Turn 136 - Zululand
still kinda in zombie mode from the intense late-night seder yesterday..... but the turn waits for no religious holiday
so the gav/superdeath war has ended.... almost completely inconclusively despite the significant amount of material exchanged, other than gav gaining and keeping possession of nutty bars. i mean, settling up aggressively around it as gav did and then attacking to secure it has worked out pretty marvelously in terms of landgrab..... and for SD it, i suppose, makes sense to drop out of this stalemated war and immediately pounce on naufragar now that we're obviously for real about helping them. wonder what the odds of that happening as soon as SD's stacks make it to the other side of their territory are...
drey did NOT return the cow/cow i sent at the end of last turn, which makes me nervous..... but actually i'm not sure they've had the turn yet so i'll be chill about it for today at least. going to send ginger an extension request as well which i would be surprised if they accept, but i mean, i've been surprised before. some combination of the those two outcomes will probably determine if i feel safe enough to (temporarily?) drop slavery for serfdom
and re GT, their reinforcing stack turned around and no attacks were made against our units. lalibela has 8 units inside, divided evenly between crossbow and pike, after another double-whip, and it looks strongly like this will be the turn to attack it
williams, as usual, your prediction has immediately come true.... GT's stack is on the move. guess we have a date with it soon

looks like their plan is to recapture lalibela and prevent us from making gains elsewhere... they can reach the city on their turn 139. most likely i'll leave just longbows on defense there and try to get them to a) take losses to crack those and then b) expose more of their stack to collateral than is advisable. if our main forces tangle directly i think we're in the regime where whoever strikes first with their cats will win
anyone know off the top of their head how engineering affects road movement in neutral territory? i remember reading about this in a recent PB, and the answer being NOT what you would expect based on the behavior of similar techs like construction, but i can't remember the exact citation as of now D: unfortunately it looks like it's about to become very relevant to our situation... we may have reached the limits of what we can take with direct force now that their stack is on the way (well, MAYbe we JUST barely have time for a sneaky capital-gondar fork still but would have to raze if so) and are entering the positional/attritive phase of the war, where we might have to rely on nauf making gains on their end and our combined 3-1 edge in cities to grind them down
yeah, aksum-gondar fork looks like it's not going to happen.... if the stack moved 3 tiles eastward as we'd expect, it's in position to JUST make it into gondar next turn, and then aksum should have at least the 4 defenders we blocked from lalibela last turn in theory if it is just that in there we could try to trade a huge number of HAs for a raze on the capital, but a) i don't really think that's going to be the case and b) even if they have just 1-2 extra pikes in there i think we straight-up won't have enough 2-movers to crack it. so most likely i will end this turn feeling successfully deterred from that plan.... although naturally dragging the stack away from nauf's increasingly formidable-looking force should do SOMEthing to progress the war state even if we get stalled on our end
ok well, i don't think i'm going to be unconvinced about attacking the 8 trapped units in lalibela anyways... or of keeping it if we manage to take it, as the city has been pretty neutralized as a productive force from whipping at this point and it is the best defensive tile around for our (two....) longbows. althoughhh GT will get the borders back if they recapture successfully which i do NOT like (does that happen immediately on the same turn as recapture?) as it could set up a situation where they could retake the city AND hit our main stack with collateral on the same turn..... or retake the city from a position such that we can't immediately counterattack in response due to the borders cutting off our road access. ugh, might have to test this in worldbuilder....
...
yeah according to wb the borders DO pop back immediately, on the same turn as a recapture. lovely. so we're thinking seriously about a raze then
well, at least we can do the fun part in parallel with the unfun part. two accuracy cats and one unpromoted cat suffice to zero out the 40% defenses at lalibela, leaving 6 available as attackers. i am not really sure whether CR1 or barrage II is the correct promotion for our first cat that will almost certainly die in the attack (4% odds to live unpromoted).... so back to WB.... which suggests that we'll get about 1 extra total damage spread across the 5-6 collateraled units from BII, which i feel confident is less than the marginal extra damage from CR2 we'd do to the one crossbow that we'd be facing if the odds to live are as long as 4% is. i expect that to change once our cats have appreciable odds of actually surviving but for now i think we lead with barrage II (and as you can see i didn't have time to really do this analysis in any detail so if it conflicts with the conventional wisdom please lmk for next time)
barrage II cat vs full health CG1 xbow (3.5%): dies, doing 0.8 damage to both the defending xbow and 6 other units in the stack. which leaves.... one full-health unit.... so from taking a battle at 3.5% odds to win, our cat reduced the total strength of this stack by 5.6, MORE than the strength of the unit we lost?? and that was the EXPECTED outcome?? damn collateral damage is powerful :D so with one more full-health xbow we clearly have to take at least one more battle at odds like those:
barrage II cat vs full health CG1 xbow (3.5%): dies doing exactly the same amount of damage. now our cats have real, 10% odds to live on the attack and perhaps this is where i start taking CR promotions. they still get the best odds by quite a bit, but at the same time we obviously we don't want to overspend them here if we can avoid it. and yeah, taking CR1 takes our odds from 10% up to
B1 + CR1 cat vs 5.2 health CG1 xbow (37.9%): LIVES on 0.2 health!
B1 + CR1 cat vs 4.4 health CG1 xbow (46.3%): dies and gives GT a GG in the capital
B1 + CR1 cat vs 3.5 health CG1 xbow (87.0%): lives, but also on very low health
B1 + CR1 cat vs 3.3 health CG1 xbow (88.4%): lives at half health, but this attack might have been a mistake as we're now at the collateral damage limit. we are also now out of cats and need to have brought some real units, none of which get amazing odds even against the fully-collateral-ed stack.... the best we can do is
C1 elephant vs 3.3 health C2 pike (58.8%): dies but does good damage
C1 elephant vs 3.3 health C2 pike (59.8%): wins for our first actual kill lol
C1 elephant vs 3.3 health C2 pike (59.8%): wins again! there is still one more 3.3 health pike but then our horsies can clean up the rest i think
flanking 1 horse vs 3.3 health C2 pike (55.5% to live): dies but i think now we're in cleanup territory for real
i then proceeded to stupidly misclick a horse into the city before promoting LMAO (75% odds) and won, but now it's on an unintended tile and GT could kill it next turn unless we try to reachily defend it. this is a significant deal as it represents a second sentry unit if it lives..... it is on the tile 1W of the city center, where the capital garrison can reach it, and was intended to be 1N of the city where it would be safe. considering using my "one oopsie" but perhaps this is not impactful enough to warrant that
then 5 straight 99% battles that we of course win, including one to create something i foolishly forgot in my mass-promotion of impis to shock, namely a MEDIC impi who will eventually become our super-healer if we go for it
i am still unsure as i make this attack whether we should raze or capture. i think it depends on a) where the borders revert to without the city and b) how many buildings are preserved....
border situation: two lines of tiles have reverted to neutrality. not much but obviously could be critical in preserving our ability to first-strike GT's stack. preserved buildings are a lighthouse for 3-food lake tiles and our empire's only forge

so not as crazy-impactful as saving the granary would have been, but the lighthouse will be close if we gain control of the lake tiles and the forge would otherwise almost certainly never have gotten built here and represents quite a few free hammers over the course of the game...
and, in what is becoming distressingly common, i realize the most important feature of the position RIGHT as my mouse is hovering over "burn baby burn" - we have a bunch of 2-movers in GT's face right now that we have some reason to believe if would be inefficient for them to counterattack, and lalibela represents such a bottleneck in GT's road network that we can negate their engineering advantage by, um.... just by pillaging the stupid roads

ok, that's good enough - i am not certain i can make it happen, but i am going to try to find a way to hold this city. we don't need to delay the counter by THAT much to get a pretty decent-sized pile of longbows in there (and we even brought along 2 archers we can upgrade if needed). and if we can delay 4 turns we can hilariously whip PRO walls and then instantly have 75% defenses here thanks to chichen itza...... i'm not sure trying to hold it is CORRECT per se, but it's certainly favored by our traits and setup relative to baseline, and the city is so burnt-out at this point that really all we'd lose by having to evacuate it is the tiles. yeah i guess that's good enough for me
so in summary we lost, what, 3 cats and an elephant to kill 8 medieval units and capture a city? ok i think i'm starting to get why everyone considers these things op as fuck now
move up a sacrificial chariot to check the defenses in the capital:
ughhh. so looks like whichever group of our roadcutters is most forward is going to get pasted by cat collateral :| maybe both if they have combat workers in the fog though i don't really expect them to.... i am pretty confident it's worth cutting the farm road; is it worth expending what is probably 2 additional units to use a 5-unit stack to lay waste to this town and cut its road as well (and thereby probably shield the 2 plains farm cutters + misclicked sentry-promotable horse, which is why it's just 2 additional units)? we have some seriously crappy units we could use to do so and it would make a big difference in defending lalibela so i think so. but ugh i need a break.... i'll come back to this after lunch
one lunch later....
well, i bit.... and then realized that we really didn't have to do that as, unlike with fighting, units can still move after pillaging in this game
so maybe i did just blunder away 5 units towards a questionably worthwhile aim.... but we will see. most of those units were crappy things like impis that we really can't expect to get much use out of going forwards. i did add two horse instead of making the stack ALL impis, so those will be a significant loss, but the point there was to force them to either use a catapult, which will probably die, or take out the horses with crossbows, one of which would be expected to die.... so basically i traded the difference between two impis and two horse for either a crossbow or a cat in expectation, which seems... maybe not a slam-dunk, but fine at least. and we did wring a fuckton of gold out of burning down the town anyways.... enough to upgrade a longbow, so we did GAIN the difference between an archer and a longbow in a hilltop border city which is not nothing either
anyways, if we're going to hold this city, or at least put GT in a position where they can't recapture without collateral exposure, what we need most is time.... time to pile it as high as we can with protective longbows on a hill that even catapults + medieval units should find inefficient to remove. we are whipping some in all our cities within plausible range, and in conjunction with upgrading the 2 archers i brought along (which i may not do, 180 gold is PRICEY) we can have 7-8 in there at around t141, plus protective walls, plus chichen itza. we have no way of knowing how many combat workers they have in position, of course, but each turn of delay will give us a significant defensive asset here. and if we can force them to burn all their cats on a big pile of super-spammable longbows, THEN we might be in position to make progress again offensively
i decide to gamble with two workers on getting a road on this PFH... they would have to have either a TON of combat worker in the fog or a few workers and a 2-mover that can beat our horse in order to punish this. if successful, we could check out the defenses at matara next turn and rapidly redeploy up to 10 horse from lalibela to hit it the turn after if it seems feasible to do so. if not, we are well past the point where 2 workers is a significant loss i think, given that ours have been sitting around roading for almost 10 turns now lmao
similarly, i trade an impi for the scouting warrior of GT's in an attempt to draw troops out of matara or at least get them to show what kind of forces they have in position. well.... ok, actually what happened is i was CONSIDERING doing this, and, um.... moused over the combat to get the odds, and then misclicked into actually PERFORMING the combat.... soooo that's not wonderful and i don't know if i actually would have gone through with it given full consideration, but too late now lol. trackpads man
forgot to mention this hilarious development in gingerland:
obviously a double holy city is amazing for them.... but i guess it's also good for us?? they won't be able to stack it with wall street until pretty late in the game, late enough that we are hoping to have conquered them by then, and prussa is quite firmly on our side of a hypothetical ottoman split with dreylin. so maybe we will end up fixing our lack of exploitable national wonder cities by building wall street in ginger's best commerce city and the heroic epic in GT's capital
another reason to have pillaged the town: notice lalibela will claim no two-food tiles when it comes out of resistance in 4 turns.... which means we MUST build a windmill on the marked tile immediately or it will starve and then not be able to whip walls (or actually, maybe it can dry-whip them thanks to the forge?). and um.... guess how many turns we are from machinery, and how many turns we WOULD be from machinery if we were roughly one pillaged town poorer.... not that i'm quite sure where the 5 (is it 5?) workers we need to 1-turn a windmill will come from but clearly it's at least POSSIBLE to do so
edit: wait no, we'll be in a golden age so that's inoperative. guess i just did it for fun then :shrug:
sheesh, what a nasty complicated mess of a turn. not too confident i played it correctly either. but so it goes!