I believe that where you get your data from is important. If you don’t know that you can’t use the data. It’s like you are doing a political poll, you can’t just go into one city and ask for their opinion. The rules for data collection must be established in advance and strictly followed, otherwise the data will be biased. You can’t clean them up with mathematical wizardry, not if the survey process was messed up from the start.
So I ask this again, how do these third party services collect the players data?
Don’t you see any problems in that in terms of biased selection?
What is your answer to that? That it was correct back then?
Careful my friend, don’t prove them wrong accidentally. Gonna end up being insulted and mass reported. Especially SvenX is a type of a guy who will never lose. He will stalk your next forum threads and always insult you if you prove him wrong once.
This thread just makes me sad because I suck as Ana and literally never pick her when I play support.
I’ve been playing since May 23, 2016 with an account with over 1,900 hours and I believe I have 7 on Ana. I also don’t play Mercy - I think my hours are 7 on her too.
I am just sending a friendly warning my friend. You are entering a dangerous territory asking them questions that they have no answer to. Cuz he is not going to prove you wrong. He is going to insult you.
Overbuff has the same source as blizzard our account data.
Blizzard is even nastier, you must agree the their collecting data to play the game.
You always have the problem of standard deviation. That the poll is not accurate that is why you make a second poll. If both polls shows roughly the same. The polls should be correct.
Our second poll is blizzard when they agree with some meta statements from overbuff.
You should see overbuff as an indicator what blizzard will probably do in the future.
With empirical data theres no need to calculate the probablility of how often or how likely one or both teams will choose Ana.
The reason why her pickrate is so high is because of how skewed her pickrates are across all ranks. If everyone could play at a Diamond level then her pickrate would be astronomically higher.
Its no question that Ana is picked more and more as you go up in ranks. There’s simply no support like Ana in the game. We had Baptiste on release who could kind of counter anti-nade (and was intended for this) but everyone hated his immortality field. Therefore with no supports in the game who can reliably counter Ana… she will reign supreme across most ranks where players can aim and shoot
Then why did you call it an opinion? You’re only arguing with yourself here.
So overwatch API allows you to take profile data from open profiles which includes playtime and with that playtime there average out of all playtime for each hero. This is confirmed by how the playtime for all heroes adds up to 100% and roles do not add up to exactly 20% 40% 40%.
Basic denial.
And it doesn’t work. Feel free to use math to do what I asked you can’t ^^.
It’s there clear as day.
I enjoy your time and I don’t see why I’d dislike you.
I’ve directly asked you to prove me wrong by asking you to use your method to calculate the distribution of all support heroes for support players so that you get 100% of all heroes. I know it doesn’t
I’ve disagreed and disproven it mathematically and you said it yourself…
So technically you can only disagree. Unless you want to mathematically disprove me.
Also…
This has been proven it can’t happen and you still haven’t corrected it.
PS. it’s 1 3, 1 4, 2 3, and 2 4.
This is at about 6th grade probability maybe a little lower
Yes and there’s a good reason to believe that, in my opinion. That’s the reason why I don’t agree with any balancing threads that try to rely on actual “numbers”.
Do you have a source on that, as far as I know they can only access the public API. The API doesn’t exist anymore. Players have to sign in to “overbuff” for example, with their battlenet account. Players with private accounts will not show up in any of their stats. This makes the data inconclusive at best. Others might call it very misleading.
I want to ask you - do you really see this in-game? Are you suggesting that you hardly ever see Ana in games? Because I see her in pretty much every single game, and from what I’ve read on here, most people see her in pretty much every single game.
Does it not make you question your math when it does not even come remotely close to the reality you are seeing? It’s odd you would suggest 4.44% per support and then play the game and see any connection between this value and the frequency you see Ana.
I would think that might make you question your reasoning?
(Note: since my previous answer seemed to set you off, I am being as nice and polite as possible because I am curious how you reconcile what you see and what you calculate because there is no way they are similar if we are playing the same game).
Blizzard has more data that is correct.
Blizzard can accurately calculate the pickrate on the 3.position after the decimal point.
But Do you need it? does it matter?
You can do nothing with the pickrate.
The only thing you want is to compare it too other heros.
So that you know what is meta.
Brownie has already told you.
You are allowed to doubt overbuff, but you should doubt your own opinion, too.
I personally think Winston is a little weak, but after looking at overbuff and the forum, maybe I am wrong.
Oh, they can probably calculate it further, but that’s not my point.
It’s not about how accurate it is. You are right about that.
My point is that if the data collection is not done correctly, the entire statement is false. The data from Overbuff could be very, very far from the data that Blizzard has.
I have explained several times why the community at large should assume that these statistics are incorrect.
It is obvious that only a special group of players, log into Overbuff with their Battlenet account to participate in the data collection. That’s not how you collect data and analyze it, because it would be wrong from the beginning.
