Discussion in 'Card Hunter General Chat' started by Deepweed, Apr 2, 2016.
Who has broken the 1900 barrier? The best from Kongregate is billiska at 1894.
[GUILD] Power of Fires
wasn't there a consensus that its pretty much impossible now?
It's should be possible to do by manipulating the matchfinding system, something that can be rather easy at high levels due to the lack of players. Slow, but possible. Doing so under normal circumstances, however, would require a deck that wins somewhere around 95% of the time, I'm guessing, and that's going to be rather hard to make for obvious reasons.
I once wrote a CH rating simulator. Input: current rating, (constant) win probability, number of games to play.
Output: average average rating over those games, average max rating over those games.
Can't access a computer for a few weeks atm (only a phone) so can't confirm it but I bet that 90% win rate is more than enough for 1900 rating.
There you have it. Log on when theres superbowl or something and play cardotron dozends of times, coz hes the only shot you got at 90% winrate.
And no need to be coy about it, selective queuing is the cornerstone of any intentional attempt to raise your elo.
Highest I ever got was somewhere in around the 1870 range (don't remember the precise number). The meta site says my best win streak was 24 wins which almost certainly happened at the same time. So probably another 5-7 wins would have put me there?
Your winning strike should have been long enough:
Stexe Max Elo 1904 Best Win Streak 18
ch33seman Max Elo 1903 Best Win Streak 24
BlackSabbath 1900 Best Win Streak 29
I doubt that. All the winning streaks and high elo I saw, where from good players, which played a consistent deck over a longer time and in an excesive way. Black Sabbath for example had a deck, which was perfect for that map cycle and he perfected the way to play it in some hundred games. In the next map cycle his elo dropped very fast, because he tried to hold on to his 'winning' deck, but the maps of that month were different and so he lost a lot.
Instead of waiting for the next superbowl I would wait for the next map cycle, try to find a very consistent deck for that cycle and start playing that deck from the very first day dozens times per day. In the current map cycle for example I would try a well profen burfft (Atomic Fireball), radcannon (Bill's rad bomb+leadership+vampire) or other distance build (Barnum & Friends).
So there is a way to learn about the highest elo ever achieved by a player?
(or only the kongregater ones?)
Hey how easy is selective queuing these days? I plopped in some code a few years ago to stop high rated players from getting too lonely, with the side effect that selective queuing should be a little harder.
You can intercept people of your elo range really easily if you know their ranked habits a bit. And if you see the someone that could give you trouble is ingame, then you prbably wont get matched with that dude. Unless its Nictus.
I've often watched high-elo games then queued for ranked when the game ends. I almost always get either opponent. So there's a way.
Thats why i rarely queue right after my previous match end. Its predictable simply because there arent enough players active simultaneously. I am sure it works at times when the game is busier.
Obernoob is right, the key to hitting 1900 is to nail down the perfect build for a given rotation and then just stomp for a month. This was how we made Power of Fires so devastating back in the "small guild kill squad" days; we'd basically determine what the "power" build of the month was, and then go on a rampage as soon as we had it figured out.
To my knowledge that's how every player who's gotten past 1900 has pulled it off more or less. From memory, I can say that this is definitely how Stexe/Black Sabbath/ch33seman/Darial Storm hit 1900.
I'd also like to add that Cardotron is actually -not- a good method for getting ELO that high, since Cardy will sometimes just get perfect draws and wreck any build that isn't specifically meant to beat him (basically nimbuses/stuff that abuses Cardy's decision-making skills).
Even winning 24 out of 25 games vs cardy is not worth it at higher ELO's; eventually you're only getting 1 ELO for beating Cardy (somewhere in the upper 1700's usually) and you lose 30 ELO for a loss. It becomes too risky. I suppose you could make an anti-cardotron build that abuses the poor AI with nimbuses in order to win every game, and then try to play super-carefully for weeks, but all it would take is one game vs another human to set the player behind another 30 games of cardy farming.
That code is working. There is just the little but of not enough players beeing online. Than it takes a while till the match will begin, but you are still playing against the expected player. But I have no idea how this could be avoided.
@HisRoyalHygiene (ThetianKnight) just broke the 1900 barrier!!!
My current win rate for this season is 94% so you were close
I've had cardo disabled for months. He ruins elo, gives little guild contribution and isn't interesting to play against.
Based on this thread I think my current 1909 is the highest rating ever
Nobody noticed @Mitsuaki sneaked past 1900 today!?
Separate names with a comma.