In my first article on mindset, I learned from Zalae how to keep a level head and avoid tilt. We talked a lot about harmful habits of thought and how they can negatively affect one's play.
We didn't talk much, however, about developing positive habits of thought. That's the subject for today. I was privileged to have Purpledrank, Archon's team coach, share with me his insights on learning from losses.
Purple's advice was simple and powerful, sometimes surprisingly so. Focus on fixing the earliest mistake in every game, beginning with the mulligan, in pursuit of flawless play. Forget decklist tinkering to "beat the meta" and just play your strongest decks perfectly, whether on ladder or in tournaments. If one of your personal top tier decks is consistently losing despite correct play, stop using it. Take advantage of expert players to learn decks before playing them.
Learning from Losses
Let's say you lose a game, and you're not immediately sure why. How do you figure out what happened and what to adjust in the future?
When I was not so good at the game, I actually spent more time revisiting losses than playing games. Now, I am more or less able to recognize my misplays immediately.
Think over the turns that seemed obvious. Most of the time, you won't find much, but every so often you realize that you could have taken a different line.
Recording games seems indispensable for looking back over decisions, but most players don't do it.
Yeah, it is. I tend to rewatch a lot of tournament games, and I jot down times while I stream to go over specific games.
Win or lose, most games are decided by the mulligan. Most Hearthstone games are decided by turn 5. If things go completely wrong, you need to revisit your mulligan.
That seems similar to Day9's advice on improving in StarCraft 2: Focus on the very first moment that things went downhill, not the final blow that sealed the game.
Yep. When I do coaching, I often hear the old "Dr. Boom is overpowered" thingy. Most of the time, if you lose, it's because you didn't set up for the Boom on turn 7. The three turns prior to Dr. Boom coming down are what decided the game, not the Boom on 7.
For me, aiming to play games perfectly, at the sign of the first mistake, regardless of outcome, the game is a failure. Even though you don't always lose that game, you are playing a suboptimal game, in which you are trying to recover from mistakes. The objective is to play a flawless game, so it's good to catch yourself on the first mistake. Try to play out each turn in your head thinking about what would have happened if you had made the correct play to begin with.
I sometimes immediately concede if I realize I made a misplay. Do you ever do that?
It's worth playing out if you are really high up in Legend ranks, because recovery is always possible, but off stream, yes, I do that a lot.
Yeah, naturally it depends on if it's at a point in the ladder season where the game matters, like climbing the last few games to Legend or pushing top 100 at the end of the season.
So, you played a game, and you recognize that you misplayed. Win or lose, what goes through your head beyond kicking yourself not to do it again?
You tend not to focus on games you win. It's actually important to set your mind to have positive reinforcement in your day, so beating yourself up on both wins and losses is going to lead to not enjoying the game.
Often, when I'm playing with Firebat, we catch mistakes, and just drop an "Oops." We make a reminder not to do that again.
So while playing, you try to let the games be over when they are over, and move on to focusing on the next one. You review previous games at a separate time?
Some games you actually do need to look over separately for sure. Most of the time, however, you make a mistake on turn 5 and 6, and you go, "Oh, we messed up because we didn't set up to deal with Emperor Thaurissan effectively." That's a good example.
How To Interpret Trends: Don't.
We've been talking about individual games so far. What about overall win/loss records? How do you tell the difference between variance and trends you can learn something from?
In the long run, all statistics can tell you is the win rate for each different matchup. In the short run, it is impossible statistically to prove with a significant r-squared (correlation coefficient) if you are playing poorly or if it is variance.
My personal benchmark is that if I can't get top 100 Legend over the course of a week, either I am playing very poorly, or I am playing the wrong deck for the metagame.
That actually ties into another question I had, which is how and when you decide to switch decks for ladder. Finding the middle ground between stubbornness and overadapting I think is one of the hardest aspects of ladder play. It sounds like your time scale for that decision is about a week?
Every player should have their own tier ratings for decks. For example, for quite some time my tier 1 has been Midrange Hunter, Freeze Mage, Oil Rogue, and Druid. If I were trying to climb, I would only play those decks. And since I value them all at tier 1, it wouldn't matter if I switched decks after every game.
That rating is, I imagine, based to a great extent on your own skill with those decks.
Yes, very few people would consider Freeze Mage tier 1. Everyone has their own strengths and weaknesses in terms of playing different classes.
That's pretty different than the way most people talk about beating the meta by finding the strongest deck in the moment. It's more similar to players like Loyan or Ryzen, who climb with the same class every season, regardless of the meta.
Regardless of a particular deck being good in the meta, if you suck at it you won't climb.
We've mostly been talking about ladder, but I suspect a few readers also play in open tournaments. How do you apply this approach to tournament play?
Open tournaments are very similar to ladder, but there are a lot more tech cards included. I wouldn't get paranoid about them. I've played in a bunch, and the odds of someone running a Conquest lineup with triple Kezan Mystic decks is really low. It's never come up for me.
Just use the decks you have been practicing, play them perfectly, and your tournament results should be similar to your ladder results.
So don't stress about lineups, just play your three best ladder decks and crush noobs.
Yep, more or less.
Hopefully, that encourages more players to try an open tournament for the first time.
Learn First, Then Refine
What about learning new decks? It is both more difficult and more important to correctly analyze losses with an unfamiliar deck. Most players don't practice with a team, so they learn by laddering.
One of the benefits of living in the Archon house is that there is someone to help you learn new decks. For example, if I were to learn Patron Warrior, I would sit down with Zalae and he would talk me through a lot of the plays.
Learning from scratch is something I would not recommend. I know most Hearthstone players don't have the same luxury as I, however. Utilizing resources like Twitch streams to learn is a great starting point. A lot of it can just be watching a turn and saying out loud what your play would be, then comparing it to what the streamer did. If it's different, what was the streamer's thought process? Keep in mind that your play will often be better than the streamer's, because they will be making mistakes as well. After doing that exercise for a while, you should be able to start playing with the new deck as if you were already comfortable with it.
The main focus should always be on mulligans and deck building. Understanding why the cards are in your deck is the foundation for figuring out optimal plays later.
That's pretty daunting for a truly unfamiliar deck, learning not just how to play the list, but all the potential card choices and why to run one over another.
Yeah, it is. Most people will already have preconceived notions of a deck just by playing against it, though. Once you've gotten 30'd out of hand by a Patron Warrior, you can kind of figure out why Whirlwind, Frothing Berserker, and Warsong Commander are in the deck!
That lesson sticks pretty hard after the first couple times it happens.
Extreme example, I know, but losing to a deck is a good way to figure out how it operates.
The picture I'm getting is very focused on playing as perfectly as possible with a deck you are extremely comfortable with and knowledgeable about. You seem unconcerned with what other people are playing. Are there aspects of your play that you adapt?
Adapt to other people? Absolutely not. Tier 1 decks have game plans and you should focus on executing that game plan rather than worry about tech cards or meta game choices. If you play perfectly, the wins will come regardless of what your opponents are doing.
"Perfect play." For most players, this is an aspiration, not a concrete goal. By taking systematic steps to improve, however, perfection becomes more and more real. If you haven't already, go install OBS; press "Start Recording" at the beginning of your next ladder session. Reviewing recordings of your games is a potent tool for self improvement. You can get the most out of it by keeping in mind what to look for, and, equally important, what to ignore; what to adjust, and what not to adjust. Focus on the mulligan, identify the earliest misplay in each game, and don't let the meta or win/loss streaks fool you into playing any other deck than your personal best.
Make sure to keep an eye out for the next installment in this series on the competitive mindset in Hearthstone!