Recommend
6 
 Thumb up
 Hide
5 Posts

Race for the Galaxy» Forums » Sessions

Subject: Robot was too smart and too lucky for me rss

Your Tags: Add tags
Popular Tags: [View All]
Greg Jones
United States
Washington
flag msg tools
I was Epsilon Eridani versus an Alpha Centauri robot on hard difficulty.

I was looking at a pretty good starting hand. I had three 1-defense military worlds, so three free builds right off the bat. What's more, they were good card generators - great for the early game: Last of the Uplift Gnarssh (green windfall), Smuggling Lair (brown windfall with Consume for 2 cards), and New Survivalists (blue Production with Consume for 1 card). I could Settle two, then Trade for 6 cards. Then after Produce, I could get 6 more, or 3 on the robot's Consume roll.

I also kept Imperium Lords. Might be nice. At least 5 points, and at least 3 cards on Produce.

Avian Uplift Race was my harder discard. I'd only need 1 more military strength to be able to Settle it, and it might go with Last of the Uplift Gnarssh on a green 6-cost development. But a bird in the hand is worth two un-uplifted avians. I also easily discarded Replicant Robots. Looked like my Settling would be military.

I thought it would be smart to start off with a double Settle, maybe whittle away a little of the robot's credit. I didn't need to Settle - Trade. I was prepared for a robot Consume. I could still Consume for 3 cards with Smuggling Lair and Epsilon Eridani, or 2 cards and put down New Survivalists.

Unfortunately I forgot that Alpha Centauri doesn't lose credit when it follows a Settle. It rolled Settle and robot anyway, so double Settle. The robot's first stroke of luck was to flip Deserted Alien Library for 5 points.

Then I called Settle - Trade. I got Tourist World, and that started to look like a strategy. Produce - Consume can be pretty questionable against the robot. It matches too much. However, if it stood a chance against any robot, it was this one. It didn't have any economy yet, and neither Produce nor Consume was its robot action. I already had enough Production capacity to power it: one windfall and one Production world.

The robot called Develop and got its first 6-cost development chip.

Next round I called Develop - Produce, to get down Galactic Imperium, load my worlds for the Consume for points, and grab a few cards for the Production bonus on Galactic Imperium.

After that I was mostly ready to go on my engine. I could call Settle and get down Tourist World, and Consume/2xVPs, but I thought better of it. A match would give the robot x2, the last roll I wanted it to get given my strategy. Anyway I would have to Produce and miss Consuming in the next round, so it wouldn't accelerate my second or further Consumes any.

So I decided to make a speculative call: Consume/2xVPs and Produce. The robot's action was Settle, so there was a good chance it would go perfectly for me. Worst case, I Consume on Smuggling Lair and Epsilon Eridani for 2VPs and 3 cards, and Produce for 3 more cards.

Here's where the robot started to get smart. It rolled x2 anyway, and matched my Produce. We both got 2VPs, and I overflowed my hand limit. The robot's economy went up to 3.

So now I was in the same situation, except I had more cards and the robot had a bigger economy, but I couldn't pull the same trick as last turn, since I might end up just really overflowing my hand limit. I called Settle - Consume/2xVPs. The robot matched me.

On the next round I called Produce to load my worlds, and the robot called Trade while I had no goods and couldn't leech it for a couple points. I also called Settle to add another cheap Production world.

It was looking bleak. With only 4 VPs left in the supply, I couldn't call Consume/2xVPs without ending the game, and I wouldn't win. I was down 19 to 33. I did have Terraformed World (5VPs) in my hand, and a lot of cards, so I figured I'd try to catch up a little on tableau points. Then I'd make a lunge at a win with a Consume/2xVPs; as long as the robot didn't match me I could gain on it.

I called Settle - Settle and the robot called Produce. So I put down Mining World (3-cost brown Production world with +1 card on Produce) in addition to Terraformed World, spending most of my cards, most of which I got back on my Production bonuses. The robot Settled 3 points worth, so I gained a little on it.

I drew Galactic Trendsetters. Now there was a way to win. Unfortunately I was one short to Settle it. So I called Explore +1/+1 and Settle. The robot called double Settle, so I got to double Settle anyway: Galactic Trendsetters and for free, Runaway Robots (1-defense brown military windfall).

Now I was set for a big Consume. 5 goods on Last of the Uplift Gnarssh, New Survivalists, Artist Colony, Mining World, and Runaway Robots. 2 to Tourist World for 3, 1 to Galactic Trendsetters for 2, 1 to Epsilon Eridani for a point and a card, and 1 to Terraformed World for 1. Double it would be 14. If the robot didn't match x2, I could beat it.

So I called Consume/2xVPs and Produce. There was no purpose to the Produce, but calling it makes a single robot match match Produce instead of x2. Thanks to
Daniel Cristofani
United States
Portland
Oregon
flag msg tools
Avatar
mbmbmbmbmb
for teaching me this tactic.

Unfortunately it didn't work. The robot rolled Produce and match. What are the odds?? The natural Produce call prevents the match from applying to Produce. It applied to x2 and the robot got 10VPs instead of 5.

Final score:

Me - 45
Robot - 48

Me:
Imperium Lords
4 = 1 x 4 military worlds
2 = 2 x 1 Imperium card

Other tableau points
17

VP chips
22

Robot:
9 = 9 x 1 6-cost development chip

Other tableau points
17

VP chips
22

The robot was too smart for me. If the robot "chooses" the right roles, it'll beat you every time. It plays by unfair rules. Your only chance is that it plays randomly, and that means sometimes it plays stupidly. Not this time.
4 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
mumu shanshi
msg tools
mbmbmbmbmb
morningstar wrote:
Unfortunately it didn't work. The robot rolled Produce and match. What are the odds??


Not that unlikely, actually.

1) For a specific roll of Produce and match, the odds are 1 in 18, or about 5.6%.

2) Given that you chose IV:x2 and V, the odds that the robot will also choose IV:x2 and V (when the robot action is settle) is 1 in 6, or about 17%.

3) Given that you chose IV:x2 and V, the odds that the robot will have IV:x2 as one of its two choices (when the robot action is settle) is 5 in 18, or about 28%.

The robot had a better than 1-in-4 chance of also consuming double in the final round.

I do sympathise, though...
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Greg Jones
United States
Washington
flag msg tools
mumushanshi wrote:
morningstar wrote:
Unfortunately it didn't work. The robot rolled Produce and match. What are the odds??


Not that unlikely, actually.

1) For a specific roll of Produce and match, the odds are 1 in 18, or about 5.6%.

2) Given that you chose IV:x2 and V, the odds that the robot will also choose IV:x2 and V (when the robot action is settle) is 1 in 6, or about 17%.

3) Given that you chose IV:x2 and V, the odds that the robot will have IV:x2 as one of its two choices (when the robot action is settle) is 5 in 18, or about 28%.

The robot had a better than 1-in-4 chance of also consuming double in the final round.

I do sympathise, though...


Yeah, I realize there are various ways it could choose x2, I just thought it was insulting that it came about it in the single most unlikely way.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
John "Omega" Williams
United States
Kentwood
Michigan
flag msg tools
designer
Avatar
mbmbmbmbmb
A very nicely detailed session report with a good play-by-play of decisions and thoughts as it went along. A close struggle too!

So in retrospect. What could, or might, you have done diffrently?
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Greg Jones
United States
Washington
flag msg tools
Omega2064 wrote:
A very nicely detailed session report with a good play-by-play of decisions and thoughts as it went along. A close struggle too!

So in retrospect. What could, or might, you have done diffrently?


Well I already admitted that calling Settle - Settle on round 1 was a goof. But I think it was the right move anyway. It got me 6 cards on the Trade, whereas starting with Settle - Trade would have only got me 4 cards, and 3 cards on the next Settle - Trade.

If I really wanted to whittle down the robot's credit, I remembered later in the game the way to do it to Alpha Centauri is actually to call Develop when it was 0-2 credit. It doesn't get a 6-cost development chip unless it matches you or calls its own Explore. That's actually fairly likely, but I think you have to take some risks against the hard robot. If it doesn't match and it flips a regular Development, it loses credit, and you can perhaps avoid it getting a 6-cost development chip.

Secondly, the whole startup of my Consume engine didn't go well. The speculative Consume/2xVPs - Produce was a good call, I think, although the robot rolled against the odds and made it not work. However I should have followed up with Settle - Settle instead of Settle - Consume/2xVPs. Then the robot wouldn't have matched my Consume (I think it rolled a match, not a natural x2). I called Settle - Produce the next round anyway. I should have called Settle - Settle then Consume/2xVPs - Produce. The outcome would have been the same for me, but worse for the robot.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Front Page | Welcome | Contact | Privacy Policy | Terms of Service | Advertise | Support BGG | Feeds RSS
Geekdo, BoardGameGeek, the Geekdo logo, and the BoardGameGeek logo are trademarks of BoardGameGeek, LLC.