A + feature? http://dotabuff.com/heroes/anti-mage doesn't you mean that "best versus" and "worst versus"?
^
Those are vs heroes. He means hero winrate with another hero on their team.
Basically the opposite of the "Best Versus" and "Worst "Versus". "Best With" and "Worst With" would be the new ones
It would be a usefull feature, every one always thinks of counter picking but there is little thought put into picking a team that works well together, beyond having having a farming order and lanes that work (at least in my trash tier). I cant imagine it would even be hard to program considering that dotabuff already tracks every hero picked in every match and what team won.
it should be added in game
like a team picks a hero than it pops out some suggestions, maybe this way we would see less 5 carries team but i still doubt it
Oh, my bad then.
Well would be interesting, it'd have to be based on some statistics or something or it could make the games even worse when your team picks pudge to counter huskar or some other dumb shit
Old bump but I checked and this isn't a + feature. Probably because it's not personalized, so it'd be hard to charge for data someone could just copy and paste somewhere else for free. I'd be hugely interested in this and given the current infrastructure of Dotabuff this should be super easy to code up. I know they're probably busy but is there a better place to post suggestions like this so DB staff are more likely to see it?
Sorry for third bump but I'd be really interested in this feature. Is there anyway I could at least contact staff about it? Given the information Dotabuff already collects about matches, it'd be super easy to implement.
Bitte melde dich an um Kommentare zu posten.
I'm sure this has been suggested somewhere else, but a great statistic to offer for heroes would be their winrates with certain heroes on their team, and their win rates vs items built by the enemy team (specifically, whether an instance of the item was built by the enemy team at any point).
It would be interesting to see whether KotL was really the best friend of Phantom Lancer, or if a Medusa really got in the way of an allied Lich. Or to see if Halberd was the best option vs an Ursa, or if a BKB really did make an enemy OD useless.
If these have been considered but aren't going to be implemented, why so? Is it to much processing on the server's end to calculate? Not useful? Thanks for your time!