Intersite Game - Turn Discussion Thread

[mostly_harmless]

Tile culture analysis tells us that there is a CFC city northeast of their capital.

There seems to be a roaded forested hill 2N-NE of Indira. Looks like a good spot to move the scout to and even have one tile movement left.

mh

[Qgqqqqq]

Agree with mh.
Couldn't the axe in forbidden fruits move to support the attack if necessary.

[Lewwyn]

Just to chime in if we save the promotion that means we can attack at 96% odds and then heal whatever damage we get from that as well. SO I like the idea of saving the heal for after killing our second warrior. If we promote and take a lot of damage and don't have a promo to heal it could delay the capture.

[kjn]

Had a d'oh moment while eating a late breakfast - we could let Dantski heal next turn - in effect both increasing our odds (more HP and higher strength) while still getting 2XP if we win on T87.

The detailed stats right now:

Dantski has 96.60% chance to win, 3.40% to lose

Dantski: 4.30 (86/100), 3/2 XP
Barb warrior: 2 (100/100), +25% tile defense, fortify, and city defense, -50% vs melee, net +25€

Healing a single turn would give Dantski 4.55 in strength and 91/100, and push us to 97.2% chance to win the next battle (risk to lose shrinks from 3.43% to 2.8%).

Healing two turns would get us to 4.8 strength, 96/100, and give a 2.3% risk.

[Azza]

That could work, but if Alemanni grew I'd attack immediately.

[Sian]

i'd lean towards keep attacking, from the reasoning that we don't know when the city will grow. We should be ready for capture as soon as the city grows (or as soon afterwards) so we should first clean up so theres only 1 warrior left in the city.

At least ... thats my 2¢

[mostly_harmless]

That could work, but if Alemanni grew I'd attack immediately.

i'd lean towards keep attacking, from the reasoning that we don't know when the city will grow. We should be ready for capture as soon as the city grows (or as soon afterwards) so we should first clean up so theres only 1 warrior left in the city.

Why is it so important to capture the city right when it hits size 2? What am I missing here?
If kjn's variant increases our chances to capture the city by delaying it for one turn, then I favour that. Unless, as I said, I am missing a crucial fact with a size2 barb city.

EDIT: And I know about the auto-raze mechanic at size 1.
EDIT 2: What if we kill the second warrior now and then have to wait 3,4,5 turns for the city to grow? Would be safer to wait and heal now, no?

mh

[SevenSpirits]

Why is it so important to capture the city right when it hits size 2? What am I missing here?
If kjn's variant increases our chances to capture the city by delaying it for one turn, then I favour that. Unless, as I said, I am missing a crucial fact with a size2 barb city.

Let's see, kjn's idea involves waiting a turn to get about +0.6% chance to win the battle. If we lose the axe we have to spend 35h to replace it and let's say 5t to move the new one over there, generously +1 turn because it builds an extra warrior in the meantime. So 0.6% of 6t, that is 0.036 barb city turns. 35h * 0.6%, that's 0.2 hammers. 1h is made up in about 0.25 turns of that city's operation, so 0.2h are worth about 0.05 barb city turns. Total expected value gained by the extra odds is 0.086 turns of operation of the barb city, which is nothing in relation to the 1 lost barb city turn if we, well, delay capturing it by a turn.

So in order for the math to come out in favor of waiting a turn, we have to think there is a 91% chance the barb city will not grow by the time we can capture it at the earliest. We knew the barb city was there t81. Since it is working the 3f tile it should grow 8t after being born. At worst it will have grown t89. At best rate we would be able to conquer it t87. AFAICT the first tracking of barb pop increases in the C&D thread was t80, and none happened t81. Since it's t85, the earliest it could have been born was t78; once the turn rolls and we have to make the decision if the city hasn't grown yet, the earliest will be t79.

So we know (rather, will know next turn, if the city doesn't grow yet) with fairly high probability that the barb city was born either t79 or t80. If we can confirm that it was not born t79, then sure, waiting a turn for +0.6% odds seems good. If we just don't have the info, I think the odds it was born t79 rather than t80 are higher than 10%. So on balance we'll be better off attacking rather than waiting.


Disclaimer: values are rough estimates. And there are always those unknown unknowns.

EDIT 2: What if we kill the second warrior now and then have to wait 3,4,5 turns for the city to grow? Would be safer to wait and heal now, no?

mh

There is also the fact that we may get an equally or more useful opportunity to heal before a future attack. So I don't think that's necessarily the case. It doesn't seem like that 0.6% odds improvement is hitting a breakpoint or anything.

[kjn]

Do note that my tracking of barb pop is quite tentative (though we might know more about its validity once we capture Alemanni). We also don't know if barb cities produce anything on their first turn of existence (unless T-Hawk or someone else can cast light on that).

Since we have so many unknown variables here, and the downside of Dantski being defeated is quite large (need a new axe, need to walk an axe to the barb city et c), I think it makes sense to minimise risk.

[Dantski]

I agree with Seven about the 0.6% being fairly trivial for waiting a single turn to heal. If you want to be totally safe you have to heal to full before attacking again and that feels like you're getting to XCOM levels of paranoia about combat odds.