Australia Game Analysis

Talk about the shiny-new post-apocalyptic Advance Wars game here.
Post Reply
GipFace
Rank: Lord of Children Games

Australia Game Analysis

Post by GipFace » Sun Jan 22, 2012 7:55 pm

For the past week, I've been playing some AW4 on random wi-fi to pass the time. This was the first time I have played seriously in perhaps over half a year, so my skills were a bit rusty. Fortunately, I streamed much of this, so I can go back and analyze my own play.

2P-60 Antipode Map, AKA "Australia", despite its lack of airports, has become my favorite pack-in map to play on. Here it is in all its glory:

Image

Days: 30 (originally 40)
Timer: 2:00 (originally 3:00; I am handicapping myself because my opponent gets the full 3:00)

My random wi-fi gauntlet had also consisted of 2P-47 Beaker River and 2P-69 Clown Island. However, it became apparent early in the week that neither of those maps created games that were as epic as the ones on 2P-60 Australia. This is because Australia is arguably one of the best designed maps. It's actually better designed than most of the custom maps on AWBW. So I forced my game to always load this map. (See here for instructions on how to change the random wi-fi rotation)

An asymmetric map, Australia is notable for having no airports and a central position that is slightly left of center, where the mountain range is. 2P begins with a 10-9 property lead. These two factors mean that 1P's first-turn advantage is virtually neutralized. Due to Australia having 39 total properties, there can never be a draw. The lack of airports, as well as the defensive mountain range, cause many of the niche metagame units to see play. Rockets, recons, war tanks, and even the carrier and battleship have all seen play from me on this map. The seaports are nicely implemented here. By not making sea the focus of the map, the seaports become a tech supplement, rather than forcing a slippery slope gunboat battle.

The map is large enough that there's plenty of tempo opportunities, though not too large for turtling (slow, defensive play with mechs) to become useless. The black partition indicates the natural split of the map. If both players capture all of their properties on their side of the split, 1P comes out on top 20-19. It is up to 2P to use his initial property lead to either capture the city left of position "3", or hold the extra city in the center. This can be done with initial aggression and bike hopping, followed up by a strategically-placed rockets. If the map becomes split, both sides will accumulate a sizable army, and one player will be forced to blink first.

Position 1 is 13mp away from 2P's lower base, which means a tank will be unable to reach it in 2 days. Position 2 is 18mp away, so when 1P is bringing up a bike to capture, if 2P is pressuring with an early tank and 1P is not, 1P must take the city 2 spaces to the right of position 2. Position 3 is 15mp away, which means that there's a possibility of 2P disrupting with a recon. Note that if a unit is placed on position 3, 1P cannot disrupt the city to the left with a bike, as the wasteland south of the right city prevents the bike from doing so.

-

Me (Waylon) vs. Brenner (Potter)

Day 19/30. I'm at 11/12 meter, while Potter is at 1/12, having lost his CO earlier. The battle lines had been set for a while, which allowed me to tech to seaplanes. His antiair were built late, so this was a perfect time to initiate a bust. What should I have done?

Image

Here's the optimal set of plays:
- Zone seaplane + md tank will 2hko his CO
- 110A tank will deal 93% to 100D rockets
- If rockets is 1hko'd, cripple north arty with tank
- 4 arty put into range 3 of south city arty
- Rockets 3N
- Move remaining tanks behind arty

It seems obvious, but when you're playing with a 2 minute timer, you don't always make the best plays due to mental and psychological factors. I did the first three plays, but then advance my useless mechs, thus blocking the rockets. My rockets only moved 1N, which meant that 5 indirects covered only a single square. I then placed a tank on the city south of the mountains, which was a silly mistake. I learned that mechs aren't too good when you need to perform a bust.

So this was what I did:
Image

On his turn, he did the best possible set of plays, which was spread crippling my artillery instead of focus firing. Now, despite my misplay on the previous day, the bust can still be successful if I make the right plays here.

Image

- Fire with outer artillery
- Invoke COP
- Target wisely with the two seaplanes (explained below)
- Move up with tanks
- Move up remaining artillery, join if possible
- Drop a mech into the mountains

You can see that the rockets misplay from the previous day had cost me, as its only good target is the 8hp artillery after I poked with the outer artillery. As for the seaplane targets, the north rockets is an obvious one, as it'll be 1hko'd. But then I horribly misplay by targeting the south rockets instead of an artillery or a tank/md tank. The south rockets seems like a juicy target until you realize it can't even fire at anything. I wasted an important seaplane shot. To make matters worse, I didn't use my fresh artillery to join. As you can see below, my bust is about to run out of steam because 1) my rockets no longer has any targets and 2) all my front-line artillery are crippled.

Image

He eventually got COP and fended off the assault. One seaplane was blasted, another ran out of gas, and the remaining were crippled. I was forced to retreat. However, he had to capture a city, and so he sacrificed his CO antiair to launch his final push. (My indirects were forced to target it over the md tanks.) It looked like he might actually take a city.

Image

But wait! I had placed my remaining seaplanes in my carrier for a 1-day heal. They were at 8hp and 4hp, and with my COP ready again, I was going to ride them to victory.

Image

After sniping his final bike (which I didn't really need to do since it was day 29), I immediately ended my turn. He got the message and surrendered.

Image

Woo woo! I barely pulled out the victory. Hopefully you got some insight as to how to manage a bust with many units.

I had some other good games on this map, so stay tuned for more analysis!

Doraleous

Re: Australia Game Analysis

Post by Doraleous » Wed Jan 25, 2012 12:30 pm

More analyses plz

simple_bob
Rank: Game Theorist

Re: Australia Game Analysis

Post by simple_bob » Thu Apr 19, 2012 4:35 pm

Antipode is skewed towards red, not only having the first turn advantage, but more properties are within reach of red in only two days, and can attack by three days.

The black line division that gipface says is the natural split is incorrect. The natural split of the stage is more vertical, and his explanation for the city next to position 1 is entirely false. Who cares if blue can reach this city with a recon in two days, because red can reach the same city and block the city in two days and kill whatever assault blue attempts to send to prevent a two day capture of the city. Even if red is slowed down to a three or four day capture of the city, it can’t be prevented; therefore red always gets the city next to position 1.

Next is position 3, which are two cities that are equidistant for both players to reach within three days for the further city and two days for the closer city. Based on strict distance this point appears to be fair, but in actuality both cities are easier to block and capture for red, because blue has the mountain range to his back and reinforcements must stand on road terrain or else blue sacrifices critical mobility for engaging red, while red gets to stand on wasteland/city terrain making for a much stronger defensive position, and red gets the first turn advantage. Red also gets the tactical advantage of being able to walk across open space versus blue that must walk around a mountain range.

The wasteland/blue city between positions 1 and 3 is a tactical mobility and barricade advantage for red, and a tactical mobility disadvantage for blue.

Position 2 is at the top of the stage and has two cities both left and right. Gipface believes the cities to the left of the 2 are for blue, but they are not, they are battle ground cities that both red and blue can defend to prevent the other player from capturing. Both red and blue require three days to reach the cities to the left of position 2 and with red’s first turn advantage he can prevent blue from ever getting them.

The cities to the right of the 2 have the same kind of positioning as position 1, where if blue attempts to disrupt red from getting these cities, it’s only temporary and with the distance disadvantage from being so far away, red’s capture of the cities can’t truly be prevented, only slowed.

Similar to position 3, red has the tactical advantage of both defensive protection and mobility to reach the cities to the right of position 2 and blue is at the disadvantage for both defensive protection and mobility to reach the cities to the right of position 2.

By this analysis, we can see that the natural split of economies becomes 17-18, with the two cities to the left of position 2 and both cities around position 3 as unclaimed. Once you start factoring in the tactical mobility advantage for red, you can see that red has an easier time grabbing the city to the right of position 3 once blue is pushed back, due to the fact that blue cannot maintain a fight from the roads versus city/wasteland terrain, bringing the economies to an 18-18 tie.

The stage is tactically setup for red to push against blue and not the other way around, because red has more open space to the right of the mountain range then blue does to the left. This allows red to fan out his troops and use the fact that he can out-number the troops to the left of the mountain range, this tactical push back allows red to capture more of the neutral cities, increasing his economic advantage.

I would like to add in position 4, which are the two cities in the south, where blue starts with ownership of the city that is to the right of position 4. When the game enters the late stages of lines being drawn around the chokepoint of the north, blue is at the an even larger disadvantage, now being economically behind, and being at a tactical disadvantage for unit placement, red will start teching to push blue back to his base in the south, leaving blue unable to defend the city to the right of position 4, and depending on how poorly blue has done tactically, or economically, blue can even lose the city to the left of position 4 (but most likely red will not get the city to the left of position 4).

This causes blue to fall even further behind economically with a 19-17 split, and three cities neutral (two cities to the left of position 2, and the city to the left of position 3). This analysis shows that red is always in the tactically advantageous position throughout the lifetime of the entire battle. Blue’s only strategy of the map is to hold out until time expires, and if red doesn’t conquer you, you won, even with the property distribution saying you didn’t.

If you truly believe blue can counter anything I have described here, try it. Play against yourself and see if you let blue do anything beyond what I have told you, that’s what I do. I would like to say; that I have both survived to time expired, because I’m super pro, and beat gipface as blue, due to him making mistakes. I can recognize that the stage is skewed and only with mistakes, does blue have a chance at winning. If gipface would still like to believe blue has a chance, I’d like to see him fix his game to place himself as blue, instead of always making me blue, the biggest proof that gipface knows, red has the advantage.
Last edited by simple_bob on Fri Apr 20, 2012 10:37 am, edited 1 time in total.

User avatar
HPD
Tri-Star CO
Tri-Star CO
Rank: Mentat
Location: The Mountain
Contact:

Re: Australia Game Analysis

Post by HPD » Thu Apr 19, 2012 8:58 pm

Yo, Text McWall, please be so kind to sort this into paragraphs. No offense, but right now it's an unreadable mess.
"So when I say the fudge shaman flies he goddamn well flies and that's that." - Narts
"My motto is that there are far too many women in the world to waste time with men." - thefalman
"It's just that I'm not really aware of how a common conversation goes." - Imano Ob, talking on MSN about talking on MSN
"As for FE8, that was IS' variant of Man Spam - Dudes with Swords edition." - Xenesis

GipFace
Rank: Lord of Children Games

Re: Australia Game Analysis

Post by GipFace » Thu Apr 19, 2012 9:08 pm

You wrote a wall of text and not once is there any mention about the CO unit factor. Wonderful. 2P will get an extra $4k over the first 4 days, thus making his CO tank equip free. This is enough to shoot down most of what you said, as early game disruption is a lot weaker when all it does is feed the enemy's meter.

position 1... Who cares if blue can reach this city with a recon in two days, because red can reach the same city and block the city in two days and kill whatever assault blue attempts to send to prevent a two day capture of the city. Even if red is slowed down to a three or four day capture of the city, it can’t be prevented; therefore red always gets the city next to position 1.
The only way 1P guarantees a capture of that city is if 1P doesn't hop with bikes. By not hopping with bikes, 1P cannot rush to the center and claim a city in which to support his CO tank. By the way, due to the initial property advantage, 2P could afford to lose a recon, though I haven't played that possibility out.

Position 2... the cities to the left... are battle ground cities that both red and blue can defend to prevent the other player from capturing. Both red and blue require three days to reach the cities to the left of position 2 and with red’s first turn advantage he can prevent blue from ever getting them.
If 1P is grabbing the city close to position 1, 1P cannot disrupt. Disrupting is a poor idea when 2P will have a city in which to retreat his CO unit.

Play against yourself and see if you let blue do anything beyond what I have told you, that’s what I do.
Hello FHQ

I have both survived to time expired, because I’m super pro, and beat gipface as blue, due to him making mistakes.
If you want to brag, also include that you pick no one but Tabitha, and this map lacks an airport for me, thus my book of anti-Tabitank tactics is severely limited, especially because I always pick Waylon or Greyfield. lolol~

I’d like to see him fix his game to place himself as blue, instead of always making me blue, the biggest proof that gipface knows, red has the advantage.
It's not possible to rig the wi-fi rotation as 2P. How about YOU get yourself a flashcard and hex edit the ROM, which is exactly what I did?

ThunderWalker
Rank: Elf
Location: Netherlands

Re: Australia Game Analysis

Post by ThunderWalker » Thu Apr 19, 2012 11:24 pm

@Gip: Well, if you know you face him, you can pic a CO that counters his Tabitha there to teach him a lesson.
My sig is a void.

simple_bob
Rank: Game Theorist

Re: Australia Game Analysis

Post by simple_bob » Fri Apr 20, 2012 3:12 pm

You wrote a wall of text and not once is there any mention about the CO unit factor.
This is because my analysis is of the map itself, not unit-battling techniques or strategies, there are far too many tactics based in units and counters for every tactic to say this is exactly how a particular battle will go. Whatever unit set you wish to say is the way to do it, an appropriate counter setup will block it. I don't factor CO's when analyzing a map, because no CO stops the fact that a unit on roads is weaker than a unit on wasteland/city, no CO stops the fact that cities are too far away to be reached, no CO stops the fact that there is more space to the right of the mountain range than to the left of the mountain range, no CO stops the fact that red has a mobility advantage of open space versus blue having to go around a mountain range.

position 1...The only way 1P guarantees a capture of that city is if 1P doesn't hop with bikes. By not hopping with bikes, 1P cannot rush to the center and claim a city in which to support his CO tank. By the way, due to the initial property advantage, 2P could afford to lose a recon, though I haven't played that possibility out.
No kidding you haven't tried it, because it does not work. Play against yourself and see. Sacrificing 4k as blue is totally un-intelligent when fighting from the disadvantage. Plus if blue sends a recon on the first day, blue can't reach the cities at position 3 quickly.

Position 2... the cities to the left...
If 1P is grabbing the city close to position 1, 1P cannot disrupt. Disrupting is a poor idea when 2P will have a city in which to retreat his CO unit.
What are you talking about? Position 2 is in the north, position 1 is in the south, different bases send troops to these locations, and you don't send troops from one base and make a tactical choice.

If you want to brag, also include that you pick no one but Tabitha ... especially because I always pick Waylon or Greyfield. lolol~
LIES!!! I have seen you play as a variety of CO's, as so have I, I mostly play with Tabitha online, because Tabitha is one the hardest CO's to use against an intelligent player, because her CO zone is the hardest zone to charge (requiring the first 30 points out of 60 to come from attacks of the CO unit, versus every other CO (except Forsythe) who can use the defensive meter building technique and other units to charge the CO zone). But if my opponent doesn't stop her CO zone from charging, the game is over, I will know how to zone-push until the opponent takes a form that allows for her CO power to clinch the win. Using Tabitha gives your opponent a strong tactical CO advantage with the fact that her entire army is 100/100(without terrain bonuses), while my opponent will have at least 110/110 units. Getting Tabitha to a CO zone is an uphill battle, because intelligent players won't let it happen, especially with a map like antipode where blue must fight from the tactical disadvantage.

this map lacks an airport for me, thus my book of anti-Tabitank tactics is severely limited
There are many ground based tactics for stopping a Tabitha tank.

Hello FHQ
Sorry for being out of the loop, but what does FHQ stand for?

It's not possible to rig the wi-fi rotation as 2P.
I know you can't fix to say always player 2, it's called not editing the ROM buddy. Further proving you know red has the advantage and you lie to others so you can keep beating them. Playing from the disadvantage is how I've become a good player, not by always playing from the advantageous position.

gipface: I don't know if this is the right place, but nice job on the advance wars 4 chart. Is the submarine labeled dark red because the submarine can dive?

When I say play against yourself, I don't mean play against the computer; I mean you play both sides, red and blue. The hardest part about playing yourself is that you want one side to win, and you will most likely skew your own battle to allow it. Don't skew, be full-force and all your skill for both sides.

Because I am playing from my 3ds, I don't have a computer version running, how can I post my stages here, should I just create some tile formula and post that, and you guys could recreate my levels? And would you guys like me to post my unit production… like this many infantry, that many artillery, and such (with the composition/budget values attached)?

User avatar
Xenesis
Tri-Star CO
Tri-Star CO
Rank: Hydrocarbon Inspector
3DS Code: 2535-4646-7163
Location: 0x020232DD
Contact:

Re: Australia Game Analysis

Post by Xenesis » Fri Apr 20, 2012 3:24 pm

Sorry for being out of the loop, but what does FHQ stand for?
He was a member infamous for arguing about 'strategies' when he only tested them against himself or the AI...most of them didn't work very well at all.
IST wrote:Even the worst individual needs to discover the joys of a chicken statue that is also a pregnant blonde housewife.

simple_bob
Rank: Game Theorist

Re: Australia Game Analysis

Post by simple_bob » Fri Apr 20, 2012 4:59 pm

Thanks for the knowledge xenesis, that's why i added
The hardest part about playing yourself is that you want one side to win, and you will most likely skew your own battle to allow it. Don't skew, be full-force and all your skill for both sides.

I guess FHQ had that problem. lol.

I see your friend code xenesis, and i'll add it, perhaps we can play sometime or trade maps.
Here's my friend code: 301019-695022 (simple_bob)

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest