Hearthpwn

The ranked system does not (primarily) use an MMR*.  You are matched with other people as close to the same rank as possible, probably without regard to any estimate of your performance.

What this means is that after the reset each month, everyone is clawing back up the ladder.  Legend players start at 16/3 stars and people who haven't played start at 25, but if someone starts at 25 and gets to 15 relatively quickly, they'll be playing the slower players from legend the previous month, and their decks.

As the month progresses, the more active players and stronger players move up fast, while others move up more slowly, and lower ranks become easier as a result.  Probably the easiest time to move through lower ranks is somewhere near the end of the month, offset perhaps by more serious but inactive players coming back in the last week to rush through getting the card back.

If you'd like to play with an MMR system that tries to match you against other similarly-performing players, Casual mode offers that.  While my normal account is in NA, I play an f2p account on EU as well, and complete quests in casual mode using basic decks while laddering with one decent quality deck in ranked.  Because casual uses an MMR, and I'm using weak decks there, I can get my quests done relatively quickly because I'm matched against other players in a similar position.  Playing ranked, I'm up against all the usual meta deck suspects.

Hope that helps with understanding what's going on.

*  It's pretty clear that Hearthstone does at least retain an Elo-like estimation of skill in ranked mode, even if it's not the main mechanism for matching.  This is how the game chooses new Legend players' rankings, and how matching is done for legend players.  I'm not sure we know if it's used as a secondary input for matchmaking at lower ranks.

Blizzard consistently manages to make games that are engaging and highly polished.  Yes there are other games in the same spaces that are more ambitious, or more rewarding for some particular subset of players, or just more successful.  But, I work for another game company, and every single person plays Blizzard games.  Maybe not exclusively, but they're all sold on the experience.

The whole "quality going down" narrative comes almost entirely from people who have warm childhood memories of the first Blizzard games they ever saw.  

Getting from rank 5 to legend takes about as many games as getting from rank 20 to rank 5, assuming you maintain the same win rate.  If you can play that many games in a day, go for it, but I myself would probably save my energy for the new season and try to get to rank 5 earlier in the month. :)

After more play:  This is horrible, it just doesn't work.  But, one game out of ten is hilarious.

I've been playing around a bit with a warlock quest deck that doesn't really try to be a zoo lock at the same time.  Instead, it attempts to use a ton of card draw to accelerate the completion of the quest.  In playing this deck, one interesting side effect is that the cards that discard others are relatively powerful and come out pretty early, which can (when things work out) lead to a pretty strong board state around turns 4-6.

It's super-gimmicky and may not have a chance to develop into something viable, but I figured I'd share it with others to see if it stimulated some conversation.

Edit:  A new article on hsreplay.net describes their stats about how the quests affect win rate.  According to them, getting the lock quest completed by turn 7 is the tipping point for a favorable win rate.  I was finding that this deck could usually complete the quest around turn 7 or 8, but a little tweaking (or better play) could improve that perhaps?

Edit 2:  Here's what I wrote up for the deck description:

After playing some of the decks floating around out there for quest warlock, a couple problems jumped out at me:

• There are a limited number of cards that discard others.
• Getting the quest requires fueling those discards with more cards (because you only get credit for cards actually discarded.)

So, I went all-out adding card draw.  This version of the deck maximizes card draw to enable discards.  Lots of card draw also gets more and powerful discard-based cards out faster, which also enables having a reasonably strong board state on early turns because those cards have better stats to compensate for the discard disadvantage.

My success with this seemed to depend on successfully drawing lots of cards and getting Lakkari Felhound out fast to buy time, ideally with a Nat Pagle behind it.

My approach to the mulligan was to hunt for 1- and 2-drop card draw cards like Novice Engineer and Loot Hoarder.

This deck is gimmicky and probably will not ever be viable on ladder.  However, it feels tremendously powerful when it works and might be the foundation of something worthwhile.

Community-sourced data out there such as VS indicates that Elemental Shaman is winning more than 50% of the time from rank 10 to legend.  You can certainly rank up with a good deck.  You'll probably need to craft Kalimos, not sure how much more dust you'll need than that.

Quote from alienoreo >>

Casual seems far more instructive and frankly it's just a lot more fun. I have difficulty understanding why people spend hours and hours grinding ranks against decks they absolutely hate, absolutely not having much fun, when whey could have much more interesting games in casual. Then again, I'm not a trophy collector.

Quote from aeson88 >>

I don't think it's that healthy, there's still way too much aggro. 

Quote from Lisca >>

Pirate Warrior is superior in many ways... it has high consistency

My experience trying to rank up with pirate warrior above rank 5 (and this is borne out by the VS data) is that it's really great against certain deck archetypes and fails horribly against others.  This means that there's more variance in win rate, and extreme streakiness.  You might shoot up to rank 3 only to fall back to rank 5 again.  (This can of course happen with any deck, but pirate warrior is less consistent than some others.)

It does have the critical advantage of being a fast deck, but a more consistent deck will help ensure you don't end up way out on the long tail of many hundreds of games to get from R5 to Legend.

I ended up switching to a murloc paladin deck, which can still be fast, has a solid win rate and is a lot more consistent.  The worst match-ups aren't as bad (and the best match-ups aren't as good) but how things go isn't as dictated by a rock-scissors-paper of deck archetypes alone.

Quote from Grumhul >>

Quote from Breweryhorse >>

A ELO ranking system like we have it for Chess would be much better for hearthstone too. 

Blizzard had a ELO like system for starcraft/broodwar. They should make it for hearthstone too:)

 Nope. You need to think about the problems generated by such a system, when adopted by a game played by 40million players. Queue times will get horrible.