Lack of good Constructed decks?!

Hi Guys,

I wanted to share some of my thoughts and see if there are more people out there who feel the same.
For the last weeks, ever since TGT has come out I've been trying to climb the ladders with our buddy.

But the lack of updated and well constructed decks on this forum is starting to get disturbing.
The only updated one, rank 4 paladin works in the last few days of the month, and on the EU it only got me to rank 10.

So instead of complaining about this, I have been trying to create some good decks, without any success do at the moment. Ive dropped back from 10 to rank 18.
I hope if anyone out there has this same problem, share your thoughts, if you have good decks that are working on ladder, PLEASE share them!!

I know there are plenty decks out there, only the most on this forum are for golden hero farming. I would really like to see some well constructed decks
that work in this current Meta!

Again, I will do my best to make some decks work, and if they do I will share them here, as this forum really needs them.

greetings,

Fiber

 

Well, let's set up some good conditions first.

1 - The bot plays control much better than agro archtypes.
2 - Control only works in the current meta with good board clear (because agro is so rampant).
3 - You need 1/2 tech cards to deal with big drops (usually).
4 - If you're control, you need to have an end game plan (which the bot can play well).
5 - The bot sucks with traps.

Now that we've covered that, how many decks are really left?

Midrange pally we've discussed inside out. Myself and Sakura have had a lot of input into the deck floating around there.
Dragon Priest has all of the above (but I do not know how well it plays it).
Midrange Hunter will work REALLY well once King's Elekk is fixed.
Mech Mage.
Midrange Druid works well. Bot plays combo well.
Zoo Lock wins more than it loses.

Try all of the above. And let me know how Dragon Priest plays.

Note: I can't promise any of them taking you above rank 5. But I know the Pally one can.

 

I just tried Agro Druid that started doing good and had some really good damage combo's, but then it got me stuck on rank 15.
At the moment im trying out Murrrlock, this is going decent but I dont think good enough to get me past rank 15 I'm afraid.
I will try out Dragon Priest next and keep you guys posted.

How about this constructed mage, seems to be a new concept.. can it perform well with the bot?!
source: https://www.youtube.com/watch?v=wY9nQL0avhM

 

Without being harsh, if you're 'doing good' until you hit Rank 15, then it stops working, then it's not 'doing good'. Other people are just being bad.

Monthly goal should be Rank 5 to get the golden epic for dust. If you're not achieving that, then you're not really 'doing good'. Not being an elitist jerk, just reminding people of what the goal is.

Also, post links to builds, not youtube, I'm not watching a video when I could look over a build in 30 seconds.

 

good comment and I totally agree. Doing good needs to be on the long run. I did say, it started doing well untill...
Maybe My comment should have been, it did ok untill rank 15 and then it sucked balls!!

Hence the reason I didnt post the decklist. as it didnt even come close to rank 5. And its def, the rank im streiving for.
Like you said Pally seems to be the only one we know off on this forum at the moment,
but I beleave strongly there must be more decks out there that should be able to achieve a winrate of 55%+ on the long run with this bot.

If you would have checked the link, you would have seen the build in the right of his screen, just as easy to see it there as on a forum, im not going to make screenshots of decks that im not sure of that will do good.
When a deck achieves that highly desired rank 5 it has earned my respect, and I will def. make a screenshot of it then for your pleasing.

 

Mmm, the problem is probably that we're all quite lazy also. Ie: We use bots, of course we're lazy, that's the whole point of them.

In that laziness however, we do not look to create many decks. We use decks that are 'good decks', but that doesn't mean the bot plays them well. Patron Warrior is an amazing example of a good deck that can't be used by the bot. Our worst enemy is kind of ourselves in that regard. I am equally guilty.

Midrange pally just happens to be how the bot likes to play. Not just that, it has enough safe guards that it allows for the bot to succeed.

Ie:
3 healing cards.
2 Aldor Peace Keeper (basically 2 removals)
2 Equality (can be used really well with Muster for Battle AND Consecration)
2 Consecration (anti agro and also for equality/boardwipe)
1 Sylvanas (uh oh button)
2 Ironbeak Owl (2 uh oh buttons)
1 Big Game Hunter (biggest uh oh button)
1 Dr Boom (50/50 win condition uh oh button)

This is why Midrange Pally works so bloody well. It's 'safe'. Maybe if we shoved a bit more healing into Druid it'd be equally good tbh. Hrmmm. >_> Need to test that.

Edit: I just looked at the Control Mage deck. Interesting, and I think it definitely has the potential to do well. The problem is it has 3 traps. I don't trust the bot to put down Effigy when it really matters. Ie: At any time at all. Try changing those cards for something else and it might work. My initial thought might be 1 Flame Strike for some clear.

 

you are being very helpful and constructive, thanks for that

ooh yea im running the Dragon Priest atm, because im kinda lazy to and I dont want to watch all the matches but do want to know what it loses from, so i can then improve the deck.
Im using: http://hearthstonetracker.com/ its nice to use when you play HS yourself, but the stats can tell you if you lost a lot vs hunter or stuff like that.
If you do let it run over night, you know exactly how it performed vs other classes.

 

Sounds wonderful, throw us a link to which Dragon Priest you're using. There's a few rotating around. I'm looking forward to your findings.

 

I cannot see how you can possibly reach rank 5 with the bot in its current state.

Like, everything it does seems to be extremely dumb. It is certainly good enough if you only care for rank 20 gold farming, but I can't see it reaching a high rank. It cant even get essential mechanics right, like dealing with hunter traps or immune heroes (iceblock, malganis).

I have used the deck that you talked about in the other thread. The best it achieved was rank 7 for me. Usually it hovers between rank 8 and 9, unless it gets a very lucky or very unlucky streak.

 

I have the bot but shitty cards so im getting stuck around 16. However my 2 friends have nice cards and have achieved 10 using the bot.

 

Lets put the Decks aside for a moment and discuss the Modification of the DefaultRoutine that brings the best results.
I already experimented with all these settings, but I cant really see significant improvements in the Bots Gameplay even if I double the number of created boards.
So is here somebody, who once reached Rank 5 with the Bot and can tell us which Values he used for the following Variables ?

@Tony: U once said, ur Bot can reach Rank 5, so please tell us ur secret.....

Behavior behave = ;

int enfacehp = ;
int mxwde = ;
int twotsamount = ;
bool enemySecondTurnSim = ;

bool playaround = ;
int playaroundprob = ;
int playaroundprob2 = ;

int amountBoardsInEnemyTurnSim = ;
int amountBoardsInEnemyTurnSimSecondStepp = ;
int amountBoardsInEnemySecondTurnSim = ;

int nextturnsimDeep = ;
int nextturnsimMaxWidth = ;
int nexttunsimMaxBoards = ;
int ImprovedCalculations = ;

bool secrets = ;
int alpha = ;
HREngine.Bots.Settings.Instance.simulatePlacement = ;

 

@EmoKlaus

This morning I played pally deck to-> rank 6, but after update i return to rank 9 :/ ( the bot have a bad play now )..

 

I recently got rank 5 with a midrange paladin deck similar to this one https://www.thebuddyforum.com/heart...guides/224658-rank-4-paladin.html#post2039094
The settings I used were as follows:

int enfacehp = 15;
int mxwde = 5000;
int twotsamount = 1000;
bool enemySecondTurnSim = true;

bool playaround = false;

int playaroundprob = 50;
int playaroundprob2 = 80;

int amountBoardsInEnemyTurnSim = 40;
int amountBoardsInEnemyTurnSimSecondStepp = 200;
int amountBoardsInEnemySecondTurnSim = 40;

int nextturnsimDeep = 6;
int nextturnsimMaxWidth = 20;
int nexttunsimMaxBoards = 200;
int ImprovedCalculations = 1;

bool secrets = false;

int alpha = 50;

HREngine.Bots.Settings.Instance.simulatePlacement = true;

The reason I use these exact settings is that obtkamer the coder of Silverfish AI uses these settings as the default one for his ai, so I took it as a kind of recommendation.

Hope this helps.

 

Thx, I'll let the Bot run with these settings + that same Paladin Deck over night.
I start at Rank 14.

I'll tell u my actual Rank after waking up ^^
Just hope it wont be 15.
X_X

 

Nice, I hope you are successful.

I forgot to mention that I also added the lines to always mulligan Ironbeak Owl and Big Game Hunter. You can find them in the Rank 4 Paladin thread mentioned above.

 

Welp, sorry, but this is the second month in a row I've hit 5+ due to the new reward system. Prior to now I hardly cared about rank. And as written above, there are multiple people using midrange pally hitting 5. Consider adjusting your hours of play? Maybe you play during pro player peak or something.

 

I edited the DefaultRoutine according to Masturcheef's settings and added the Mulligans and used the recommanded Paladin Deck.

I started at Rank 13 and woke up at Rank 14

Maybe the last Patch realy did break the Bot ?

 

@Masturcheef

Thx I'll test it

 

Today I noticed something similar:

Decks with a decent winrate now perform very poorly and barely get any wins, so I suspect it has something to do with the new update (other threads also mention this problem).

 

13/26 atm for me 50%