Some bad loses and worse wins

Back in May I posted a status of my first 20 games on iC Cup. I was 5-15 and felt bad about it. I never really got around to another 20 consecutive games. That is until now.

Recently I've played another 20 and the result is the somewhat better 8-12. Not a huge improvement, but still significantly better than before. I've switched from Terran to Protoss In between the different series and I hope that is part of the reason.

I promised to post some of my recent stupid loses. Here are four games to compare and contrast.

• Here I get owned by a Zerg.
• In this one I mess up a game versus a Protoss.
• This game I win pretty surprisingly. Actually the Zerg opponent disconnected right when I attacked.
• My Protoss opponent here claimed to be pretty bad. I think he was right.

So, dear readers, is there something for me to learn in the above games? Other than to pay attention and not fuck up the games. I clearly remember that in the two games I lost, I felt crowded and under pressure most of the time. Looking at the replay it does not seem so bad.

So what have I learned? Don't forget the second pylon. I do that too often. And don't forget to tech even when fast expanding. You need templars or reavers as early as your opponent.

So what is next on iC Cup for me? I don't know. I'd like to play a lot more, but I think they are going to reset the ladder pretty soon and I think I'll let it settle some before venturing out again.

APM Spam

I've suffered some stupid loses to Zerg on iC Cup recently and I decided to work on my mechanics a bit. First by playing the same build again and again against the computer. There is not really much opposition, but it gives me something to beat while I concentrate on my multitasking skills.

I've just finished my last such game and the game felt good. With some anticipation I went to bwchart to check out the result. An APM of 120 and most of the game slightly above the 100 APM mark. It is an improvement.

When playing a Zerg, I don't feel completely safe before I have my high templars and a couple of canons at each base. It is a balancing act to do stuff really fast while making sure to tech as well. Pumping low tech units is an easy way to get a high APM, but you need that aggressive tech build as well. I haven't tested this against a live opponent yet (Oh, sucky Zerg opponent; where are thou?), but it feels like it is almost there.

I still suck - a rant about ranking

I've been quiet for a while playing 1v1s again on iC Cup. And I still suck. For a while I thought that everyone except me was playing as Terran and I was thinking that I was slowly improving.

It may sound as a small accomplishment, but my APM once a rock solid 60 has climbed to 85 and I am eyeing 100, wondering if I will ever get there. I've also gotten some build-order wins. Not many, but a few. Then I ran into a bunch of Protoss players and realized that I still suck at mirror matchups.

The skill level difference on D is huge and at least two players who beat me was playing with APMs above 240. They were waaay better than me. It makes playing depressing. I think what is wrong is that many people start out new accounts just to beat up sucky players like me or maybe they are just cruising through the lower levels until they reach their level. I think there must be a better system.

Back when I was playing chess competitively I know that clubs, to some degree, could hand out start ratings to new players. To sort of seed them right. Some were given 1100 and others 1400. It made it more fun for those at 1100, that they did not have to play other newbies who were clearly a lot better.

Recently I got in to the Quake 3 Live beta. They do a calibration when you create your account and then they use that to match you against other humans. I guess they measure stuff like how long does it take you to get 20 kills, what is your aim accuracy and other things. I don't really know how well it works for Quake Live, but I like the idea. Obviously it can be abused, but I would like to see something similar for Starcraft II.

Hotkeys

A major part of my recently improved APM has to do with how I use my hotkeys and yet I feel I still have a while to go. In a game recently my ally pointed out that around 20% of my actions were selections and less than 10% was hotkeys. His stats were almost the opposite. I think i prefer to click on the buildings when the building is on the screen rather than cycle through hotkeys. So I use my hotkeys to build units while I am in combat, but when I am in the base I click on buildings and I probably shouldn't.

One other reason why I don't use my hotkeys optimally is probably the system I have for assigning builds and units to keys. Right now I put my start nexus on 1, the first three gateways on 2, 3 and 4, my initial scout on 5 and later groups of units on 5 and up.

Recently I've realized that some of the more specialized buildings needs constant attention too and is worth a hotkey. Thus 0 is now my robotics facility and that works okay even if I tend to forget I even have one! I guess that is a side effect of sucking so much at this game...

The above stuff tends to work okay for the first 5 to 8 minutes of the game, but at one point I may be at 5 or more gateways and the system collapses. Especially when I also have 2 or 3 control group of units. At this point I probably should reassign key 1 to a new gateway.

The problem is that I rarely get 5 groups on hotkeys so 5-9 is often underused. Could I use just 8 and 9 for unit groups then I would have room for 7 gateways on hotkeys. The problem here being that it is slightly faster to access hotkeys on the left hand side of the keyboard, so it makes it slightly more difficult in the beginning of the game.

What about your forges? Shouldn't they be hotkeyed too? I think I'm loosing my mind over how to optimize this. Oh how I suck at this.

You probably can't take him

So I was playing another 2v2 with my partner in crime AiurAdun. Us as white and yellow Protoss against a brown Protoss and a red Terran. I think this games demonstrates my understanding of 2v2s very well. Ie. you have to rush one opponent in a 2v1 situation and if you do that, you'll probably win in the end.

In this case I zealot rushed the brown Protoss. My ally thought that was a bad idea since he went 3 gate versus my 2 gate build. And AiurAduns respect for my 1337 micro-skills is not that great either. His advice was "du kan nok ikke tage ham" or "You probably can't take him on/out".

It turned out that I could and I am pretty proud of my micro in this part of the game where I think I am quite successful in making my zealots survive. It does help that brown was abysmally bad at controlling his units, but a win is still a win.

After the initial success we then managed to spend more than 35 minutes defeating red. A Terran that has dug himself in and is defending 3 bases is pretty tough although we eventually did take him out too.

When I watch the replay now I can evaluate the cost to my macro-game of doing intense micro in the start. Looking at unit counts I am around 32 halfway through the attack, whereas red is around 45. As I don't loose a lot of units and I think that could easily do a build and massing like red did, it probably cost me around 8 food count in lost attention to my base. That must be improved.