SGOTM 12 - One Short Straw

[bcool]

The battle was very very close (and I realize now that we probably would have lost Orleans to Churchill if I didn't win 70% and 83% odds on the last two units.

So we are significantly behind schedule on Mao

We have
Macemen (7)
2 CRIII
1 CRII with 11/10 xp
1 6/5 xp CRI
1 5/0 xp
1 3/0 xp
1 6/10 CI MedicI

catapults (3)
2 5/0 xp
1 3/0 xp

swords (2)
2 5/2 xp CRI

chariots
1 3/0 xp

axes
1 6/10 xp CII
1 0 xp

2 Galleys
1 Trireme (next turn)

We probably need the chariot and axes for city garrisons (Pigs, GP Farm, Rheims?)

Fort situation (due to the delay of Orleans falling) (Not going to be the reason we are slowed down I think)
T156 finish 1 fort 1S of Paris and 3/7 of another
T157 finish 2nd fort SSW of Paris
T159 finish 3rd fort and 3 spare worker turns (could bring over another worker to finish on T158 if necessary but need a space in a galley)

Fish could build 2 catapults by T161-T162 (accuracy catapults are as good as unpromoted Treb at knocking down walls) Maybe more if we want to whip that poor city more.

Moscow could build 1 3-xp Macemen/Treb 3 5-xp Treb by T162

Paris can build 2 more Galleys by T161?

If we still want 2 Macemen and 4 catapult to land behind Shanghai (I think T164 they can land 1 catapult would be 3 xp)

SoD from Moscow with 4 Treb could be on Mao's borders by T166

 

[bcool]

I have to go for the rest of the day, but I thought someone might have time to look at/ finish the test game

This is pretty accurate and it is the turn before the real game. I had some doubts about how I did MM on Fish and I think there is an extra grass tile in the test game.

I think we need 1 more worker turn on fort near bahamas to match real game too.

 

[babybluepants]

That battle was uncomfortably close, and reinforces my opinion that when we invade China we should have a few more maces than originally planned to avoid something like this happening again.

What's your estimate?

 

[LowtherCastle]

If we closed borders with Churchill to block his stack how badly would this hurt our economy? Churchill can be bribed into peace with Churchill in exchange for CS and theology, perhaps this would be a more cost effective option?

Closed borders with CHurchill costs us 1gpt in trade route income, I think. But it also severs our one trade route path to Gandhi and Roosy, if we lose the path through Mao with us and Roosy DoWing Mao and us iceballing dG.

If Churchill captures Marseilles, Orleans gains the bananas and gems. Seems like a good deal to me. But it also makes us a land target of Churchill next time he rolls the DoW dice...

Not sure I want to get Churchill one step closer to Macemen and turn his uber-capital into an UBER-CAPITAL with Bureaucracy.

Btw, guys, the reason my calcs for Nanjing's border expansion were so terribly off is that I was using +2 for the colosseum. That expansion may not happen till T160 or later. Oh well...

 

[LowtherCastle]

That battle was uncomfortably close, and reinforces my opinion that when we invade China we should have a few more maces than originally planned to avoid something like this happening again. If the RNG had been only a little less kind to us we would have had to break of the attack, the delays this would cause us if it happens in China should outweigh the extra time it takes us to build a bigger offensive stack.

Apples and oranges. The Orleans Campaign was an impromptu campaign patched together with a ragtag army to create a path through dG to Mao, while we put together a juggernaut to bulldozer Mao.

bcool courageously took on 13 (14?) units with 6 CRII maces, 3 cats, 2 swords, and 2 chariots. Against difficult odds, he succeeded. Kudos to bcool!

Now we will have 3 CRIII maces, 3 CRII maces, 1 medic mace, 4 accuracy cats (or 2 cats + 1 treb) + 4 CRII and 1 CRI trebs to cowardly bash down on 3-4 longbowmen. With a steady stream of 2 CRII trebs every 2 turns arriving thereafter. We will have 2 galleys protected by a CI trireme to run a Pincher's attack on Shangahi, allowing us to threaten Shanghai for only 2 turns before we capture it. We will have all the odds in our favor.

Since Mao is steadily building city improvements in every one of his cities and still has improvements to build in Shanghai, there's no reason to believe he'll change, unless Roosy happens to DoW him. But we will have a spy informing us of the exact standing of Shanghai before we commit ourselves to the DoW.

1. There is absolutely no comparison between the two campaigns.
2. If anything, we'd need more trebs, not maces, but the steady stream can include 1 or 2 maces every so often, to replenish defeats at long odds.
3. Since the trebs will be coming steadily and we see the state of Shanghai before DoW, I see no reason for any concern.

 

[mdy]

If we never have to face more than 3-4 longbows in a city I agree that the force LC described would be sufficient. My concern is that if we are faced with a much larger stack similar to what we faced in Orleans, only with the more difficult to destroy longbows we could run into problems.

By this point in the game the AI's normally have at least one modestly sized stack (even if we are confident he hasn't been building units recently he could have upgraded an archer stack he built some time ago.)

I think we should attempt to scout China's core cities before we have to commit to the force size for the Mao campaign to resolve this uncertainty.

If we do build another couple of maces we could make them combat 1 cover which are almost at good at taking cities and would allow us to take out longbows in the field with better odds than our CR maces.

 

[babybluepants]

2. If anything, we'd need more trebs, not maces, but the steady stream can include 1 or 2 maces every so often, to replenish defeats at long odds.

Relatively speaking, we do need more siege than mace, but I think we need both.

If we never have to face more than 3-4 longbows in a city I agree that the force LC described would be sufficient. My concern is that if we are faced with a much larger stack similar to what we faced in Orleans, only with the more difficult to destroy longbows we could run into problems.

By this point in the game the AI's normally have at least one modestly sized stack (even if we are confident he hasn't been building units recently he could have upgraded an archer stack he built some time ago.)

I think we should attempt to scout China's core cities before we have to commit to the force size for the Mao campaign to resolve this uncertainty.

If we do build another couple of maces we could make them combat 1 cover which are almost at good at taking cities and would allow us to take out longbows in the field with better odds than our CR maces.

+1. This sounds right to me. I think we're planning on the Shanghai assault without thinking of it as part of a war. We did the same in Paris - just barely took it in 1t, then struggled to keep going. It's reasonable to assume that we'll have to kill 7-8 units taking Beijing and probably absorb a bit of counter, IMO.

At this stage of the game, I would expect to blast Shanghai, absorb Mao's SOD hit and/or kill some units in the field and then immediately keep going for Beijing. Commitment to a steady stream from Moscow+Fish can let us attack early and try CF-ing to heal maybe, but we should keep in mind what's required for the whole invasion, I think.

The DOW-CF cycle did also result in DG spamming a lot of units, btw. Typically, he shouldn't have more than 2 archers and maybe one more unit in cities like Orleans and Marseilles.

My main question here is still whether we need unit production from other cities. Once a SOD of a certain size (25-30?) is needed, it should be more efficient to spam from multiple cities at once, so as not to incur ongoing costs.

 

[LowtherCastle]

@mdy

I agree with that. Now that Mao finally built some roads for our spy, he can go 3NW this turn and see both Beijing and Shanghai. That's plenty of advance knowledge for the Shanghai attack. If he has a stack, it's most likely on his western border, where he has common borders with Roosy.

If there's a SoD at Guangzhou or his western border, we can find out about that later. In any case, our SoD will gradually grow in size, so it will be prepared to handle anything. After the 4th or 5th suicide treb, our trebs will have positive odds of withdrawing so we shouldn't lose more than about 5 trebs per city, max. our maces can alway take two turns to capture the city, if needed. Furthermore, we're not planning to wipe Mao off the map.

 

[LowtherCastle]

Note to bcool: I advanced the game save. Pls, download the new version for your turnset.

Since bcool said he was gone for the rest of the day, I took the liberty of moving the spy 3NW so we could see into Beijing. It has 5 lbs, an archer, and a settler. The new save is of course uploaded.

 

[LowtherCastle]

I tested suiciding trebs against Beijing with 5 lbs and an archer. Our 5th treb had every lb down to its minimum collateral damage (3.6/6). That gave our CRIII maces 91% odds, CRII maces 67% odds. I suicided 1 more treb and 2 cats, and the odds for our CRII maces were 99%. A third of our seige attacker survived.

Of course, this will be different each time, but if you suicide enough trebs, the numbers even out. One time no damage to the lb, one time he's knocked down to 1.5/6, etc.

The only reason we need more than 7 maces is if we're against a stack of 15 and we aren't satsified capturing the city in 2 turns. To me, the question is, how many trebs do we want? We've got plenty of maces already.

Since maces cost 10h less than trebs, maybe there's breakeven point where trebs become less useful than maces. Here's another way of comparing them:

attack                       Odds of surviving
------   ----------------------------------------------------------------
         CRII treb  CRII mace  CRIII mace  mace  treb  CRI mace  CRI treb
         ---------  ---------  ----------  ----  ----  --------  --------
  1         25%         8%        21%
  2         45%        18%        38%
  3         53%        34%        45%
  4         82%        38%        77%
  5         85%        69%        80%         
  6         90%        64%        91%       51%   60%     57%       69%      

Note: The 64% was against a CGII DrillI lb
  but the 69% was against a CGI DrillII lb

For the 6th attack, I added the odds for units built outside of Moscow/Fish.

Since seige also have a withdrawal capability, I think over the long run, trebs are simply better.
.

 

[bcool]

Quick break here at my Mom's house. Do we want to trade engineering for feudalism to Gandhi?

So focus on Treb builds even in Fish then?

edit: acknowledging LC's message about the save.

edit: out of curiousity how do knight's compare LC (with the possible trade to Gandhi, Guilds might not be too far away)

 

[Mitchum]

Just so you know, I'm following the action on my phone... but it's hard to type too much and I don't have access to the game. Keep up the good turnset, bcool. Things seem to be shaping up even though the RNG has been working against us a bit.

 

[mdy]

I wouldn't trade engineering to Gandhi. If he's researching engineering he can't be researching education.

We might want to check the border cities with America as well, just in case.

Using suicide trebs is clearly more hammer effective than using maces, though I still think it might be beneficial to build 1-2 combat1 cover maces to allow us to attack longbows in the field, and to give us a few reserves in case of bad RNG.

 

[LowtherCastle]

Quick break here at my Mom's house.

Happy Thanksgiving, bcool!!!

To you and everyone else who celebrates it!

 

[babybluepants]

Maces will be useful for a long time, and we could use upwards of 30-40, I think.

Our world conquest should go something like:
1) Mace + Treb
1a) Mace + (Musket) + Treb
2) Upgrade to Cannon
3) Knight or Cuirassier + Cannon + left over Maces/Muskets
4?) Upgrade to Cavalry / Rifle (I think entirely unnecessary - Rifling is off our beeline anyway)

The critical step is (2). In other words, I think we can kill all AIs with basically just maces and knights, assuming we get to Steel fast enough (to kill them all before they get Rifling).

Horse Archers are an intermediate option, btw, for being upgradable to Knight-Cui-Cav. Over-building siege and then chasing it down with 2-move units can be great where the length of supply line is significant - such as we will have here soon.

We certainly don't wanna self-tech Guilds, though, so we're stuck with 1-move units for the China war and maybe further. I think we should build enough maces right now to not have to stop warring faraway from home at some point in the near future.

 

[babybluepants]

No Engineering to Gandhi. We have no real use for Feud right now anyway.

 

[babybluepants]

Happy Thanksgiving, bcool!!!

To you and everyone else who celebrates it!

In Canada we're a month ahead of you guys.

 

[LowtherCastle]

probably absorb a bit of counter ... absorb Mao's SOD hit and/or kill some units in the field

Mao has no metal. What counter are you expecting? Horse archers? I guess we could have 1 pikeman in our stack.

I think we're planning on the Shanghai assault without thinking of it as part of a war ... we should keep in mind what's required for the whole invasion ... assume that we'll have to kill 7-8 units taking Beijing.

Okay. Let's assume 4 lbs in Shanghai, 8 lbs in Beijing, 10 lbs in Gunagzhou. The other cities we don't really care about.

Looking at the map, we have this sequence:
T+0 DoW, sea pinchers move 2 tiles SoD has 8 trebs
T+1 Sea pinchers land W, SoD moves to Shanghai-2SE
T+2 bombard + SoD to Shang-1SE gain 2 trebs
T+3 capture Shanghai (lose 4 trebs)
T+4 Sea pinchers move next to Beijing, SoD (w/8 trebs) leaves Shanghai gain 2 trebs
T+5 bombard, SoD moves next Beijing
T+6 capture Beijing (lose 6 trebs) -- cease fire gain 2 trebs
T+7 heal + fort Shanghai-SW
T+8 heal + fort Shanghai-SW gain 2 trebs
T+9 Shanghai borders expand, galleys move through fort
T+10 DoW, Galleys enter Mao's waters, SoD crosses border gain 2 trebs
T+11 bombarders land next to Guangzhou, galleys pick up more trebs, SoD moves
T+12 Bombard, more units land, SoD moves
T+13 Capture Guangzhou

Conclusion: Our initial SoD needs 8 trebs, including ~5 CRII, the rest whatever. Everything else comes from Moscow, turn by turn.

My main question here is still whether we need unit production from other cities. Once a SOD of a certain size (25-30?) is needed, it should be more efficient to spam from multiple cities at once, so as not to incur ongoing costs.

Yes. Fish has to build one accuracy treb for bombardment (separate from 8 trebs above). Other cities close to Mao need to produce 3 no-promotion trebs. FOr example.

 

[babybluepants]

Somewhere in that mess of posts last night, I asked about the Bahamas settler. Where is it supposed to go?

 

[babybluepants]

@LC,

Ok. You're right. I do too much cannon warfare, so I instinctively assume a much higher siege survival rate.

I still think that if I was playing this myself, I'd go with something like 10-12 CR and 3-4 Combat-Cover maces. Maybe that's too conservative, but there's a valid point. If we're assuming 20 LB kills, we have to assume 1-2 freak-odds losses. There's also the occasional mace getting huge damage, and healing 2-3t slower than the rest. One of the existing maces is Medic, which shouldn't do any fighting below 99% and should stay with the main stack. After we capture Shanghai, we'll have to leave a unit behind in case there's 1-2 units coming up from Nanjing. Then there's the LB's and archers we should be able to pick off in the open (not great odds with CR units). If we go with the current 7, it's easy to imagine a situation where we're down to only 2-3 that are able to hit a city on a given turn. Just doesn't seem like enough.

If there's no convincing you guys, I'll drop it. The plan is very tight, IMO, but is workable.

As for secondary production, I simply don't get why we'd want unpromoted units at this stage. Barracks is 50% less than the cost of a single unit. It almost pays as soon as we build a couple of units in a city. Unfortunately, we don't have a good production site, with the exception of GP Farm (which should obviously start running mass specs soon). Bahamas could maybe build units. I don't see any reason to settle the remaining islands anytime soon. We'll have 20 cities pre-Communism as it is. We could be building a unit every 5t there, and ferrying with a single galley. Paris could potentially get barracks, as well. I realize it's currently on a galley-trireme kick, though.

Edit:
I'm figuring that when we capture Washington, we'll get at the very least 100 base commerce per turn from GLH, btw. I think we need to do that as soon as possible after China.