Yeah the site might not be the most accurate or most up-to-date, but still gives an OK read on the state of heroes. The trends of Overbuff have been the same for multiple years.
I wish Blizzard gives us an official statistics website though, similar to Rivals.
Yeah but even those mentioned stats (as inaccurate as they might be), still reflect reality to an extent.
Most of those heroes are unpopular and/or obnoxious, and they also happen to perform poorly.
Devs have to be more proactive when it comes to fixing them.
To be honest, the hero design team seems scared and play it very safe.
There are tons of ways to try and fix those heroes in a cost-effective way. No new VFX needed.
it doesn’t tell how good a hero is, it tells u how good the general population is with them. having a bad pickrate makes it even more unlikely to be good data
That’s not how statistics work. What matters is the sample size. If the sample size is big, and some heroes have a low pick rate despite that large sample size, it means that hero is unpopular. It doesn’t mean the data is unreliable.
Yeah but the game is made for the general population. If the hero doesn’t fit the general population, it means the hero sucks.
Heroes are not made for aliens.
low pickrate means precisely a lower sample size, what are u on aboutXD
buff a hero that doesnt need it cause players are bad with it and watch as ppl make ur matches unplayable by abusing said hero. if u want to do data analysis go ahead but dont get upset when ppl call out bad methodology
Bro.
The sample size is not exclusive to the heroes with a low pick rate.
If we take a sample of 1 million OW players, and only 1000 of them play Symmetra, this means that Symmetra is only played by 0.01% of the players. It doesn’t mean the symmetra pick rate stat is unreliable.
Please ask google or chat gpt what sample size means.
Most of the time, buffs are not needed. Struggling heroes need reworks, not buffs.
yeah thats not pickrate tho, pickrates doesnt tell u how many players play what. u dont know the first thing bout data analysis and dont even understand the data u are looking at, but by all means keep making posts bout pick and win rates
i dont need chatgpt, i ACTUALLY KNOW, plus if i needed i wouldnt use chatgipity for something so basicXD
debatable, bastion and echo are def not struggling
No they do not. They’ve gone through multiple periods where data was missing. They are currently missing multiple game-changing factors. Not including Hazard’s data or including perks are huge indicators that their data is likely unreliable. If I recall correctly they also lagged on adding other heroes before as well.
Please stop putting so much stock into Overbuff. It’s really, genuinely not that reliable. I wish it was but it is not.
ill tell u right now that a sample size of 1k is not that reliable
and as someone else already told u, its unreliable data, which can be worse than no data
means the data is less reliable than a high pickrate hero and u went for the lowest pickrates.
considering the unreliable nature of the data, u went for the smallest samples sizes possible,no time series, no rank stratification, prety easy to see that its a useless venture to try and gleam any insight from this
take