The viability and strength of a hero is not only decided by that hero’s pickrate. There are many other factors, like how much fun that character is, or how they fit into certain team compositions.
Please come up with some original arguments to explain how your main needs a buff, or how Dva needs a nerf.
People using pickrates as their main strategy to back up their claims is just annoying.
EDIT/DISCLAIMER: I mean specifically pickrate and winrate, not other stats like damage per 10 minutes or other things like that. I almost never see anyone use an overbuff stat that isn’t either pickrate or winrate, and those are the ones that annoy me.
Right, but using damage rates, solo kill rates, and final blow rates, and healing rates should be a damn good way of comparing some heroes.
What SHOULD we use? A bunch of whining forum goers, or ACTUAL data from the game?
I just came out of a thread, where people were arguing that Brigitte did more damage then Mcree. I mean, what the hell do you say to that apart from, Well… we have stats…
If a character is picked across all ranks in the top 5 most played and their stats are waaaay over any other then yes, using overbuff stats is fair. Same if Briggitte has the 7th least kills and the 5th least damage in game and is tied in heals wirh zen. But people still want to nerf her because she is unfun. How about McCree’s flashbang? Oh wait, people are intelligent enough to know how it works and get away from its range. Bringing up stats is fair game as long as they arent the only argument. Also before you accuse me of being a filthy dps main yada yada, ad hominem yada yada i main healers and tanks, especially D.Va.
Win rates are screwed from overbuff anyway, the blizzard API (if you can call it that) just gives what the last hero someone was playing for the win or loss.
This is why Syms rates are so high, since usually people only Sym for point A. and if they hold point A, then they win the game.
If the lose point A, they switch, and the loss doesn’t impact her winrates.
It is why tracer is so low as well, because people switch to tracer as a last minute stall…
Oh, if you REALLY want to stop people from wanting to use pick rates for balancing… let them know how paladins balances their heroes, and what that means to the hero balance. It is funny as all hell.
The upshot of using pick rate to balance heroes is for EVERY DPS you add to your team in Paladins, your win rate goes down.
This includes the FIRST DPS you add.
The highest win rate teams are ALL tanks and supports. Because they have the lowest pick rate…
Now… if people WANT overwatch to follow the same path…
You know Overbuff isn’t just pickrates, right? There’s winrates, E:D ratio, Damage output, healing output, pretty much every stat that is common to many heroes. Is it perfect? No. But without access to the official figures it’s the best we’ve got.
Win rates are screwed from overbuff anyway, the blizzard API (if you can call it that) just gives what the last hero someone was playing for the win or loss.
There isn’t an official blizzard API for Overwatch
All of those sites work by scraping the data of your playoverwatch profile.
So, what Overbuff does is it takes a snapshot of your profile, parses your data on each hero. And then, after a few days it does it again. And then what it does is it compares the difference between snapshots to figure out how many games you played on each hero and with what outcomes. That’s how the winrate is calculated.
This is why Syms rates are so high, since usually people only Sym for point A. and if they hold point A, then they win the game.
If the lose point A, they switch, and the loss doesn’t impact her winrates.
You get credited with every hero you played proportional to the time played. Hero switching does affect Symmetra, but if she played Symmetra for 50% of the game and then lost, she will still receive 0.5 games lost.
It is why tracer is so low as well, because people switch to tracer as a last minute stall…
Last minute stall is only a fraction of total game and it doesn’t affect winrate drastically. The real reason why tracer is so low is because in most ranks, people can’t play her at all.
Pickrate is a perfectly viable indicator of staleness and repetitiveness though. Why play a game if it’s pretty much the same every time? Everyone is going to have different tolerance levels for it, but having the exact same configuration each time does wear on people and cause them to enjoy the game less than they used to.
So? It still gets counted. Her games played also gets increased by the same number, not by 1.
Meaning, if you played Sym for 60% of the game and Hanzo for 40% of the game and then you won the game, your games_won on Hanzo will go up by 0.4 games, on Sym 0.6 games and your games_played will go up 0.6 on Sym and 0.4 on Hanzo. You won’t get credited with a full game played on either hero.
The reason why winrates on Sym, Torb and niche heroes get skewed is related to switching but it’s not exactly what you described. Give me a minute to find an old post of mine that explains it.
The only place where hero switching matters is when it comes to niche picks.
Why? Here’s an example.
You play Symmetra on 1st point of Hybrid or 2CP map. You have 4 minutes. You manage to hold it. At best, you can draw or win the game. The game will last 4 more minutes at most. 4 minutes from the defense round and 4 minutes on the attack round makes the game last 8 minutes at most. It will likely last less than 8 minutes. Now lets say that you win the game. And you’re playing some other hero during the attack round. Not Symmetra. If you win the game, you receive (4 min) / (4 min + how long it took you to attack) games_won on symmetra. So, if it took you 2 minutes to get a tick and win, you’ll receive 4/6 games won on symmetra which is 0.666 games won.
Ok, now lets look at an another scenario.
You play Symmetra on 1st point of Hybrid or 2CP map. You have 4 minutes for 1st point. The point gets breached and you lose it. You likely switch off Sym to a healer instead. Your actual time played on Sym is less than 4 minutes. The defense round now lasts extra 4 or 5 minutes. Regardless if you won or if you lost, the games won or games lost on Symmetra will be adjusted by time played on symmetra divided by the total length of the game. Time played on Sym will be less than 4 minutes and the actual duration of the game will be at least the duration of the defense round + 4 minutes if you don’t even manage to cap the 1st point on attack. For a concrete example where you got breached after 2.5 minutes and switched off sym and then you held them just before the last point, the duration of the defense round will be at least 8 minutes. And then you still get an attack round which can either be 4 minutes if you fail to attack the 1st point or even more. So you have 2.5 minutes divided by at least 12 minutes. Ergo, you will receive 0.2 games lost.
That’s how her winrate gets skewed. Similar thing with Torb. And Orisa since she also often gets picked into those compositions. But that’s why I said. For niche picks, the winrate is not a reliable indicator for those characters.
But for characters such as: Mercy, Zen, Soldier, Tracer, McCree, D.Va, Winston, Reinhardt, Roadhog, Genji, Ana, Zarya that’s not the case. Some of them see more usage on certain maps but most of them see similar usage on defense and offense. Their winrate will usually be a reliable indicator.
It’s negligible. Swapping to Tracer happens in last 30 or less seconds and both the losing and winning team do it. One fully played game on Tracer will outweight the contribution of 10 or so swaps.
Also, where solely using pickrate or solely using winrate isn’t that good an indicator, using them together is. Low pickrate, high winrate = situational hero. High pickrate, good winrate = strong hero. Near 100% pickrate, near 50% winrate = mustpick. Low pickrate, low winrate = garbage.
Statistics in this game never go well for argumentation. They’re really unreliable and dismiss certain heroes power since that effect cannot be measured. For example, the impact of a hack or a Biotic Grenade cannot be measured in statistics and the impact of a player like Widowmaker or Reaper, assassins.
There is an effect though. People are complaining about far less than a percent on the forums.
Hell, I had someone try to argue that Brigitte was OP, since she did more healing than Zen. (she did less than he last week) but when they made that argument, there was 50 health / game in it.
