So far the new focus on strategy has yielded 2 wins out of 2 games, and both against what I would call "good" bronze league terran players - ie with 50+ wins in the league and a decent win rate.
Both times I've made mistakes. The first game I had some micro mistakes that cost me quite a few units - separating my infantry from my sieged tanks at one stage cost me four tanks, for example, but generally my macro was pretty strong - by keeping him on one base I had such a huge macro lead that I could afford to lose the units. In the second game I had some much better micro - including an initial push to snipe a supply depot without loss and a drop which took another depot without loss. Both of those attacks supply-blocked him. I also dealt with a hellion drop in my own base quite easily. I eventually just rolled over his base with my tank superiority. I took far too long grabbing my third and fourth gas though, I will have to remember that next time - I ended up with loads of minerals and no gas, because the only units I could make for long periods were marines and turrets.
Still - it felt good to have two really solid wins. I'm sure I'd have managed to lose at least one of those games previously, or at least they would not have been so one-sided, and it makes me think that with the right strategy I might improve further.
Wednesday 23 February 2011
Tuesday 22 February 2011
Strategy
So, I've been thinking a lot about my build order, and watching a lot of Terran replays, and what I've seen is that there are lots of different ways to play terran. Cloud, for example, relies on mainly infantry and vikings. He doesn't seem to build tanks - marines and marauders form the backbone of his armies - and he's aggressive about expanding early. Sjow, on the other hand, seems to love hellion play, particularly with the weapons upgrades, and drops. I think it's important to pick a style that you are comfortable with, and to have in mind an overall plan. With that in mind, I have the following general plan and I'll try to stick to it in future games:
Overall objectives
To keep opponent on one base for as long as possible. On maps where that is difficult (Delta Quadrant or Jungle Basin) to attack with a strong tank/viking push and try to break the front.
Rough build order
- Build 2 barracks, 1 with a tech lab and 1 with a reactor
- Build marines and marauders
- Research stim
- Build a factory with a techlab and a starport with a reactor
- Move out at around the 5:30 mark with whatever I have and push gently at the front of his ramp
- While pushing out save up for an expansion
- Build another 2 barracks, another factory with techlab and an armory
- Try a drop at around the 8 minute mark
- Mass troops for a big push at about 14 minutes
Execution
- It's important that before the first push engages I have some idea of whether I'm being rushed, so I'll scan the top of his ramp before I get there. I'll try to snipe off a couple of units with stim, or take a supply depot. The important thing is that he thinks I'm going to be aggressive.
- Leave a couple of units near his natural expansion if possible, so that if he expands blindly I can kill off his worker. Again, this should worry and discourage him from moving.
- It's important that for the early push and the drop I try to save my units wherever possible. The purpose of these attacks is not to win the game or even kill workers, it's to worry him and encourage him to turtle.
- For the big push I'll siege my tanks up near the front of his base and use my vikings to try to get vision up the cliff to allow me to shell any buildings near the edge. I'll build missile turrets at my contain and near my base.
- On large maps I might take a third once the contain is up.
Variations
If my opponent tries a big attack early such as a roach rush or a 4-gate then I'll need to shelve the expansion and play defensively. I'll need to scout that if possible.
On Delta Quadrant against Terran I might go for something completely different like a 4 barracks marine-only attack. I've tried that before and it usually works if I have about 8 or 10 marines at 5:30. Otherwise it can be brutal to try to break the frontal base defences.
Conclusion
I'm not trying to say this is some kind of uber strategy or that this is how you should play Terran. This is how I'm going to try playing for a bit though - hopefully being more rigorous and trying for consistency will help me improve.
Overall objectives
To keep opponent on one base for as long as possible. On maps where that is difficult (Delta Quadrant or Jungle Basin) to attack with a strong tank/viking push and try to break the front.
Rough build order
- Build 2 barracks, 1 with a tech lab and 1 with a reactor
- Build marines and marauders
- Research stim
- Build a factory with a techlab and a starport with a reactor
- Move out at around the 5:30 mark with whatever I have and push gently at the front of his ramp
- While pushing out save up for an expansion
- Build another 2 barracks, another factory with techlab and an armory
- Try a drop at around the 8 minute mark
- Mass troops for a big push at about 14 minutes
Execution
- It's important that before the first push engages I have some idea of whether I'm being rushed, so I'll scan the top of his ramp before I get there. I'll try to snipe off a couple of units with stim, or take a supply depot. The important thing is that he thinks I'm going to be aggressive.
- Leave a couple of units near his natural expansion if possible, so that if he expands blindly I can kill off his worker. Again, this should worry and discourage him from moving.
- It's important that for the early push and the drop I try to save my units wherever possible. The purpose of these attacks is not to win the game or even kill workers, it's to worry him and encourage him to turtle.
- For the big push I'll siege my tanks up near the front of his base and use my vikings to try to get vision up the cliff to allow me to shell any buildings near the edge. I'll build missile turrets at my contain and near my base.
- On large maps I might take a third once the contain is up.
Variations
If my opponent tries a big attack early such as a roach rush or a 4-gate then I'll need to shelve the expansion and play defensively. I'll need to scout that if possible.
On Delta Quadrant against Terran I might go for something completely different like a 4 barracks marine-only attack. I've tried that before and it usually works if I have about 8 or 10 marines at 5:30. Otherwise it can be brutal to try to break the frontal base defences.
Conclusion
I'm not trying to say this is some kind of uber strategy or that this is how you should play Terran. This is how I'm going to try playing for a bit though - hopefully being more rigorous and trying for consistency will help me improve.
Thursday 17 February 2011
Noob terrans
One of the things I find most frustrating about Bronze League is the massive variation in player skill. Take for example the two zerg players in the post below. The first player scouted me, and then successfully executed a cheesy but still valid early rush. The second guy fast expanded, held off my counter push (although with some losses) and then beat me with a proper baneling bust strategy. Both of those guys knew what they were doing.
Since then I've played two terrans. The first allowed me to pin him in his natural expansion. I had total map control, dealing with his one attempt to take a third. He made no effort to break out and attack me - even when I (with true noobish comedy effect) nuked my entire frikking army, he sat there and let me replace the contain before eventually starving him out.
The second terran I played had only played about 8 games ever, none of which he had won, and again he sat in his base and allowed me to lock him in. There was a half-hearted attempt to take a second base, which was easily crushed by the small force I'd left just next to his natural, but other than that he just sat there and waited for me to crack his (not very well defended) base.
I think that my league position in Bronze is probably artificially inflated by wins against these guys. I seem to always beat the real newbies, but it's the harder games where I really need to work and focus.
The problem is that when you start the game you don't know how good your opponent is! The last terran game in particular when I saw zero aggression my mind was racing - I was expecting cloaked banshees, I was checking the islands for hidden expansions, I was dropping and checking for thors (I actually saw an armory, and it turned out later from the replay that he did in fact research the 250mm strike cannons upgrade, but he never built any thors...). In the end though I could have won the game in 10 minutes with a siege tank push, but my worries about macroing properly meant that it took closer to 20 minutes...
I sometimes try to infer from my opponent's portrait how good they are - if they have the medivac portrait, for example, they must be reasonable. That doesn't always hold true though - I personally sometimes use one of the basic portraits on the basis that people might think I'm even worse than I am.
Since then I've played two terrans. The first allowed me to pin him in his natural expansion. I had total map control, dealing with his one attempt to take a third. He made no effort to break out and attack me - even when I (with true noobish comedy effect) nuked my entire frikking army, he sat there and let me replace the contain before eventually starving him out.
The second terran I played had only played about 8 games ever, none of which he had won, and again he sat in his base and allowed me to lock him in. There was a half-hearted attempt to take a second base, which was easily crushed by the small force I'd left just next to his natural, but other than that he just sat there and waited for me to crack his (not very well defended) base.
I think that my league position in Bronze is probably artificially inflated by wins against these guys. I seem to always beat the real newbies, but it's the harder games where I really need to work and focus.
The problem is that when you start the game you don't know how good your opponent is! The last terran game in particular when I saw zero aggression my mind was racing - I was expecting cloaked banshees, I was checking the islands for hidden expansions, I was dropping and checking for thors (I actually saw an armory, and it turned out later from the replay that he did in fact research the 250mm strike cannons upgrade, but he never built any thors...). In the end though I could have won the game in 10 minutes with a siege tank push, but my worries about macroing properly meant that it took closer to 20 minutes...
I sometimes try to infer from my opponent's portrait how good they are - if they have the medivac portrait, for example, they must be reasonable. That doesn't always hold true though - I personally sometimes use one of the basic portraits on the basis that people might think I'm even worse than I am.
Wednesday 16 February 2011
Heart of the swarm
So, after the bright new dawn of me deciding to blog about my SC2 experience, I logged on last night and played three times. I play Terran, by the way. Two of those games were against zerg, and there were two defeats. In my other game I played and beat a pretty awful turtling terran (I should say that I beat him even though at one stage I managed to nuke my own army by accident - I was trying out ghosts for the first time!).
Game 1 - vs Random on Shakuras Plateau
This guy rolled zerg and 6 or 7 pooled me. I always, always wall off when I face zerg, and do so slightly earlier than normal - I build my supply depot on 9 supply instead of 10 supply, and then send that worker to scout. The reason for this is that cheese is so common in lower leagues, especially by zerg players. I get those 6 zerglings turning up early about 50% of the time I think.
The problem here was that I built my second supply depot in the wrong place so there wasn't a complete wall off, and took a while to cancel it and rebuild. At this time my scout reached his base and I saw a couple of zerglings - I thought "shit, 7 pool!" and frantically tried to rebuild my depot but the zerglings arrived about half a second before I replaced the supply depot in the right place. In my panic I hadn't started building my marine from the barracks and the zerglings were in my mineral line before I even started to do that. Sigh.
What to learn then? Well make sure the wall off is in place! I might also not scout with my worker after the first depot is built but wait until after the barracks is built. It's a slightly later scout, but should mean I have more minerals and can therefore complete the wall faster.
Game 2 - vs Zerg on Blistering Sands
Another defeat. This time against a non-cheesing fast expanding zerg, who baneling busted me off two bases and killed me with a pretty large swarm of roaches and mutas.
I actually started quite well here, with an infantry push on about 6 minutes that killed the queen at his natural, the lings protecting it, and a bunch of drones. I then left him alone for ages (huge mistake right there) and when I pushed again at the destructible rocks at his back door I saw a few roaches lings and banelings shuffle past on the way to my base. I stopped hitting his back door to intercept the push, and I killed all the roaches and banelings although the zerglings cleaned up my infantry. Again though, I left him alone, and when I eventually expanded (so, so late) he just charged forward in a mass and went through the front door like a hammer.
So what did I do wrong? I was so, so late in expanding. I didn't harrass the FE enough, I had far too much gas and not enough minerals (hint - build SCVs!), I didn't tech to siege tanks in time to repel his push. I was too slow too slow too slow. Bah. This one hurt to lose, because I *knew* I was messing up. I perhaps got complacent after the first push, even though I know that if your opponent expands and you hit them hard at their expansion, any damage you do is only making up your economic disadvantage, not putting you ahead.
Nevertheless, I had been determined to push at about 5.30 with whatever infantry I had, and I did that and it worked well. I did the same in my next game against the terran player and took two supply depots and put him firmly on the defensive. The difference was that in the second game I followed it up with more aggression and eventually won. I think I'll do that all the time in future.
I think I need to decide on a build order and try to stick to it. At present I'm trying to do lots of different things ("maybe I should make some hellions..."), or to decide on the fly how many production buildings I need or when to expand. This not only makes all my builds a bit rubbish, but it's hard to learn what I could do better each time. My next post will be my build order from now on.
Game 1 - vs Random on Shakuras Plateau
This guy rolled zerg and 6 or 7 pooled me. I always, always wall off when I face zerg, and do so slightly earlier than normal - I build my supply depot on 9 supply instead of 10 supply, and then send that worker to scout. The reason for this is that cheese is so common in lower leagues, especially by zerg players. I get those 6 zerglings turning up early about 50% of the time I think.
The problem here was that I built my second supply depot in the wrong place so there wasn't a complete wall off, and took a while to cancel it and rebuild. At this time my scout reached his base and I saw a couple of zerglings - I thought "shit, 7 pool!" and frantically tried to rebuild my depot but the zerglings arrived about half a second before I replaced the supply depot in the right place. In my panic I hadn't started building my marine from the barracks and the zerglings were in my mineral line before I even started to do that. Sigh.
What to learn then? Well make sure the wall off is in place! I might also not scout with my worker after the first depot is built but wait until after the barracks is built. It's a slightly later scout, but should mean I have more minerals and can therefore complete the wall faster.
Game 2 - vs Zerg on Blistering Sands
Another defeat. This time against a non-cheesing fast expanding zerg, who baneling busted me off two bases and killed me with a pretty large swarm of roaches and mutas.
I actually started quite well here, with an infantry push on about 6 minutes that killed the queen at his natural, the lings protecting it, and a bunch of drones. I then left him alone for ages (huge mistake right there) and when I pushed again at the destructible rocks at his back door I saw a few roaches lings and banelings shuffle past on the way to my base. I stopped hitting his back door to intercept the push, and I killed all the roaches and banelings although the zerglings cleaned up my infantry. Again though, I left him alone, and when I eventually expanded (so, so late) he just charged forward in a mass and went through the front door like a hammer.
So what did I do wrong? I was so, so late in expanding. I didn't harrass the FE enough, I had far too much gas and not enough minerals (hint - build SCVs!), I didn't tech to siege tanks in time to repel his push. I was too slow too slow too slow. Bah. This one hurt to lose, because I *knew* I was messing up. I perhaps got complacent after the first push, even though I know that if your opponent expands and you hit them hard at their expansion, any damage you do is only making up your economic disadvantage, not putting you ahead.
Nevertheless, I had been determined to push at about 5.30 with whatever infantry I had, and I did that and it worked well. I did the same in my next game against the terran player and took two supply depots and put him firmly on the defensive. The difference was that in the second game I followed it up with more aggression and eventually won. I think I'll do that all the time in future.
I think I need to decide on a build order and try to stick to it. At present I'm trying to do lots of different things ("maybe I should make some hellions..."), or to decide on the fly how many production buildings I need or when to expand. This not only makes all my builds a bit rubbish, but it's hard to learn what I could do better each time. My next post will be my build order from now on.
Tuesday 15 February 2011
Background
Rank 20 in Bronze League division Grizzly Oscar, 76th percentile in Europe, win ratio 51.4%
Some background first. To paraphrase the mighty Totalbiscuit, my name is Everblue and I suck at Starcraft 2.
I'm a keen gamer although historically most of my gaming has been offline. Some offline games I've loved:
Civilisation
Football Manager
Pro Evo Soccer
Halo
Star Wars - Knights of the Old Republic
Goldeneye
My only real experience of online gaming (apart from some fairly epic LAN parties with the original Halo game back around the turn of the millennium) was World of Warcraft, which I've been playing since 2005. As I type this I am leader of a raiding guild, and despite being casual in terms of playtime I've raided to a reasonable level - top 10% of the raiding population approximately which means top 5% of the player base, and my guild was top 20 on the server at the end of the most recent expansion.
Recently though I've become disillusioned with WoW, and I started looking for a change of pace. I discovered Starcraft 2 through the "I Suck At Starcraft" series created by Totalbiscuit and I've become hooked ever since. I am bad though. Really bad.
My macro is terrible, I am too defensive, and I barely micro at all (running away from a vastly superior force counts as good micro from me). This blog is about me hopefully learning to play better. My first goal is to get out of bronze league and into silver, and establish myself there. The idea behind this blog is to force me to think critically about my game, to analyse some of my defeats, and perhaps to get some advice.
Some background first. To paraphrase the mighty Totalbiscuit, my name is Everblue and I suck at Starcraft 2.
I'm a keen gamer although historically most of my gaming has been offline. Some offline games I've loved:
Civilisation
Football Manager
Pro Evo Soccer
Halo
Star Wars - Knights of the Old Republic
Goldeneye
My only real experience of online gaming (apart from some fairly epic LAN parties with the original Halo game back around the turn of the millennium) was World of Warcraft, which I've been playing since 2005. As I type this I am leader of a raiding guild, and despite being casual in terms of playtime I've raided to a reasonable level - top 10% of the raiding population approximately which means top 5% of the player base, and my guild was top 20 on the server at the end of the most recent expansion.
Recently though I've become disillusioned with WoW, and I started looking for a change of pace. I discovered Starcraft 2 through the "I Suck At Starcraft" series created by Totalbiscuit and I've become hooked ever since. I am bad though. Really bad.
My macro is terrible, I am too defensive, and I barely micro at all (running away from a vastly superior force counts as good micro from me). This blog is about me hopefully learning to play better. My first goal is to get out of bronze league and into silver, and establish myself there. The idea behind this blog is to force me to think critically about my game, to analyse some of my defeats, and perhaps to get some advice.
Subscribe to:
Posts (Atom)