THE DREAM - Soddisfazioni e riflessioni
Attenzione: Se effettui il login puoi cambiare la lingua con cui visualizzare le carte di Hearthstone!
  • Peppu 26 maggio 2015, 18:14 Peppu
    BattleTag: Peppu#2154
    Messaggi: 2856

    ''I usually WAT alone'' [cit.]
    Perchè l'Ancient WatcherGuardiano Antico ha -0,24 come valore? rofl rofl rofl 1
  • FeriandRedazione 26 maggio 2015, 20:05 Feriand
    Messaggi: 9780

    Responsabile della sezione Hearthstone di BattleCraft
    Ho cercato il post degli autori di heartharena, questo interessa anche a @PokerFace ;)

    I cambiamenti portati nell'arena da GvG. Post Completo
    GvG Meta Changes
    A More Sticky Board. This is the concept that will govern much of your advanced GvG play. The number of minions in a typical draft that require 2+ hits to remove from the board, or that removes a minion and stays on the board, has more than doubled. This means that a new consideration, once an afterthought in the Arena, has now come to dominate high-level thinking in the game: # of attacks per turn. If you have 2 minions on the board and one removal in your hand, you have 3 possible attacks for the next turn. If your opponent plays enough minions/removal to absorb/initiate 4 attacks, then you will no longer have full board control the next turn. That's a Harvest Golem and a Piloted Shredder, a Bomb Lobber and an Annoy-o-Tron. Heck, that's two Razorfen Hunters. Even wonder why Muster for Battle and Impo-losion are such great cards despite more or less being properly valued mana-wise? It's because they guarantee board presence the next turn. Many considerations about how to play your cards in anticipation of your next turn changes with the necessity of more attacks to control the board (leads to intelligent "over"-extending, higher value for small removal and pings, etc). It also plays into how you play your own cards, so that you can play on your opponent's need to anticipate your board, and so you can actually get away with anti-tempo turns to set up super-sticky boards on future turns by saving certain cards to be played together on the same turn. Attacks needed was rarely a relevant concept in the Arena pre-GvG, but incredibly important now, and it piggy-backs on tempo to allow for quicker lethal and give more consistency to buffs (a staple in up-tempo decks even pre-GvG). When combined with taunts, which redirects the limited attacks you do have, it's a deadly combination.
    Faster Starts, Stronger Finishes. Everything snowballs!!!!! Curves are better now!!! Whoever doesn't curve out loses!!! The first two are true statements about GvG. The third one is a huge exaggeration. The game has not fundamentally changed. Missing a drop was always a huge disadvantage, and it is no moreso now than before... unless you do not have the proper answers with initiative. GvG allows for a quicker push for lethal with the same fast starts as in pre-GvG, but these pushes are just as vulnerable to initiative from the control game of an opponent. Now more than ever, that Frostbolt or Darkbomb is probably better than an actual 3-drop. With the drop in consistency of drafting removals, large minions played in the late game on even boards are more powerful than ever. Large minions played late game on advantageous boards are game-ending. So, while large minions are a bit more situational now than they used to be, their powers are actually magnified compared to pre-GvG from a combination of scarcity of big drops and scarcity of large removals. This is of course, doubly true for large taunts (which removes the situational element and can be effectively played even on disadvantageous boards). The biggest mistake you can make in the GvG meta is to push all for small minions. That was a winning strategy in week 1 of GvG. It is a highly inconsistent strategy now without the proper cards offered. GvG magnified the effectiveness of extreme strategies on both ends of the spectrum, while shifting the median slightly to the more aggro side. It did not by any means destroy the control game. As per pre-GvG, it is better to go with your draft, than to force things in a particular direction from pick #1. Take stock of your 10, 11, 12 win opponents. They are still less than 50% comprised of aggressive decks. And thus, we establish the RPS meta of GvG. With more diversity comes more completely unwinnable games. The 12-win rate of top players will drop, and most likely the average win rates as well (although, that has a higher chance to even out once the RPS effect is properly accounted for by raising the floor of bad decks as well).
    Yeti = Win. We always had Flamestrike, Truesilver, Eviscerate, and Swipe. Now, we also have Bomb Lobber, Flamecannon, Tinker's and dramatic increased appearances of Warlocks & Paladins. The two classes that were once able to handle 5 health minions with ease (Warrior and Druid) have both dropped in popularity. So, the 5-health minions have become the new win condition. About two thirds of your games as a mid-ranged deck will be determined by who has more 5-health minions + large removals, if both players curve out. Similarly, 4-attack minions are also on the rise due to 3-drops having 4 health now, and the rise of Priests' popularity. Together, these two effects mean that Yetis are even more dominant in the meta than they used to be, despite no longer being the best neutral common.

    Alcuni aggiornamenti. Post completo
    Our methodology is to look at outliers and find patterns with groups of cards that share common characteristics to determine rules of value and usage. The Tier List is a web of these values, and it is impossible to move any one piece without seriously impacting other pieces. We also only interpret the stats through our own experience, and certain cards (several of which will be noted below) are not getting their proper statistical due in our tier list, because we simply don't think they're that good. As usual, we're adhering to our philosophy that an expert tier list based off of mathematical models adjusted by meta experience and statistical analysis will be superior to a purely statistical or mathematical based tier list.
    Anyway, after 100+ combined manhours, we are finally done with the Tier List Update! Full change log is here, but no need to sort through all that unless you're seriously interested. Almost 50% of cards have had their values adjusted at least slightly, with about 5% adjusted more than a tier. We've also removed some class adjustments that just weren't getting results, and added ones that the stats indicate strongly, while readjusting others to scale more precisely. So, the red/green landscape of each class will look significantly updated as well.
    Observations are below.
    We were generally right! Most adjustments were slight and ones we were planning to make before looking at the statistics, less than half a tier. It's a rare case when two cards of the same mana cost had their relative positions switched. Actually, the raw tier list values of our neutral minion markers (e.g., 4/4, 3/3, 7/7) matched the statistics so precisely that it was creepy. Further, the mass majority of our post-GvG meta adjustments were accurate or been pushed even farther in the direction we were going. ::pats self on back::. That being said, with 500+ cards, there were a few big misses and we will cover the major ones below. Overall, following our old tier list would have gotten you on average 90%+ of the same cards. A/A-
    The Legendary Yeti is not THAT good. What can we say, we're suckers for tradition. The Chillwind Yeti has stood atop all other classic/naxx cards in just about every single neutral Tier List since the days of Trump and Ant1Gravity. We're not changing that... yet. The data is really not necessarily statistically significant to move them down relative to tier peers, but it's looking much more like a 0-4 point difference among the top class commons than our old 14-point differential, and with the Yeti closer to the middle than the top (Dark Iron and Sen'jin both have higher variance but also higher win rate). We cut the whole thing down to a 8 point differential, but statistically, it should be just 4 points. Those top neutral commons are good, but they're not that much better than the next tier of Shattered Sun and Stranglethorn Tiger. Similarly, the top legendaries like Ragnaros or Ysera are not really that much better than lesser heralded cards like Troggzor. They're better, but it's not as night and day as you might think. Sunwalker in Warlock, for example, is better than Ragnaros. Side Note: As expected, better players do better with Mechanical Yeti than Chillwind Yeti. It's a wash for the average player. Gotta learn how to use those spare parts, and when not to play that Yeti, guys!
    It's Still Very Very Good to Draw Cards. We were never huge fans of card draw, and with GvG's more tempo-based meta in the Arena, it seemed like most players had come around to our thinking (or even gone further than that into straight up aggro). Well, two classes aside (Hunter and Warlock), cards that draw other cards are still very very good. If we stuck straight to the statistics, Acolyte of Pain would be one of the top 3 common neutrals. Novice Engineer statistically is one of the top 3 2-mana minions. Shiv is great, Hammer of Wrath and Starfire are certainly not "too slow", and even friggin Call Pet is not an awful card (that's right, we're actually artificially deflating Call Pet's score, to stay sane). I don't think anyone's successfully calling their pets... it's more that the ability to cycle a card means there's one less "below average" card (your alternate choice in the draft) in your deck, therefore improving your deck quality. If the average card in a deck is 65 points, any card that cycles itself is worth 65 points if you don't care about the tempo loss (which, if your deck is going to play the late game, you really don't care all that much about). That's not to say Call Pet would be a 64 Tier Score card in Mage/Druid... but it certainly wouldn't be where we had these cards.
    The Great 3/2 vs 2/3 Debate. So, which is better? Here's the answer to this age-old question: Drum roll...... they're the same. The Arena is actually perfectly balanced in its neutral 2-drops if you're a high-win rate player. Go Blizzard! The addition of Shielded Minibot really changed up the effectiveness difference between the Raptor and the Croc. BUT, interestingly, if you're an average player, 2/3s are actually slightly better for almost all classes. Why? 3/2s are more easily removed, so there are more times when you'll have to come back onto the board, and there're more things to play around (cleave, forked, consecration, holy nova, swipe hero power, explosive shot, stormforged axe, glaivezooka). Better players are able to dodge these bullets, which then let them take advantage of the extra face damage and ability to eat up Mana Wyrm, Northshire, and Voidwalkers, while trading up with 4/3 and 3/3s.
    Average Players Try to be too Fancy. There's a reason we made it a point to divide stats not just by how many wins a run gets, but more importantly by how good the player is playing the deck. It's because average players are noticeably... average. With almost every card that is not a play-on-curve-drop, or a respond-to-situation-X type card, better players make better use of relative to their other cards. In fact, having positive text on the card for minor effects is actually WORSE than not having the positive text at all! A Goblin Sapper performs about 1 entire tier better for the average player, in secrets classes, than Illuminator. But, lo and behold, for classes without secrets, there is no meaningful difference between the performance of the two cards. That's right, by giving players the option to play for the "value" of a little bit of healing.... the average player will take the bait and lose a bit more. On the other hand, as you might expect, high win-rate players perform better with Illuminator in secrets classes than Goblin Sapper, because it is obviously a better card. Remember, it's the win rate of the player we're changing here, so the deck secrets composition is assumed to be a constant between the two types of players (especially when most people stick pretty close to what HearthArena suggests). We'll release statistics showing why that's mostly a good idea later =P Just a small anecdote that we saw repeated everywhere with almost every card that was not a straight up "respond to situation X" card (like Kodo). Interestingly, Taunts made some of the biggest skill gap differences. So, that may be a good place to start if you're looking to get a better understanding of the game.
    Dr. Boom is OP. Yup, even in arena. I'm not going to go too deep into it, but we're not giving it its proper due in this Tier List due in part to the slight RNG, and due in part to leaving some space for potential better cards. But seriously, Blizzard. Every result ever gathered in Hearthstone with Dr. Boom is pointing in the same direction. I was never a huge fan of the card (and only used it once), and so we didn't have it top-tier rated. But, the stats for this are as ridiculous as the stats for Ancestor's Call (I've said it on stream before, but if we could, we'd give this thing a -30 tier score, max, so that nothing heartharena does could push it over a hungry crab). Oh, and speaking of Hungry Crab. It's pure statistical value is about a 44 Tier Score. We are not joking. The sample size is surprisingly decent as well. Please post your personal testing results.
    The Mysterious Case of the Flame Leviathan. Speaking of things that punched wayyy above its weight. . . presenting: Flame Leviathan. 7/7 for 7 mana, with a totally RNG 2-damage board clear. This card should not be much better than a War Golem. We did not move it. But, it definitely wins the award for the biggest wtf stat. If rated purely by statistics. Flame Leviathan would be rated a 94 (150 run sample size, so on the small side, but not worthless).
    Pandas < Birds. We thought good players would use them better than their textless Raptor and Tallstrider counterparts. We were wrong. Even the best players average a significantly lower win rate with Brewmasters (although, they did close the sizable gap by more than 50%). Just too much tempo lost.
    Best/Worst Class? Mage, duh. Kidding. That's only true for infinite arena players. For the average player, it's actually a toss up between Mage and Paladin. Also, overall, better players do relatively better with Rogue, while less skilled players do better with Priest. Warlock and Warrior tie for lowest winning class, regardless of your skill level.
  • PokerFace 26 maggio 2015, 20:22 PokerFace
    BattleTag: qazasa#2586 Morgana#1981
    Messaggi: 485

    Leggere può seriamente danneggiare la tua ignoranza
    @Feriand mattone O_O ma interessante :good: 0
  • FeriandRedazione 26 maggio 2015, 20:55 Feriand
    Messaggi: 9780

    Responsabile della sezione Hearthstone di BattleCraft
    @PokerFace due bei wall of text, ma capendo come pensano gli autori del tool il mio uso di esso è migliorato notevolmente :good: 0
  • BloorMost Valuable Poster 26 maggio 2015, 21:06 Bloor
    Messaggi: 4035
    ci si riprova col 12 win!
  • FeriandRedazione 26 maggio 2015, 21:27 Feriand
    Messaggi: 9780

    Responsabile della sezione Hearthstone di BattleCraft
    @Bloor non posso seguire adesso, ma se non usi heartharena non ha molto senso postare qui :) 0
  • Magnus 16 giugno 2015, 09:43 Magnus
    Messaggi: 971
    Eccoci qua con un altro 12! (ci è mancato poco davvero che rifacessi il dream)
    Ero su Ipad e non so perché non mi ha mostrato la lista sulla destra alla fine, ma va beh
    Lista mazzo

    Analisi HearthArena

    Le scelte
  • 7ornado 16 giugno 2015, 10:06 7ornado
    BattleTag: antani#2993
    Messaggi: 526

    Iscritto il: 08 aprile 2015, 13:48
    mazzo poco forte eh...congrats 1
  • kenjiste 16 giugno 2015, 10:42 kenjiste
    BattleTag: kenjiste#2776
    Messaggi: 428

    wubba lubba dub dub!
    @Magnus grande :) :) 1
  • FeriandRedazione 16 giugno 2015, 10:48 Feriand
    Messaggi: 9780

    Responsabile della sezione Hearthstone di BattleCraft
    Io dopo le 10 vittorie dell'ultima volta non ho più ottenuto grandi risultati, ma sto giocando molto meno in questo periodo. 1
  • Peppu 16 giugno 2015, 15:10 Peppu
    BattleTag: Peppu#2154
    Messaggi: 2856

    ''I usually WAT alone'' [cit.]
    Ma guarda un po', un Fel ReaverVilrazziatore per le 12 wins! :D
    Non mi sembra un mazzo eccessivamente sgravo (non c'è un'arma che sia una per esempio), un motivo in più per farti grossi complimenti :D
  • Magnus 16 giugno 2015, 15:41 Magnus
    Messaggi: 971
    Peppu ha scritto:Ma guarda un po', un Fel ReaverVilrazziatore per le 12 wins! :D
    Non mi sembra un mazzo eccessivamente sgravo (non c'è un'arma che sia una per esempio), un motivo in più per farti grossi complimenti :D

    @Peppu Grazie! Il VilrazziatoreFel Reaver mi ha fatto vincere 3 partite e perderne 1
    Oggi invece con un mazzo che ti assicuro era da 12-0 tranquillo ci ho fatto un 5-3...sfiga nera...
  • Peppu 16 giugno 2015, 16:00 Peppu
    BattleTag: Peppu#2154
    Messaggi: 2856

    ''I usually WAT alone'' [cit.]
    Io sono un fan del Fel ReaverVilrazziatore, a differenza di tanti diffidenti qui nel forum :D 1
  • Magnus 20 giugno 2015, 12:40 Magnus
    Messaggi: 971
    Altro 12-1, l'unico rimpianto è quella sconfitta persa veramente male.. (mi sa che la lista che non appare è un bug aggiunto con la nuova patch (boh) )


    I premi

    Scelte fatte
  • Yardeh 20 giugno 2015, 13:05 Yardeh
    Messaggi: 9338

    Iscritto il: 21 aprile 2010, 19:16
    Poco oro, vedo :D . 1
  • kenjiste 20 giugno 2015, 13:48 kenjiste
    BattleTag: kenjiste#2776
    Messaggi: 428

    wubba lubba dub dub!
    Come sempre, ben fatto :good: :good: :good: 1
  • Magnus 02 luglio 2015, 16:11 Magnus
    Messaggi: 971
    Beh, che dire, ecco un'altra arena run degna del titolo del post ;)
    Le ricompense

    La lista

    Le scelte fatte

    L'analisi di Heartharena
  • kenjiste 02 luglio 2015, 20:51 kenjiste
    BattleTag: kenjiste#2776
    Messaggi: 428

    wubba lubba dub dub!
    @Magnus mamma come vai :) :good: 1