SGOTM 12 - One Short Straw

[LowtherCastle]

Okay, here's the significance for your turnset, mdy, as I see it.

Principle 1: We now have an extraordinary synergy between spamming spies/Confu Misses and our short- and long-term strategy.

For example, a mature GP Farm puts 1h into a spy while growing like mad. Then it 2pops the spy for max overflow (33h, I assume, because of the forge). It then builds MoM for 1 turn at +14hpt, putting 47*2.5=117h into MoM. Twelve turns later it has ~449h in MoM. Same goes for Moscow, Fish, Siberia, Pigs, Bahamas, Bermuda, Fish/Corn, Bananas, Paris, Orleans, etc., all to varying degrees of effectiveness.

Our priorities were already:
1. warmongering
2. infrastructure
3. EPs accumulation

Now, this adds

4. Confu spread to major cities asap
5. Failed gold harvesting

If you want overflow into NE, build a Confu Miss in Pigs for 1h and 2pop into NE immediately. Assuming it succeeds in spreading to GP Farm, it means the difference between finishing the GP Farm ch before or after the uni poprush:

GP Farm:
T+5: Forge(3 pop whip)
T+7: Library ( 2pop whip)
Start courthouse, switch to university once we have education, whip as soon as we can and still work all food sources.

-or-

Using your T+1 format from your PPP, with T+5 being T140.

The Pigs Confu Miss will hopefully spread COnfucianism on T+8 ([COLOR="red"]in red[/COLOR]).


Turn build                    food  fpt prod   b-hpt over/whip  population  citizens work
---- -----                    ----- --- -----  ----- ---------  ----------  -------------

T+5: Forge done, build lib    18/30 +13  0/90   +1    +4b-h         5       all food tiles(!)
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 
T+6: lib(1t), build ch        16/32 +14 11/90   +1                  6       all food tiles(!)
T+7: ch(1t), build lib        30/32 +11 1/120   +5                  6       food tiles + mine
[COLOR="Red"]T+8: lib, 1pop whip now       25/34 +14 22/90   +1    +30b-h       7>6      all food tiles(!)[/COLOR]
T+9: lib done, 3pop ch now    23/34 +12 1/120   +1    over+90b-h   7>4      all food tiles(!)
T+10: ch done, build uni      21/30 +13 ?/200   +1    over(?)       5       all food tiles(!)
T+11:                         19/32 +14                             6       all food tiles(!)
T+12:                         17/34 +15                             7       all food tiles(!)    
T+13:                         32/34 +15                                     
T+14: 4pop whip uni now       30/36                                 8

T+14 it T149 so the uni is done T150, as per your schedule.

Note that a 3f tile =~7.5h in a food-rich city with Confu+forge, when you're poprushing at low population, whereas the 4h1c tile = 6h1c. GP Farm should work farms rather than the mine, except if you needed the hammers to finetune the hammers in the build.

Similar in Fish: a 1f3h mine = 2.5h+4.5h=7h, whereas a farm = 7.5h while we're poprushing. Fish needs farms! Moscow needs two more farms. Rheims needs farms to irrigate the corn. Farms everywhere!

--------

xpost
.

 

[LowtherCastle]

In that case we still have some control, however we ware still partially limited to when it is convenient to build the wonders.

True.

I thought we were going to commit Moscow semi-permenantly to a unit spam after we built Oxford/heroic Epic there. How would this fit with plans to build MOM there ourselves?

Again true. Moscow will be a fine art. Either we don't build any more wonders here, or we squeeze something in at some point when we've reached a saturation point with units. I included Msocow on that list simply because it can put hammers into a wonder so fast and people have mentioned building wealth there.

Once we have built our uni's we can put a few hammers into Oxford, while Moscow builds other things for the overflow, and maybe the NE whilst we build the university in pigs, other than that, in the near future we simply don't have the spare production for failure gold.

Well, for your turnset, I think the focus is 1) getting our cities mature asap and possibly 2) delaying NE a bit (not even sure about that). Otherwise, I agree that Moscow probably shouldn't divert production into failed gold now and we really don't have any mature city for your turnset.

We just need to be really wise about it. An immediate COnfu Miss from Pigs for GP Farm pays off hugely because it means GP Farm will be mature that much sooner, thus free to start putting poprushed hammers into failure gold.

A Confu Miss into Bahamas asap would also be great, if anyone can figure out how. Maybe build/1pop a galley next in Bananas asap?

---------------

I'd just like to emphasize that OrgRel+Confu spam is useful far beyond just infrastructure (the usual reason) because of failure gold AND spamming spies/ConfuMisses for stealing techs. It's extremely useful for this map because we are so commerce-poor and not planning to build an cottages. It's a Godsend.

 

[LowtherCastle]

Okay, if we're prioritizing a GE in Moscow, then let's hire the engineer immediately and re-adjust the MM accordingly.

(edited)
.

 

[mdy]

I am going to be very busy over the next few days, and realistically I am not going to have time to play again until next monday, so I think it would be best if I handed over to the next player now.

 

[babybluepants]

Ok. I guess I'm up then. Might not be able to do a provisional PPP until tmr morning or so. We'll see...

@LC,
So, you are proposing all this for after Universities? Sure. I think that wealth-building is in order after we have the basics, namely gran-lh-lib-(uni)-for, in our case also ch. Along with taking care of all necessary unit building, though Moscow's mostly in charge of that.
Edit: I'm still wondering about our decision not to adopt Caste-Pac. It may be the biggest potential contributor to research pace towards Lib under extreme deficit spending. We're not proposing a very massive unit build-up, so Pac would barely even cost us more, I'd think. The OR bonus in building cash is relatively minor next to +125% from marble/stone and forge.

Did we decide to put Moai in Cuba, btw? That's another potential fail-cash-cow. I still think either Bahamas or Bermuda is better long term. Bahamas seems to have about 50 base hammers with Moai, post-workshop.

 

[babybluepants]

ROSTER UPDATE

babybluepants Played T27-40 T75-90 UP
Mitchum Played T0-27 T90-96 ON DECK
bcool Played T40-48 T96-115
LowtherCastle Played T48-57 T115-135
mdy Played T57-75 T135-140 JUST PLAYED
shyuhe M.I.A.
ZPV M.I.A.
mystyfly M.I.A.
Silu lurking
Soirana erasing

 

[Mitchum]

@mdy Can you please advance the test game forward to the current turn while you still remember what you did with all 5,000 workers we have?

 

[Mitchum]

Okay, if we're prioritizing a GE in Moscow, then let's hire the engineer immediately and re-adjust the MM accordingly.

(edited)
.

If we hire an engineer specialist in Moscow starting on T+0, our odds on T+12 will be 49% (GPro), 40% (GE) and 12% (GS).

If instead, we push to have Pigs pop the next great person and hire an engineering speicialist as soon as possible without beating Pigs, our odds on T+29 will be 46% (GE), 45% (GPro) and 9% (GS).

So it looks like we only gain 6% by not popping a great person in Moscow next. I agee that it makes the most sense to push Moscow for the next great person by hiring an engineer specialist. If we get a GE (or a GS), we can do our money dance. BTW, what would we do if we get a GPro? I'd be inclined to settle him for the 2/5 base in Moscow.

If we don't get a GE and continue to work an engineer specialist without building any additional gpp-polluting wonders in Moscow, our odds will be 78% (GE) and 22% (GPro). If we start building more wonders in Moscow after Oxford (which alone will drop the GE % to below 75%), things start looking worse.

 

[babybluepants]

BTW, what would we do if we get a GPro? I'd be inclined to settle him for the 2/5 base in Moscow.

If we get a GP, we're saving it for the 3 or 4 GP GA.

If we don't get a GE and continue to work an engineer specialist without building any additional gpp-polluting wonders in Moscow, our odds will be 78% (GE) and 22% (GPro). If we start building more wonders in Moscow after Oxford (which alone will drop the GE % to below 75%), things start looking worse.

We'll also actually have to make it to 700-800? gpp without running other specialists, and lose the potential benefit of Rep sci with OU. I think it's probably more realistic to think of that one as a 50-50 or so.

 

[babybluepants]

Just thought I'd check...

GAME SCHEDULE

Turn	Date	T/Day	Player
000	Sep 14
012	Sep 25	1.09	Mitchum
027	Sep 30	3.00	Mitchum
040	Oct 03	4.33	bbp
048	Oct 06	2.67	bcool
057	Oct 13	1.28	LC
075	Oct 18	3.60	mdy
090	Oct 25	2.14	bbp
100	Oct 29	2.50	Mitchum
114	Nov 5	2.00	bcool
135	Nov 11	3.50	LC
140	Nov 16	1.00	mdy
152	Nov 22	2.00	bbp
164	Nov 26	3.00	bcool
178	Dec 1	2.80	LC
181	Dec 6	0.60	bbp
-----------------------------------
155	Nov 23	2.14
170	Nov 30	2.14
185	Dec 7	2.14
200	Dec 14	2.14
215	Dec 21	2.14
225	Dec 28	1.43	
235	Jan 4	1.43
250	Jan 11	2.14
265	Jan 18	2.14

Average up to now = 2.16 turns per day
Remaining (assuming ca. T270 finish) = 1.97 turns per day
Remaining (had we not gotten an extension) = 4.06 turns per day

I get very excited about stats...

Posts per day = 29.81
LC's posts per day = 8.76
Posts still needed to break MW's SGOTM 7 record = 1,379 (30.64 per day until deadline)
Posts for LC to break his SGOTM 7 record = 439 (9.76 per day)

 

[bcool]

If the plan is to beeline Communism, wouldn't it be better to focus on GS? How long will it be before we need a great engineer? 60+ turns no?

The GS will help us immediately.

If we do decide to go for Moscow next, the Great Engineer will just sit (and cost us 1 gold per turn)
The Great Prophet probably should be saved for a Golden Age later or I think the Shrine would be better than settling him (We are planning on building WallStreet I assume and Fish isn't a bad place for it)

How much do we really decrease the odds of getting a great engineer if we don't pop the next GP in Moscow? Is it really that horrible to run an engineer in Pigs and in any other city that has a chance to produce a GP for us?

I'm a big fan of getting the Great People we need when we need them if we can control it. And I think we can...

 

[Mitchum]

If we get a GP, we're saving it for the 3 or 4 GP GA.

Really? We'd hang on to a GPro for that long and lose out on the benefit of the settled GPro (I forgot about the base 3 under Representation for the settled GPro)? I think it will be quite easy to generate mutt great people as the game goes on based on past experience... GSs, GMs and GAs are quite easy to get when running Caste System, although trying for GArts would be a pain. We will get the free GE from Fusion. If we get a GSpy, GArt or another GPro along the way, that's one less mutt we need to generate. My thinking is that our golden ages would be triggered by:

GS
GS + GM
GS + GM + Gx
GS + GM + GE + Gx

We only need two mutts... which shouldn't be that difficult if we put our minds to it.

We'll also actually have to make it to 700-800? gpp without running other specialists, and lose the potential benefit of Rep sci with OU. I think it's probably more realistic to think of that one as a 50-50 or so.

I'd hate to bank the gold laurels on a coin flip!!! If we don't get a GE and another top team does, I think we can kiss the gold goodbye.

EDIT: x-post with bcool

 

[babybluepants]

I'm a big fan of getting the Great People we need when we need them if we can control it. And I think we can...

For starters you can't control getting a GE. The more pops we get from Moscow the better the overall odds.

Ideally we need something like:

#	Type	Use	Needed by
03	GS	Printing Press	T+25
04	GS	Chemistry		T+30
05	GS	Sci Method	T+35	
--	GM	Sid's Sushi	
06	GE	Mining Inc.
07	GS	Golden Age 1	T+64
08	GS	Golden Age 2	T+76
09	G?	Golden Age 2	T+76
10	GS	Golden Age 3	T+88
11	G?	Golden Age 3	T+88
12	G?	Golden Age 3	T+88
13	GS	Golden Age 4	T+100
14	G?	Golden Age 4	T+100
15	G?	Golden Age 4	T+100
16	G?	Golden Age 4	T+100

Plus, maybe some more, I'd hope.

The turn estimates are complete guesswork - they seem too fast maybe. That's a total of 6 mystery GP's, of which 3 could be type 1, 2 type 2 and 1 type 3. We're bound to get an accidental Artist from Pigs, and we could certainly farm a 2-3 Merchants in a city with some gold multipliers (perhaps dedicating GP Farm to something like this?). We'll also likely pop an accidental spy. That may need to be used to infiltrate, though.

Moscow's our best bet for the odd ones out. It would need 2 more pops after this one by T+100, most likely at the magnitude of: nextGPCost + 1000-1400 + 2000-2400. I figure we can generate approximately right around 4000 total gpp in Moscow by T+100 (and that's beyond rough - making unfounded assumptions about Caste-Pac, corp spread, etc).

Yes, I'd prefer having all the GS we can before Lib, but I think delaying the Moscow pop until then has potentially negative repercussions.

Edit: Maybe I'm way off assuming 4 GA's?

Edit: x-post

I'd hate to bank the gold laurels on a coin flip!!! If we don't get a GE and another top team does, I think we can kiss the gold goodbye.

One at 40% + one at 50% isn't a coin flip.

 

[bcool]

For starters you can't control getting a GE. The more pops we get from Moscow the better the overall odds.

I understand this, but is a ~50% chance from Moscow and a bunch of ?15% chance from Pigs, others significantly worse than 2 ~50% chances from Moscow?

 

[bcool]

Worst Case scenario we will/could get Ironworks built about the same time as we need a Great Engineer and aggressively run engineers in Moscow if the ~50% in Moscow and 2? 15% chances in Pigs don't work out.

 

[babybluepants]

Maybe not, but 2 x 50% from Moscow + a bunch of 15% elsewhere is even better. Also, as I've said before, I believe that getting one out of the way right now gives us more freedom later.

I could certainly be convinced, though, and it seems that the consensus is shifting towards GS anyway.

Edit: x-post - not if we build OU+HE there. IW elsewhere with planning for this ahead of time is doable, though.

 

[Mitchum]

One at 40% + one at 50% isn't a coin flip.

You're right. It's two coin flips...

Actually, I get multiple percentage chances confused. So are our actual odds in Moscow are something like [1 - (1 - .5 )(1-.4)] = 70%?

Then, if we throw in two 15% odds we get [1- ((1 - .5 )(1-.4)(1-.15)(1-.15)] = 78%?

 

[Mitchum]

Edit: x-post - not if we build OU+HE there. IW elsewhere with planning for this ahead of time is doable, though.

IW in a city with 10+ workshops and a few mines will be better than IW in Moscow anyway...

 

[babybluepants]

You're right. It's two coin flips...

Actually, I get multiple percentage chances confused. So are our actual odds in Moscow are something like [1 - (1 - .5 )(1-.4)] = 70%?

Then, if we throw in two 15% odds we get [1- ((1 - .5 )(1-.4)(1-.15)(1-.15)] = 78%?

I never got probability. Simply added seems fine, I think, so 90% in this case. If you flip a coin twice, for instance, you should presumably be getting heads 1 out of 2, not 75% chance of 1.

We had the same issue last game:

We focused on the UN GE in Delhi to the exclusion of everything else, then failed at 93%. Meanwhile, the Ducks got 2 GE @ circa 40%, without even really trying. I calculated after the game that we could have had something like 2-3k extra beakers had we run scientists in a sustainable manner.

I've had games where I've popped (no exaggeration) 3 GA in a row from 10% NE odds. Others where I get everything I want, despite odds. Getting the GE is pure luck, IMHO.

Don't really know what's best.

 

[Mitchum]

I never got probability. Simply added seems fine, I think, so 90% in this case. If you flip a coin twice, for instance, you should presumably be getting heads 1 out of 2, not 75% chance of 1.

You definitely don't add them. In two coin flips, the possible outcomes are:

heads + heads
heads + tails
tails + heads
tails + tails.​

So, there is a 50% chance you will get one heads, 25% chance you will get two heads and 25% chance you will get zero heads. Overall, the odds of getting a heads is 75%, not the 100% you'd get by adding them.

Unless you have 100% chance in a city, there is always a chance that you won't get what you want. I too have had a lot of GAs from the NE even when running 10+ scientists in a city. We need to do what we can to increase the odds of a GE without sacrificing the other parts of our game.