever, like FF. A 70% wirate in bronze is kindof terrifying for an item. Sounds like only a matter of time before that gets abused at higher levels.
My big worry about Feral Flare is that it's going to be absolutely broken for lower skill levels. Junglers already get relatively well farmed there because nobody can CS properly, games last a long time because it's rare for teams to close out a game (meaning FF can get well stacked), and kills per team are also generally very high (again, meaning more stacks for FF).
It's apparently got a 70% win rate in Bronze, so, yeah. Strong item down there.
How do you calculate the win rate of an item? 70% of winning teams build it? The losing team could have built it too. Is it that 70% of games where only one team bought it resulted in a win for the team buying it?
As far as I know, most of the sites doing stats for LoL use a "this is how often a team that has a player that has built this item wins" rule; so yes, the other team in that game could have built the item, too.
But then doesn't that just make the item popular? Dorans blade probably has a 95+% win rate.
No, if it was popular it would trend towards 50% (because half of the teams that build it lose). Warding totem has almost exactly a 50% win rate, for instance.
If so, then that is a bit confusing.
The calculation has to be done in one of two ways. It is either:
Games in which Feral Flare won / Total games featuring Feral Flare
or
Games in which Feral Flare won/Total number of wins.
From the answers so far, I am confused as to which of these is correct.
70% is a statistical landslide either way it's calculated
I wonder how common it is relative to the spirit items though from division to division....... effort
It's not. If every team always builds an item it has a 100% win rate, but that doesn't make it overpowered. Normal boots fit into that category, aside of course from salty teemo
If you track wins/losses overall the. Popularity pushes win rates to 50% not 100% since the losing team also builds the item for every game the item is purchased on both sides 1 loss and 1 win would be recorded.
As items become less popular the "has item other team does not" win % and the "has item" win% converge.
If an item is wining 70% that is massive. It's especially telling in this case. Generally there is an endogeneity problem with doing item analysis. It that doing well means more items which means the win rate of late game items is inflated compared to early game items (you only can build the closing item if you win). The only way to correct for this would be pretty restrictive(if even possible) as far as I can tell.
But feral flare/wriggles is cheap and bought first so finishing it isnt (as) correlated with the outcome since you're almost assuredly going to get 1650 gold in a game.
My big worry about Feral Flare is that it's going to be absolutely broken for lower skill levels. Junglers already get relatively well farmed there because nobody can CS properly, games last a long time because it's rare for teams to close out a game (meaning FF can get well stacked), and kills per team are also generally very high (again, meaning more stacks for FF).
It's apparently got a 70% win rate in Bronze, so, yeah. Strong item down there.
How do you calculate the win rate of an item? 70% of winning teams build it? The losing team could have built it too. Is it that 70% of games where only one team bought it resulted in a win for the team buying it?
As far as I know, most of the sites doing stats for LoL use a "this is how often a team that has a player that has built this item wins" rule; so yes, the other team in that game could have built the item, too.
But then doesn't that just make the item popular? Dorans blade probably has a 95+% win rate.
No, if it was popular it would trend towards 50% (because half of the teams that build it lose). Warding totem has almost exactly a 50% win rate, for instance.
If so, then that is a bit confusing.
The calculation has to be done in one of two ways. It is either:
Games in which Feral Flare won / Total games featuring Feral Flare
or
Games in which Feral Flare won/Total number of wins.
From the answers so far, I am confused as to which of these is correct.
I think it's the first, "When FF is built, that team wins XX% of the time", and given that 50% is the number if both team build it every game, 70% is an astounding number.
My big worry about Feral Flare is that it's going to be absolutely broken for lower skill levels. Junglers already get relatively well farmed there because nobody can CS properly, games last a long time because it's rare for teams to close out a game (meaning FF can get well stacked), and kills per team are also generally very high (again, meaning more stacks for FF).
It's apparently got a 70% win rate in Bronze, so, yeah. Strong item down there.
How do you calculate the win rate of an item? 70% of winning teams build it? The losing team could have built it too. Is it that 70% of games where only one team bought it resulted in a win for the team buying it?
As far as I know, most of the sites doing stats for LoL use a "this is how often a team that has a player that has built this item wins" rule; so yes, the other team in that game could have built the item, too.
But then doesn't that just make the item popular? Dorans blade probably has a 95+% win rate.
No, if it was popular it would trend towards 50% (because half of the teams that build it lose). Warding totem has almost exactly a 50% win rate, for instance.
That just makes me even more curious about how the stat is calculated. Is it one point if you win with it, one point if you lose with it, and then w/(w+l)?
I think this is what they do. I'm fairly sleep deprived so I'm doing a pretty piss poor job of explaining myself.
I'm also damned if I can figure out how they're calculating item popularity so that warding totem is at ~130%, unless the majority of the game's population is much worse at buying trinkets than I thought.
My big worry about Feral Flare is that it's going yOu to be absolutely broken for lower skill levels. Junglers already get relatively well farmed there because nobody can CS properly, games last a long time because it's rare for teams to close out a game (meaning FF can get well stacked), and kills per team are also generally very high (again, meaning more stacks for FF).
It's apparently got a 70% win rate in Bronze, so, yeah. Strong item down there.
How do you calculate the win rate of an item? 70% of winning teams build it? The losing team could have built it too. Is it that 70% of games where only one team bought it resulted in a win for the team buying it?
As far as I know, most of the sites doing stats for LoL use a "this is how often a team that has a player that has built this item wins" rule; so yes, the other team in that game could have built the item, too.
But then doesn't that just make the item popular? Dorans blade probably has a 95+% win rate.
No, if it was popular it would trend towards 50% (because half of the teams that build it lose). Warding totem has almost exactly a 50% win rate, for instance.
That just makes me even more curious about how the stat is calculated. Is it one point if you win with it, one point if you lose with it, and then w/(w+l)?
I think this is what they do. I'm fairly sleep deprived so I'm doing a pretty piss poor job of explaining myself.
I'm also damned if I can figure out how they're calculating item popularity so that warding totem is at ~130%, unless the majority of the game's population is much worse at buying trinkets than I thought.
Maybe on average 6.5 people per team buy it? It would make sense since sometimes you might sell and buy it again.
70% is a statistical landslide either way it's calculated
I wonder how common it is relative to the spirit items though from division to division....... effort
It's not. If every team always builds an item it has a 100% win rate, but that doesn't make it overpowered. Normal boots fit into that category, aside of course from salty teemo
If you track wins/losses overall the. Popularity pushes win rates to 50% not 100% since the losing team also builds the item for every game the item is purchased on both sides 1 loss and 1 win would be recorded.
As items become less popular the "has item other team does not" win % and the "has item" win% converge.
If an item is wining 70% that is massive. It's especially telling in this case. Generally there is an endogeneity problem with doing item analysis. It that doing well means more items which means the win rate of late game items is inflated compared to early game items (you only can build the closing item if you win). The only way to correct for this would be pretty restrictive(if even possible) as far as I can tell.
But feral flare/wriggles is cheap and bought first so finishing it isnt (as) correlated with the outcome since you're almost assuredly going to get 1650 gold in a game.
I'm under the informed opinion that Riot's analysis is pretty robust, and they can tell if an item/summoner spell is out of line in different stages of the game.
Skin Codes that have already been used in NA. They may work on the EU/OCE/Turkey servers though. They will unlock the champion if you dont already have it as well. Give a shout if you use them there please
Riot Ward Skin
RWQPHQ4NNHRV4Y
Pax Hecarim
WWE4H3YDXWDM9G
Riot Blitzcrank
WWJ6W2AJCGYFQK
Artic Ops Varus
WWQ9UYG63QDURR
I may have another Arctic Ops Varus which will work on NA servers if my friend doesnt respond soon. The codes all go dead once patch 4.6 goes live.
Grabbed Arctic Ops Varus on NA.
Has anyone told you you're a cool dude lately? Because you're a cool dude
Edit: I didn't notice you said they've already been used NA. If you gave it to someone else and they just hadn't used it yet let me know and i'll be happy to buy them the skin in game.
Since I got Arcade Hecarim and Riot Blitzcrank thanks to this thread from a previous PAX, I think it is only appropriate that I do the same for the first person to PM me.
Edit:
Both gone
GONG-00 on
Black lives matter.
Law and Order ≠ Justice
ACNH Island Isla Cero: DA-3082-2045-4142
Still waiting on Dan "Man of his Word" Ryckert to eat a hat
Neither of those works. The first gives a 100% rate to ward totem, and so does the second.
Yeah going through it again, you're right, that can't be it. You could do it by team, to make the the math easier, but that's messier. I think your method also works.
70% is a statistical landslide either way it's calculated
I wonder how common it is relative to the spirit items though from division to division....... effort
It's not. If every team always builds an item it has a 100% win rate, but that doesn't make it overpowered. Normal boots fit into that category, aside of course from salty teemo
If you track wins/losses overall the. Popularity pushes win rates to 50% not 100% since the losing team also builds the item for every game the item is purchased on both sides 1 loss and 1 win would be recorded.
As items become less popular the "has item other team does not" win % and the "has item" win% converge.
If an item is wining 70% that is massive. It's especially telling in this case. Generally there is an endogeneity problem with doing item analysis. It that doing well means more items which means the win rate of late game items is inflated compared to early game items (you only can build the closing item if you win). The only way to correct for this would be pretty restrictive(if even possible) as far as I can tell.
But feral flare/wriggles is cheap and bought first so finishing it isnt (as) correlated with the outcome since you're almost assuredly going to get 1650 gold in a game.
I'm under the informed opinion that Riot's analysis is pretty robust, and they can tell if an item/summoner spell is out of line in different stages of the game.
Its a data limitation. You can look at how powerful different items are based on the duration of the game to get an idea of whether or not an item is better in the lategame but you cannot actually do a statistical analysis of the power of an item because items are stronger and are purchased more often when you're already winning.
Win% = f(Power Item)
Power Item = g(Win%)
If you want to do analysis on items its much easier to just do qualitative analysis. How do you know Feral Flare is ridiculously strong? Feral Flare provides 1760 gold of stats before its passive for 1650 gold. Elder Lizzard provides 1400 gold of stats before its passive for 2000 gold.
Before Feral Flare there was no good attack speed based clear item and we can see that just by looking at items. Keep in mind that in order to be a decent item to clear the jungle it must give attack speed and some sort of damage boost. The best analogue would be Wits End which has 42 magic damage and 42% attack speed. It costs 2400 gold. New Wriggles has 12 attack damage, 30% attack speed, and 100 damage to monsters and 10 HP/attack for 1650.
You know Feral Flare is hilariously strong because its more powerful than the offensive capabilities that Wits Ends gives you (which was the quintessential bruiser attack speed item until the more slot efficient and sustain containing BotRK came around) for cheaper while also infinite scaling and being a ward and having sustain for amazing slot efficiency. You don't have to do a statistical analysis on it.
i'm fairly certain that if you grabbed the win rate for spirit stone, what you'd actually be seeing is the win rate of people who never upgraded their spirit stone
that's one of the tricks of checking the winrates of early game items - doran's items will be lowered because in long games they get sold, boots will be low because they get upgraded, etc
My big worry about Feral Flare is that it's going to be absolutely broken for lower skill levels. Junglers already get relatively well farmed there because nobody can CS properly, games last a long time because it's rare for teams to close out a game (meaning FF can get well stacked), and kills per team are also generally very high (again, meaning more stacks for FF).
It's apparently got a 70% win rate in Bronze, so, yeah. Strong item down there.
How do you calculate the win rate of an item? 70% of winning teams build it? The losing team could have built it too. Is it that 70% of games where only one team bought it resulted in a win for the team buying it?
As far as I know, most of the sites doing stats for LoL use a "this is how often a team that has a player that has built this item wins" rule; so yes, the other team in that game could have built the item, too.
But then doesn't that just make the item popular? Dorans blade probably has a 95+% win rate.
No, if it was popular it would trend towards 50% (because half of the teams that build it lose). Warding totem has almost exactly a 50% win rate, for instance.
If so, then that is a bit confusing.
The calculation has to be done in one of two ways. It is either:
Games in which Feral Flare won / Total games featuring Feral Flare
or
Games in which Feral Flare won/Total number of wins.
From the answers so far, I am confused as to which of these is correct.
There are three ways
1) Games Feral Flare Won/Total Games Featuring Feral Flare
In this a game in which Feral Flare was on both sides would results in 2 total games and so as popularity goes up this number converges to 50%.
2) Games in which feral flare won/ total number of wins
for this popularity pushes the number to 100%. This is a pretty useless statistic
3) Games in which feral flare won when the other side did not have feral flare / total number of games containing feral flare on one side
This is basically "raw power against a non feral flare jungler" as popularity goes up this number stays constant in expectation but increases in variance (because the total number of games is decreasing)
My big worry about Feral Flare is that it's going to be absolutely broken for lower skill levels. Junglers already get relatively well farmed there because nobody can CS properly, games last a long time because it's rare for teams to close out a game (meaning FF can get well stacked), and kills per team are also generally very high (again, meaning more stacks for FF).
It's apparently got a 70% win rate in Bronze, so, yeah. Strong item down there.
How do you calculate the win rate of an item? 70% of winning teams build it? The losing team could have built it too. Is it that 70% of games where only one team bought it resulted in a win for the team buying it?
As far as I know, most of the sites doing stats for LoL use a "this is how often a team that has a player that has built this item wins" rule; so yes, the other team in that game could have built the item, too.
But then doesn't that just make the item popular? Dorans blade probably has a 95+% win rate.
No, if it was popular it would trend towards 50% (because half of the teams that build it lose). Warding totem has almost exactly a 50% win rate, for instance.
If so, then that is a bit confusing.
The calculation has to be done in one of two ways. It is either:
Games in which Feral Flare won / Total games featuring Feral Flare
or
Games in which Feral Flare won/Total number of wins.
From the answers so far, I am confused as to which of these is correct.
There are three ways
1) Games Feral Flare Won/Total Games Featuring Feral Flare
In this a game in which Feral Flare was on both sides would results in 2 total games and so as popularity goes up this number converges to 50%.
2) Games in which feral flare won/ total number of wins
for this popularity pushes the number to 100%. This is a pretty useless statistic
3) Games in which feral flare won when the other side did not have feral flare / total number of games containing feral flare on one side
This is basically "raw power against a non feral flare jungler" as popularity goes up this number stays constant in expectation but increases in variance (because the total number of games is decreasing)
63% win rate linked above is kinda telling. That's Ezreal level season 2 numbers.
More interesting is the wriggles win stats actually fell. I guess this is because
1) Feral actually doesn't get finished in every game
2) games in which wriggles transforms aren't counted
This makes the analysis a bit harder because its evidence of the endogeneity i talked about earlier that throws off stats. Granted this only happens in 2.2% of all games (that someone completes wriggles but does not transform it) so the effect might be pretty small.
for a game where wriggles is built but not FF it pretty much just has to be games where the losing team either surrendered or got rolled in 25 minutes, i'd think.
for a game where wriggles is built but not FF it pretty much just has to be games where the losing team either surrendered or got rolled in 25 minutes, i'd think.
Wriggles still has a 33% win rate but yes It should be amazingly hard to win without transforming wriggles because winning means more access to jungle creeps which means a faster transformation.
Posts
honest to god, I don't think this is a fuck-around mode at all. It looks to me like it was an internal balance tool released into the public.
Joe's Stream.
helicopter hec was hillarious.
raid boss gp, triforce, shiv, all ap was *amazing*.
Raid Boss taric was a god as well.
Joe's Stream.
I'll still be using spirit stone items most of the time, because they're more flexible and more immediately powerful and encourage a farm / gank / farm strategy instead of farm / farm / farm / oh hey is dragon up strategy.
If you powerfarm with it, it turns into a very strong item
but if you go 8/1/7 as the jungler you just buy very strong items
It's apparently got a 70% win rate in Bronze, so, yeah. Strong item down there.
How do you calculate the win rate of an item? 70% of winning teams build it? The losing team could have built it too. Is it that 70% of games where only one team bought it resulted in a win for the team buying it?
They track *everything*, so yeh, what are the odds you win if you build an ff?
70% is kinda bananas.
(time to schedule some time abusing ff)
Joe's Stream.
As far as I know, most of the sites doing stats for LoL use a "this is how often a team that has a player that has built this item wins" rule; so yes, the other team in that game could have built the item, too.
I wonder how common it is relative to the spirit items though from division to division....... effort
But then doesn't that just make the item popular? Dorans blade probably has a 95+% win rate.
It's not. If every team always builds an item it has a 100% win rate, but that doesn't make it overpowered. Normal boots fit into that category, aside of course from salty teemo
That's what I was thinking too. Though I haven't checked the stats on that. Would be interesting to see a breakdown of items with associated winrates in the top 10 or so.
Red Potion probably has 99.9% winrate.
We could look at winrate of Feral Flare versus winrate of spirit stone to get an accurate picture.
No, if it was popular it would trend towards 50% (because half of the teams that build it lose). Warding totem has almost exactly a 50% win rate, for instance.
That just makes me even more curious about how the stat is calculated. Is it one point if you win with it, one point if you lose with it, and then w/(w+l)?
Riot Ward Skin
Pax Hecarim
Riot Blitzcrank
Artic Ops Varus
I may have another Arctic Ops Varus which will work on NA servers if my friend doesnt respond soon. The codes all go dead once patch 4.6 goes live.
yeh, obviously some items need to be discounted, like boots, red and blue pots, etc, etc.
but items then there are items that arent built by more than two people in a game, ever, like FF. A 70% wirate in bronze is kindof terrifying for an item. Sounds like only a matter of time before that gets abused at higher levels.
Joe's Stream.
If so, then that is a bit confusing.
The calculation has to be done in one of two ways. It is either:
Games in which Feral Flare won / Total games featuring Feral Flare
or
Games in which Feral Flare won/Total number of wins.
From the answers so far, I am confused as to which of these is correct.
If you track wins/losses overall the. Popularity pushes win rates to 50% not 100% since the losing team also builds the item for every game the item is purchased on both sides 1 loss and 1 win would be recorded.
As items become less popular the "has item other team does not" win % and the "has item" win% converge.
If an item is wining 70% that is massive. It's especially telling in this case. Generally there is an endogeneity problem with doing item analysis. It that doing well means more items which means the win rate of late game items is inflated compared to early game items (you only can build the closing item if you win). The only way to correct for this would be pretty restrictive(if even possible) as far as I can tell.
But feral flare/wriggles is cheap and bought first so finishing it isnt (as) correlated with the outcome since you're almost assuredly going to get 1650 gold in a game.
I think it's the first, "When FF is built, that team wins XX% of the time", and given that 50% is the number if both team build it every game, 70% is an astounding number.
Joe's Stream.
I think this is what they do. I'm fairly sleep deprived so I'm doing a pretty piss poor job of explaining myself.
I'm also damned if I can figure out how they're calculating item popularity so that warding totem is at ~130%, unless the majority of the game's population is much worse at buying trinkets than I thought.
Maybe on average 6.5 people per team buy it? It would make sense since sometimes you might sell and buy it again.
But that would seem absurdly high.
I'm under the informed opinion that Riot's analysis is pretty robust, and they can tell if an item/summoner spell is out of line in different stages of the game.
Joe's Stream.
Grabbed Arctic Ops Varus on NA.
Has anyone told you you're a cool dude lately? Because you're a cool dude
Edit: I didn't notice you said they've already been used NA. If you gave it to someone else and they just hadn't used it yet let me know and i'll be happy to buy them the skin in game.
Edit:
Both gone
Law and Order ≠ Justice
ACNH Island Isla Cero: DA-3082-2045-4142
Still waiting on Dan "Man of his Word" Ryckert to eat a hat
Yeah going through it again, you're right, that can't be it. You could do it by team, to make the the math easier, but that's messier. I think your method also works.
Its a data limitation. You can look at how powerful different items are based on the duration of the game to get an idea of whether or not an item is better in the lategame but you cannot actually do a statistical analysis of the power of an item because items are stronger and are purchased more often when you're already winning.
Win% = f(Power Item)
Power Item = g(Win%)
If you want to do analysis on items its much easier to just do qualitative analysis. How do you know Feral Flare is ridiculously strong? Feral Flare provides 1760 gold of stats before its passive for 1650 gold. Elder Lizzard provides 1400 gold of stats before its passive for 2000 gold.
Before Feral Flare there was no good attack speed based clear item and we can see that just by looking at items. Keep in mind that in order to be a decent item to clear the jungle it must give attack speed and some sort of damage boost. The best analogue would be Wits End which has 42 magic damage and 42% attack speed. It costs 2400 gold. New Wriggles has 12 attack damage, 30% attack speed, and 100 damage to monsters and 10 HP/attack for 1650.
You know Feral Flare is hilariously strong because its more powerful than the offensive capabilities that Wits Ends gives you (which was the quintessential bruiser attack speed item until the more slot efficient and sustain containing BotRK came around) for cheaper while also infinite scaling and being a ward and having sustain for amazing slot efficiency. You don't have to do a statistical analysis on it.
that's one of the tricks of checking the winrates of early game items - doran's items will be lowered because in long games they get sold, boots will be low because they get upgraded, etc
From LoLKing:
Spirit Stone: ~40%
Feral Flare: ~63%
SotAG: ~50%
SotEL: ~50%
SotSW: ~50%
that is hmm.....
yeah. ty feral flare based god.
There are three ways
1) Games Feral Flare Won/Total Games Featuring Feral Flare
In this a game in which Feral Flare was on both sides would results in 2 total games and so as popularity goes up this number converges to 50%.
2) Games in which feral flare won/ total number of wins
for this popularity pushes the number to 100%. This is a pretty useless statistic
3) Games in which feral flare won when the other side did not have feral flare / total number of games containing feral flare on one side
This is basically "raw power against a non feral flare jungler" as popularity goes up this number stays constant in expectation but increases in variance (because the total number of games is decreasing)
63% win rate linked above is kinda telling. That's Ezreal level season 2 numbers.
1) Feral actually doesn't get finished in every game
2) games in which wriggles transforms aren't counted
This makes the analysis a bit harder because its evidence of the endogeneity i talked about earlier that throws off stats. Granted this only happens in 2.2% of all games (that someone completes wriggles but does not transform it) so the effect might be pretty small.
I'm down to 5 Riot Blitzs, 6 Arcade Hec, & 2 1 Arctic Varus. I've been trying to match up an NA PM with a EU PM, so keep them coming.
Riot Ward Skin Contest: Make me a haiku about your favorite lane, or jungle camp. PM's only please, and I'll post the best ones.
Wriggles still has a 33% win rate but yes It should be amazingly hard to win without transforming wriggles because winning means more access to jungle creeps which means a faster transformation.
My gut reaction is the damage is a bit too much for too little effort but I'm hardly an expert.