[G] Why you should play ladder on two accounts

Hey, what's up, everyone? This is Sean Plott, a.k.a. Day[9], and today I want to talk about why you should play against worse players. Now, I know 90% of you hear that and you just scream, "Blasphemy! Crucify Day[9]! He's dead wrong! You should only play against the pros!" People argue that, somehow, by playing against worse players, you will stagnate in this disgusting cesspool of newbies, and that you'll never learn about "gosu builds" and "gosu tricks" so that you can become "gosu x2". Obviously, though, I'm not saying that the path to improvement involves finding ten people who suck and only playing against them. And I'm not even arguing that you shouldn't play against players who are somewhat better than you, if not way better than you. However, I really want to emphasize that, not only is it OK to play against players worse than you, but, in fact, it is an integral component to your improvement. It is critical that you regularly play against players that are worse than you, to improve to higher and higher levels.

With that said, in this talk I want to do the following:

First, I want to spend some time talking about this "common mindset" I was referring to, this idea that you should only play against good players.
Second, I want to take some time to break that argument down and demonstrate why that doesn't quite hold up; that is, I want to illustrate the flaws in that line of reasoning.
Third, I want to spend some time backing the logic of why it's good to play against worst players,
and then last, I want to wrap everything up by giving a concrete example, by kind of walking you through one of my own personal experiences of improvement, and why playing against worse players was so important.


Let's go back to the beginning: let's start off by talking about what the current mindset of playing worse players is. One of the big things I hear people say a lot is that when you play against a worse player, you can simply have fundamentals that are strong enough and end up winning without having the correct strategy. So, for example, you log onto ICCup, you play against a D-level player, and he just has so few units that, really, it doesn't matter what you do. You just walk out, and you win every single time. Since you're winning so easily, your opponent isn't forcing you to change. He's putting no pressure on you to win; he's putting no pressure on you to adjust your play to something better.

Kind of along the same line of reasoning, people say that, therefore, you should play against the best players you can: that when you play against a truly good opponent, he'll be crushing you so badly that you'll start to realize that a lot of your tricks don't work, and you'll start to see which ones are remaining somewhat effective, and then you'll end up converging towards a better style of play.

For the most part, though, this whole view can be summed up with the idea of, "I want a challenge. I want to force myself to overcome some obstacle," because that's what people's idea of improving is: overcoming obstacles. With that in mind, I want to take the opportunity to move on to part II of this audio.

Let's break down this argument and see where some potential issues are. In Starcraft, build orders and styles play a key role in play, and a lot of times people will work on one style and then, for whatever reason, want to try a different style. If you're trying this new style for the first time ever, it makes no logical sense to try it against the best player possible. You want to go out and play it a few times and get the opportunity to run through your build from start to finish and build up some comfort with it, and then, that way, you'll start to say where your advantages and where your disadvantages are. A huge danger is that a lot of times players will dabble in certain styles, but they'll be playing against players who are too good, and these good players will crush these new styles, and then what happens is the player thinks to himself, "Ah. These new styles didn't work out so well. Look how badly I got crushed. As a result, that player ends up mislearning, because all his attempts at these new styles have resulted in failure.

Moreover, let's say you did stick with one style; let's say you really wanted to work something out. If you're playing against a better player, you just don't get the opportunity to become comfortable. The better player is spending so much time convincing you that a big attack is going to come here (and doesn't) or that a big drop is going to happen that doesn't, that a lot of times you start getting into your own head--there's these mind games that are going on that totally throw you off. On a less complicated note, sometimes, just the fact that his army is moving around the map, in good patterns, can make you feel so pressured that, you know, you forget overlords, you stop looking at the mini-map, and you might even get thrown off your build a little bit.

Most importantly, though, good players are very good at hiding their weaknesses. There's that Protoss player who will have two cannons at his front and the Zerg player will feel too intimidated to run in with zerglings, when he totally could have. I mean, how many games have you played when you go look at the replay and you say, "Oh man, if only I'd attacked here, I would have won!" Against better players, that thought oftentimes never happens when it should. You'll watch a replay, and at time x, if you attacked you would have won--but as you're watching the replay, that just doesn't register. You see the situation, and you think to yourself, "No, no, I definitely couldn't have attacked."

So clearly, in terms of learning, if you're only trying to play against the best of the best, or even players who are slightly better than you, you not only end up learning in the wrong direction by discarding potentially good strategies, but the strategy that you are choosing to work on, the build order you're doing every game: you might miss out on the opportunities that are present in that build, and you might not even be executing it as well as you could be. With that in mind, let me move on to the most important part of this audio.

Part III: Why it's important to play against worse players.

This entire section can be summed up with the following: Against better players, you learn what not to do. Against weaker players, you learn what to do. There's a great discussion about this in David Sirlin's book, Playing to Win. And what he says--which is true across a huge number of competitive games--is that good players are very good at hiding their weakness. Good players don't look weak when they actually are. Weak players, on the other hand, wear their weaknesses and exploits on their forehead--you know exactly how to crush that player. So what you want to do is you want to play against some worse players and see where those huge weaknesses are, or those glaring holes are, and start to abuse those. And then, as you start playing against better and better players, you still attempt those same abuses. And you adjust them to see if they work--or to try to force them to work.

It's true that playing against only good players will force you to readjust your strategies and your tactics and your tricks. The problem is that if you're only playing against good players, you're not expanding that enough; you're not expanding out to new builds and new tactics and new tricks; you're kind of sitting in this stagnant pool. And the problem with this is something I call working yourself into a logical corner.

Players stuck in "logical corners" are oftentimes the ones you see who just teeter between C+ and B-, with a record of like 150-180. And they just play tons and tons of games, and you look at their APM and they're fast players, so clearly they have the hand-speed, but for some reason, they just can't seem to broach[sic] those higher levels. You'll hear them say things like "No, I can't do that because I'll lose to this, and then I can't attack here, and then I have to make these-all defenses now, or else I'll lose to this"--and the problem is, though they've had experiences where these things were true, it doesn't make them rules that have to dictate the rest of their play for the rest of their life.

This kind of player should stop and say to himself, "What are some big problems that I'm having?" For example, let's say in Protoss vs. Zerg: "I don't feel like I'm attacking enough; I feel like every time I want to, I either get scared or I convince myself through some past game that I can't and that it won't work." So let's say that this player has been early-expanding in Protoss vs. Zerg and going corsair/reaver into mass ground--sort of the Bisu style of Protoss vs. Zerg. Let's suppose then he wants to be more aggressive: he should drop down to D-level and he should say, "OK. I'm going to start really harassing at this stage of the game."

Let's say he starts harassing at stage x or time x, just to throw, you know, a variable out there. Let's suppose that against D-level players he gets absolutely smashed when he does this: his harassment doesn't work, it's totally defended, and he thinks to himself, "Wow, that clearly is not a good idea, if it's already losing to D-players." However, let's say he readjusts it, and starts doing it a little later than time x, and he says, "Wow, I have so many more corsairs! But now I'm starting to do a huge amount of damage. In fact, now I'm winning games almost way too easily." What's great now, is that this player can then work his way up until he gets that time just right, and then, when he's back at that C+/B- level, he now has this time-frame where he's abusing this corsair/reaver aggressive harass. It might be a time that he never even thought of or saw in his previous games, because he was so scared of so many Zerg units.

The fact that he's abusing this reaver/corsair timing might help him bump from B- to B. But most importantly, he will gain a huge amount of confidence in Protoss vs. Zerg now, because all of a sudden, Zerg doesn't feel invincible. And that confidence will show in his play; he will play much more comfortably with his style, and that comfort might be what helps bring him from B to B+.

In short, it's always very useful to alternate between playing against weaker players and playing against stronger players. Playing against the weaker players will allow you to flesh out your ideas and see new opportunities that you couldn't before. And then you'll want to take those ideas and you'll want to test them against better players, once you're really solid. And once you've tested them, that's how you adjust them and change them and choose what works and doesn't. Alternating is what keeps the mind fresh and makes sure that you're always learning new things and not working yourself into some dreaded "logical corner".

I'd like to finish up this audio recording by sharing a personal story of mine: an experience where I played against worse players and I ended up improving far more than I would have had I just stuck with playing very, very skilled players. A few years ago, I was preparing for the WCG tournament, and this is when they had removed Estrella as a map from the map pool and they put in Peaks of Baekdu. So now the maps were Gaia, Azalea, Paranoid Android, and Peaks of Baekdu, so since WCG was coming up, I was naturally only playing on those four maps. And PGTour (or ICCup--I still tend to call it PGTour) was really, really useful, because that was one of the maps of the week, so I could always get people to play me, and it was against random opponents, so I was getting a huge variety of styles, so again: just totally normal, preparing-for-tournament style stuff.

However, I encountered a huge problem, because the first few weeks, the maps of the weeks were Azalea, Gaia, and Paranoid, the old three WCG maps. And on Week Four, I'd worked my way up to A-level, but the new Map of the Week was Peaks of Baekdu--a map that I had never played any games on. So when I began, I started by taking builds I liked on other maps and trying to translate them onto Peaks of Baekdu, and even though it felt kind of OK, I still got smashed by all these other A-level players. And in fact, I started to lose almost every single game, and I started to try different builds, and whackier strategies, and literally nothing was working.

And I remember I had just dropped down to A-, and I encountered a player, and we ended up playing about ten games, and of the ten games, I lost seven of them. And I remember that it just felt so hard, and even the wins I did get didn't really feel satisfying: I kind of had this {shudder} feeling about Peaks of Baekdu. And there came a point where I was getting close to dropping down to B+, that I said to myself, "I really just shouldn't even be playing against these A-level players. I'm clearly not good enough at Peaks to be here. And I'm not even learning anything. I'm just losing and getting pissed off."

Since school had just ended that week, I created a new account at D-level, and started to slowly work my way up again. So I was playing tons of Peaks of Baekdu--I was actually playing probably like 12 hours a day--and I started to really see what was good about the map and what sort of abuses there were, and I watched a whole lot of VODs, and formulated builds, and got to practice them against these weak players. And some of the builds that I started off with felt really uncomfortable even against the D-level players, and again: I immediately said to myself, "I'm totally not going to do this." And then, I converged to the style I really liked, that felt really strong, and I worked my way all the way back up to A- by the end of the week.

And when I was A-, something amazing happened: that player who I previously went 3-7 against, I encountered him again, and we were going to play another set of games together. And it had only been a week, but I had had a huge amount of time to sort of reformulate my play against these weaker players. And when I played against him, he was doing the same sorts of styles that he was doing before, but I ended up crushing him, like eight games in a row. And in fact, in those games? It felt so effortless. All his units felt slower and stupid, and his entire build felt glaringly flawed, and there were all these exploits that I abused all game long, and when his big attack came out, I had like five times as many units as he did.

In fact, in every single one of those games, I never felt like I was going to lose at any point in time. And that, right there, is probably the best feeling in Starcraft. Because it's really difficult to get an absolute measure of how good you are or how much you are improving, because even statistics like your PGTour or ICCup rank--that kind of varies depending on who you're playing against and whatnot. But if you get the opportunity to play the same player twice in a short period of time, and the second time you play him you feel overwhelmingly better, you suddenly realize that you have improved. You have taken your play to the next level and have left him behind.

The way I was able to do that was by playing against bad players. I reset my account and it allowed me to really see at a very, very basic level what was going right and what was going wrong. And then, once I had all this sorted out in my mind, I had the period of time from the C level up to the A- level to become comfortable with this build, to work out all the nuances and all the subtleties. Moreover, the style of play I developed was something I would never have seen just playing at the A-/A levels, constantly losing and saying to myself, "Oh, I've got to suck it up and, you know, just try to deal with these losses and learn." Only by dropping down to the D-level play was I able to actually see things much more clearly. As a result, Peaks of Baekdu became, eventually, my best of all the WCG maps, and actually still remains one of my favorite maps of all time.

In short: I want all of you to remember that by playing against bad players, you end up learning what to do: you end up seeing the true advantages and disadvantages of your play. And then you should take what you learn and test it against better players. It's always good to be bouncing back [and forth] between worse and better players, and in fact, if you can get that range of play to be right around your skill level, like players slightly worse than you and players slightly better than you, then you'll find yourself skyrocketing up, in terms of improvement, and skyrocketing up to higher ranks on PGTour.

So, I hope this audio rant was useful. Good luck, everyone. Cheers.