This bot is in NO WAY ready for use against real players

[Middge]

This post is not intended to cause offense to any developers, but I feel I should put this information out there for anyone who may want to try HearthBuddy. I fully understand this is a beta, and people are still working on this product.

If you are buying this bot with the intention of climbing the ranked latter or even playing other players in casual mode, TURN AWAY (for now). As of right now, the logic is so bad as to OBVIOUSLY be a bot. Either that, or a 6 year old who literally cannot read and understand all the cards. There are so many flaws in the logic that I'm not even sure where to begin...

I've seen the bot use Leeroy Jenkins to charge a lone 2/1 on the enemy field on turn 4.
I've seen the bot fail to attack anything at all despite the opportunity to attack directly or remove enemy monsters.
I've seen the bot completely ignore cards in hand even while ending the turn with FULL MANA.
I've seen the bot use cards like Cruel Taskmaster on a card that was summoned that same turn for no reason, and then end the turn.
I've seen the bot use Unleash the Hounds and then attack a hunter directly even though he obviously has explosive trap down.
I've seen the bot lose consistently to the advanced AI.

These are just some of the things I've seen it do that I can remember off the top of my head. I say these things, not to insult the devs or to discourage people from buying the product, but to put into the light the obvious short comings thus far witnessed.

I am a HUGE fan of Bossland and their products. I use Honorbuddy and Demonbuddy on a regular basis and I am constantly astounded at how well they work. I cannot wait until HearthBuddy can stand next to those products as an equal. Keep up the good work guys!

 

[hankerspace]

Actually, there is card game. Which means, there are 2 variable where bot can't act :
- Luck
- Deck building

Obviously, luck can't be managed, and deck building is the responsibility of the user. If you want to optimize the bot : know which cards are played well and which cards are hazardous at the moment.
In a second way, i think lot of users complaining or not able to have a good winrate with bot are just users who dont know game and who are not able to build their decks well. In a such case, the "guide" section provide some decks idea (i havent take a look, i dont know if they are good, check users reviews). But, obviously, you are a good hearthstone player and thats not an issue for you.
Finally, - I think some people can confirm - bot provide good winrates when deck is good. My accounts reach 40/45% winrate in ranked, that makes me stagnate between rank 15 and 16. 65/70% reached in normal games at level 40.

And, you can't beat Expert AI - without a lot of luck - because the only difference between normal AI and expert AI is the expert AI know your cards in hands and act knowing that. Which makes him better than out HB AI.

As you said, this bot is still a beta. Users know what are the ocnsequences of buying a beta product - which is cheaper than the final product - and most of them HELP devs instead of crying this way. Your title 'This bot is in NO WAY ready for use against real players' is, for me, fully representative of your ability to build a descent deck.

 

[zuabros]

Stop troolling. The objective of this bot is to farm gold. You can do that with the desirable 50% win already. End.

 

[Middge]

Actually, there is card game. Which means, there are 2 variable where bot can't act :
- Luck
- Deck building

Is that to say that you don't intend for this bot to ever support Arena? I feel like it should be possible to develop some logic that allows for intelligent deck building in Arena. I don't think I am alone in thinking that it would be a very attractive feature some day, and possibly a more efficient way to farm gold.

In a second way, i think lot of users complaining or not able to have a good winrate with bot are just users who dont know game and who are not able to build their decks well.

I've achieved between rank 1-5 every season that I've been a part of the HearthStone Beta, and I am legendary this season. So I will take what you say to mean that I don't know how to build a deck that the AI will use well. If that's what you meant, then I agree. I have no idea what cards the AI is proficient with. I will check to see if any other botters have come up with something that can improve my win rate in normal games.

As you said, this bot is still a beta. Users know what are the ocnsequences of buying a beta product - which is cheaper than the final product - and most of them HELP devs instead of crying this way. Your title 'This bot is in NO WAY ready for use against real players' is, for me, fully representative of your ability to build a descent deck.

Again, I'll choose to take your meaning as a lack of understanding of how your AI handles specific cards. It wasn't my intention to sound like I was complaining... I was just putting out my experiences with this bot so far. I never said anything that wasn't true, and I listed quite a few reasons to back what I've said.

Stop troolling. The objective of this bot is to farm gold. You can do that with the desirable 50% win already. End.

Who is trolling?

Sure, farming gold is the primary objective of this bot, and it does it. Inefficiently and slowly, so far. Also, this 50% winrate everyone is talking about has thus far eluded me. I am lucky to achieve 25-30% with the decks I've built so far. Yet, I admit that Hankerspace brings up a good point. I am trying to use decks that work well for me as a human. I should explore decks that work to this bot's logic. I will try some of those (if I can find any) and report back.

 

[steone]

Is that to say that you don't intend for this bot to ever support Arena? I feel like it should be possible to develop some logic that allows for intelligent deck building in Arena. I don't think I am alone in thinking that it would be a very attractive feature some day, and possibly a more efficient way to farm gold.

He never said that, and btw i think it'll be quite easy to support the arena "logic" since it's only a matter of choosing the best "stat card' on each deal. And there is already some list of cards which are available to see what card is good or not in arena (like https://docs.google.com/spreadsheet/pub?key=0AifXEOqTcGcLdFVvWk1GRjVJTHJUaTVLcGViR1RRTFE&gid=11)

I've achieved between rank 1-5 every season that I've been a part of the HearthStone Beta, and I am legendary this season. So I will take what you say to mean that I don't know how to build a deck that the AI will use well. If that's what you meant, then I agree. I have no idea what cards the AI is proficient with. I will check to see if any other botters have come up with something that can improve my win rate in normal games.

Start by reading changelogs and other posts made by user....U'll see which cards are good and which one are not well supported yet.


And besides, if you're ranked legendary don't forget that people are usually "better" than people at lvl 10-15... So it would be normal to get low winrate with a bot... But again check the guide section, there are a few deck available.

 

[shinoowns]

He never said that, and btw i think it'll be quite easy to support the arena "logic" since it's only a matter of choosing the best "stat card' on each deal. And there is already some list of cards which are available to see what card is good or not in arena (like https://docs.google.com/spreadsheet/pub?key=0AifXEOqTcGcLdFVvWk1GRjVJTHJUaTVLcGViR1RRTFE&gid=11)





Start by reading changelogs and other posts made by user....U'll see which cards are good and which one are not well supported yet.


And besides, if you're ranked legendary don't forget that people are usually "better" than people at lvl 10-15... So it would be normal to get low winrate with a bot... But again check the guide section, there are a few deck available.

That's not how you build a deck for arena. You could choose the best 'stat' wise cards every turn. HOWEVER, lets say all of the best 'stat' based cards are 8 costs. Well now you are fucked and can't play anything but your hero power till turn 8. The intelligence of making a deck in arena is VERY specific.

 

[Fibula]

Sure, farming gold is the primary objective of this bot, and it does it. Inefficiently and slowly, so far. Also, this 50% winrate everyone is talking about has thus far eluded me. I am lucky to achieve 25-30% with the decks I've built so far. Yet, I admit that Hankerspace brings up a good point. I am trying to use decks that work well for me as a human. I should explore decks that work to this bot's logic. I will try some of those (if I can find any) and report back.

When I started to use Bot of Stone at first I had 30% win rate. But it got better with every dozen game played. After around 100-200 games it got to 52-55% win rate. It was all with the same deck and bot I did not changed nothing. You need to lower your MMR so that you will face easier opponents. Then your win rate will go up.

I have really weak rogue deck atm and Hearthbuddy gets 50-52% win rate with it. It's not working perfect, but it gets that 50% anyways just because I face really weak people.

 

[Julskinkan]

Stop troolling. The objective of this bot is to farm gold. You can do that with the desirable 50% win already. End.

50%? Im at 80% win rate now with hunter ^^

 

[Mario27]

50%? Im at 80% win rate now with hunter ^^

i gues u play casual and not ranked at ranked you cant have 80% winrate because the bot does sometimes other things then a person like attacking a minion with lower damage and not killing the minion with higher damage in ranked matches it wil cost you the game

also the paladin equality is messed up it trowns out equality when there are no minions and equality is to change all minions health to 1 then kill it with consecration or first wild pyromancer and then equality it needs better board control

 

[Julskinkan]

i gues u play casual and not ranked at ranked you cant have 80% winrate because the bot does sometimes other things then a person like attacking a minion with lower damage and not killing the minion with higher damage in ranked matches it wil cost you the game

the bot is not made for ranked... ofcourse im playing casual

 

[zuabros]

50%? Im at 80% win rate now with hunter ^^

You can get for a little time at 80%. If you win too much, you will rank up, start playing stronger players, and eventually win only 50%.

*ONLY possible exception would be playing at max rank (rank 1 or legendary), ofc not your case.

 

[Mario27]

the bot is not made for ranked... ofcourse im playing casual

ye but the bot should not cast equality when there are no minions its just acting then like a bot because a real person wouldnt trown equality if there are no minions and the bot also kills a lower minion with a high damage minion from my self it should attack the face or a higher priorty minion if there is one on the board and not the lower minion