Multiplayer Replay Analysis

Share and discuss strategies for playing the game, and get help and tips from other players.

Moderators: Forum Moderators, Developers

Post Reply
User avatar
beetlenaut
Developer
Posts: 2347
Joined: December 8th, 2007, 3:21 am
Location: Washington State
Contact:

Re: Multiplayer Replay Analysis

Post by beetlenaut » February 3rd, 2014, 12:42 am

Instructions written by committee tend to get longer and longer, and therefore less likely to be read. We should be careful about that. The same list can be written with less than half the words, and I don't think it changes the meaning:
Spoiler:
I also removed all but one of the "don'ts" so it sounds more positive.
Campaigns: Dead Water,
The Founding of Borstep,
Secrets of the Ancients,
and WML Guide

User avatar
HaChol
Posts: 50
Joined: September 27th, 2013, 12:48 am

Re: Multiplayer Replay Analysis

Post by HaChol » February 3rd, 2014, 1:23 am

thumbs up !
The fruit of a righteous man is the tree of life, and the wise man acquires souls.
And I said, I will perish with my nest, and I will multiply days as the chol.

Velensk
Multiplayer Contributor
Posts: 3983
Joined: January 24th, 2007, 12:56 am

Re: Multiplayer Replay Analysis

Post by Velensk » February 4th, 2014, 12:00 pm

I agree about the commitee effect.

I think we have essentially all the ideas we need to get the thread launched all that's left is to sort out specifics which in many instances will come down to personal preference.

I for example can see the point of avoiding "don't"s but would be willing to trade it in a few cases to be more specific. I kind of favor Turuks last list of all the one's listed. That said, Beetlenauts list would come in second because I also do value brevity and a 'positive' phrasing.

It may be good to just put it to vote and give it a couple weeks, maybe with a set time limit to give it a definative end.
"There are two kinds of old men in the world. The kind who didn't go to war and who say that they should have lived fast died young and left a handsome corpse and the old men who did go to war and who say that there is no such thing as a handsome corpse."

abhijit
Posts: 96
Joined: October 4th, 2011, 1:27 am
Contact:

Re: Multiplayer Replay Analysis

Post by abhijit » February 20th, 2014, 3:48 pm

I welcome the idea of such a thread, very nice initiative.

I post two replays of my recent games.
I only put my thoughts on my overall ( think initial :) ) strategy to allow for original thought while
doing a turn by turn analysis/criticism.

1. SilverHead Crossing - Knalgans (p1) vs elves.
------------------------------------------------
Some background:
Things i would like help with:
------------------------------

1. What is a good initial start on this map as p1 knalgans?
2. Mass elf fighters seemed to work really well. Any insights on this strategy and its flaws?
3. It seems I didn't use the gryphons really effectively. I would like to know especially how and around what turn
could I have used their mobility better?
4. Generally if the initial plan fails, how do you guys rebuild?
5. Any other suggestions?
2p_-_Silverhead_Crossing_replay.gz
Game 1. Compatible with wesnoth 1.10 stable
(31.95 KiB) Downloaded 311 times
2. Caves of the Basilisk - Undead (p1) vs Knalgans
----------------------------------------------------
Background:
Things I would like help with:
------------------------------
1. Undead seem to need more careful econ. management compared to other factions. Any suggestions on that?
2. There are some bad tactical decisions during combat in this game. I would like your inputs in case I missed anything
myself :)
3. I really cringed at the way this game ended, somehow i consider it was an ok move to make, can we count this as a
case of bad rng or is it a serious mistake? This final turn could have been played better, may be someone would like
to take the trouble to play it and put up the replay? :)
2p_-_Caves_Of_The_Basilisk_replay.gz
Game 2. Wesnoth 1.10 stable
(39.23 KiB) Downloaded 268 times
General questions:

1. How do you calculate a reliable kill?
2. Some ideas on counting money? Is instinct ( the way i go for it ) enough?

Hopefully I have made this post acc to the thread rules but please point out things that weren't required/undesirable.
Winner of The Alternate Frontier

Velensk
Multiplayer Contributor
Posts: 3983
Joined: January 24th, 2007, 12:56 am

Re: Multiplayer Replay Analysis

Post by Velensk » February 20th, 2014, 6:55 pm

I’ll reply to your replays in the order you posted them and discuss the questions you ask after going through the replay. I’m going to be a bit more nitpicky here than I would normally because you demonstrate a better understanding of the basics than many who post replays.

Turn 1: Any time you commit to two griffons you are likely committing to early offense. That being the case I’d think you’d want to support it with more aggressive units like thieves. Other than their village stealing ability grabbing two footpads cuts into the threat of your offense. The fighter isn’t bad but he’s on the slow side.

Turn 2/3: No Comment

Turn 4: At this point I think the first issue shows itself, this offense would be a lot more scary if that footpad was a thief instead.

Turn 5: This is a very small finicky detail but the kind of detail that can make a huge difference, if you’d tried the attack with the resilliant griffon and, when discovering that it wouldn’t work, would have left the quick griffon on the mushroom patch, you would still have the ability to threaten his central village. It probably would not have been a good idea to actually steal it but it’s the kind of steal that’s hard for an opponent to anticipate and the flexibility of the position is always a good thing. (Also naturally, it’s less risky to attack with the resilient griffon)

Rarely against elves will you need more than one ulfserker before the front. The purpose of the ulfserker is mostly to limit your opponents good options and/or to give you a way to break through a tough position on limited hexes to attack from. It’s especially the case here that you likely won’t have enough work for them with that many elvish fighters already on the field.

Turn 6: Personal preference but I think getting a guard for the south was unnecessary. Elven fighters may be good for holding you off but they are not a particularly strong offensive force (and neither are scouts) that would require a guardsman to hold off. I’d favor either a thunderer or a fighter for that kind of defense as they can not only hold the line but threaten a solid counter attack.

Turn 7: I don’t think that there was any need to put the northern fighter that far forward. The thing about woses is that until it’s time to kill them the scratches they take from retaliation don’t matter much.

Turn 8: I feel that the southern defense could have been much better managed. You had a very good opportunity to kill off three units in one turn and screen your ulfserkers from the counter attack, and even if you’d missed the most ideal result you still could have reliably killed off two enemies and shielded the ulfserkers. There are very few enemies that are worth tradeing 1vs1 with ulfserkers but ulfserkers don’t have to be one short deals if you can shield them after they make a kill.

I also feel that it was a mistake to send your leader there. You have that situation more than contained; in order to keep attacking there he is going to have to commit to a bad situation, in the meanwhile in the center and in the north he is making genuinely threatening attacks where a steelclad and whatever reinforcements he can bring could make a huge difference. If he can secure the northern villages particuarly, it’ll be a pain to take them back.

After watching his turn: lucky ulfserker there.

Turn 9: No comment.

Turn 10: Mexican Standoff. Unfortunately time is more on his side right now. This may be one of the rare cases where it’s good to leave an ulfserker in range of an enemey melee unit. A strong ulf is actually a pretty even match for an elf fighter and on that terrain you would have had equal defense, and this way he would have to defend his leader it he puts her on the keep.

Turn 11: No comment

Turn 12: I think you saw your own mistakes there.

Turn 13: I find the griffon recruit to be an odd choice as I would have favored something more durable but it has it’s advantages.

Turn 14: At this point, unless you can make something happen he can afford to just sit there and let his advantages lever him a victory.

I watched the rest of the replay and although there were small things I could point out, at this point it was essentially over as you never really had the resources to turn things around. You could crush him in one area but only at the expense of another and some wear and tare to your army. The rest of the game was essentially that taking it’s course.

As for your questions:
1: The thing I like about this map is that there really are a ton of approaches to it (most Wesnoth maps essentially dictate that at the start one needs to focus on controlling certain critical areas). I generally like to get a griffon and a fighter for control of the center and some outlaws to grab the edges. As I mentioned at the start, if I was going for double griffon play I’d probably support it with a lot of thieves.
2: Elf fighters are the most efficient unit the rebels have in terms of hp/damage to cost. They are also one of the only two units in the faction not vulnerable to ulfserkers (and the other, woses, is pretty slow). It’s only natural that you’ll see a lot of them regardless of what style your opponent is playing. I wouldn’t even consider it a strategy unless he was lacking any kind of auxiliaries (which wasn’t the case). Fighting them is a lot like fighting pretty much any elf unit and they suffer from the same problems. They lose to dwarves in a slug-out in the open, they don’t do an incredible amount of damage under any condition (though strong ones will hurt your exposed ranged units and elusives quite a bit). They can be overcome by being more efficient (via terrain use, special abilities like backstab, ect) or by bringing more power to the field.
3: I think you just needed to gear up for offense better. Mobility doesn’t help as much when you’re playing defensively, though griffons aren't bad on defensive against elves either as a counter attacking unit.
4: Depends what the enemy is doing. I just try to remember that any defense should be building towards an offense.

Next game:

Turn 1: No Comment. I would comment on his build but that isn’t the point of this.

Turn 2/3: No Comment.

Turn 4: I think it might have been better to send the ghost to the right. I can see why the two footpads and fighter would seem scary but I think that in general they still wouldn’t be able to kill the skeleton archer in the village quickly just efficiently. The ghost can trap these units but it can’t really hurt them all that well. On the other hand, it could be very useful for controlling footpads or holding a village against knalgans.

Turn 5: No Comment

Turn 6: I think you could have held the village at least for a turn. If you’d used the bat to block the thief you could also have reliably retreated it. And if he’d tried to chase you off that would delay any offense.

Turn 7: I’m perplexed by the way in which you abandoned the village. I probably wouldn’t have abandoned it in the first place (my reasoning being that as the fighter couldn’t reach it, the footpads wouldn’t do enough damage to really hurt it all that much. They could trap it but even at day adepts can destroy footpads that break formation. The ulf would probably claim at least one but as soon as the ulf is tagged everything there isn’t in a great position to stick around) but assuming the decision to abandon it has been made I would think that you wouldn’t want to put the skeleton where the enemy still has so much easy access to it.

After his turn: Well I guess it didn’t matter.

Turn 8: On the left, I don’t think you needed to abandon that village. Sure your enemy had 4 units that could target that but those units are two footpads, a poacher, and a fighter and it’s day. The fighter is the only one that can really hurt you and if it attacks it will be poisoned (the rest can be poisoned on your turn). Poison will hurt them a lot. The alternative of giving up the village means that you’ll have to reclaim it which will be rough as the rest of your army is a long way away. On the right: There is frequently a purpose in moving your units next to the enemy without attacking for position advantages but in this case I think you’re just giving that fighter a chance to smack you from cover.

Turn 9/10/11/12/13: No Comment

Turn 14: I think you’d have gotten a bit more bang for your effort trying to poison the footpad than the trapped poacher.

Turn 15: No Comment

Turn 16: Style comment. I think you might be getting too many ghosts. Ghosts are really good against knalgans but they’re good because they’re mobile and don’t die. They don’t really give you the ability to kill or threaten anything except uflserkers.

Turn 17/18/19/20/21: No Comment

Turn 22: I think that even with your leader in the mix, you may be offering him a counter attack that may be just a bit too good. A berserker actually takes down ghosts pretty well and if he can kill the ghost and the skeleton and cover the berserker again then even with nightime it could be a pretty rough situation for you.

Turn 23/24/25/26: No Comment

Turn 27: That was pretty brutal. I don’t blame you for not seeing it coming.

Turn 28: I don’t think his position was as bad as he indicated but I do agree you had the advantage.

1: I don’t think undead need to be especially econ careful. To a greater extent than any faction they are ToD dependent and have access to magic which means that they can force just about anybody off their villages come night. I think rather, undead are initiative susceptible as they are very ToD dependent and lack mobility making it hard to take initiative back from anybody who takes it from them. You did a pretty good job this game, maintaining it on the right flank for most of the game.
That said, undead do have the best econ-cheese plays of any faction. I’ve seen people successfully pull off undead play going with ‘no-upkeep’ armies comprised of bats, WC, and as many adepts as they have villages. If you’re able to maneuver around the inherent inefficiency of your level 0s you can keep a higher income for longer than your opponent.
2: I think your opponent made more tatical blunders than you did. The only major one I observed was you over-extending yourself on the right ~turn 12.
3: I don’t think it was either a mistake or bad RNG. Pretty much everything in Wesnoth is a risk and a trade-off. In this case, you were taking a risk of immediately losing the game for the chance to have a better position if that didn’t happen. You had better odds of surviving than not however, it’s not as though if you did not take that risk you would have certainly lost.
-In wesnoth it is frequently better to take risks immediately than to avoid them because they are risks and be put in a situation later when your odds are worse. For example, frequently I find that as undead it is worth forcing an engagement with drakes early even at great risk because I know that if the game drags out I’ll be up against an ever growing force of fighters and burners that I won’t be able to chase down at night which will be able to slowly pick me apart.

General questions.
1: I calculate how badly I need it vs how much it will hurt to try it. There are sometimes when the odds of getting a kill aren't very good, but if there is no cost (because of access to retreat or other factors) then it’s as reliable as it needs to be. The cost is similar to what economists call ‘an opportunity cost’ as in, it’s not a measure of money or resources but rather it is whatever choice of action is the most appealing that could be done otherwise.
Using a simple example: if you’re in the middle of a losing battle, at the wrong time of day but you have a 5% chance to kill the leader (you need 4 hits from an elven archer or something), you are measuring that 5% chance to kill the enemy leader vs your odds of winning the battle/game through any other means you can devise. The 5% frequently wins out. Now, if you do manage to win, that doesn’t mean you didn’t get lucky and it doesn’t mean that your opponent made a mistake in placing him there. It could very well be that the 5% risk of losing he was taking is more than compensated for by how much using his leader in that fashion helped his odds of winning the battle (at which point his decision would be odds of winning the battle/game when using leader in that way [~95%] vs his odds of winning his battle without using his leader that way [maybe only 90% because you’d have one more powerful unit that would make a lot of difference].
-Sometimes you just lose out despite favorable odds. There is no real thing as truly reliable except ulf vs adept. Generally things aren't all that simple as you cannot abstract an entire battle as being hopeless if you cannot kill the enemy leader but that’s how I do it.
2: For money I just go by advantage. You can use the status screen to tell you how much you’ve killed/lost. Each village amounts to roughly 6 gold per turn (+2 base +1 support for you, -2base -1 support for your enemy). Leaving aside banking/lvl 0 strats, lvled up units, and whatever advantage/disadvantage was gained by grabbing villages earlier you can use the village value and the difference in gold killed to figure out who has more gold on the field and about how much.

I don’t see how it makes a big difference though. Instinct works just fine.
"There are two kinds of old men in the world. The kind who didn't go to war and who say that they should have lived fast died young and left a handsome corpse and the old men who did go to war and who say that there is no such thing as a handsome corpse."

abhijit
Posts: 96
Joined: October 4th, 2011, 1:27 am
Contact:

Re: Multiplayer Replay Analysis

Post by abhijit » February 21st, 2014, 1:35 pm

Thanks for the thorough analysis, Velensk. Some very useful information in there.

The idea behind the guardsman was that since i saw so many fighters, I wanted to plant him on one of his south vills, and force him to get mages/woses. And the steelclad going south was to intimidate him into not attacking so that i could bring my ulfs at his vills with full hp on them, which unfortunately didn't matter to him. But, you are spot on in saying that it left the north flank vulnerable with little to no support and I under-estimated the threat there.

Thanks for the help,
Abhijit
Winner of The Alternate Frontier

abhijit
Posts: 96
Joined: October 4th, 2011, 1:27 am
Contact:

Re: Multiplayer Replay Analysis

Post by abhijit » February 22nd, 2014, 5:45 pm

Hi, today I present a replay from a ladder match b/w me and Cremember.

Weldyn Channel - Undead( p1 ) vs Cremember as Elves
----------------------------------------------------------------
Strategy:
Things I would like to know:
----------------------------

1. Where in the course of the game did I let it slip away?


Important:
------------
Around turn 8, there was a disconnect, and so I couldn't see where the wose went, and it ended up doing a 2/2 on my adept. I would like
that it be ignored, unless it had a significant impact on the game, but it would be very interesting to get an analysis of
the way I played the next turn. What could have been done better?


regards,

abhijit
Attachments
2p_-_Weldyn_Channel_replay.gz
Cremember vs abhijit ( wesnoth 1.10 compatible )
(26.36 KiB) Downloaded 277 times
Winner of The Alternate Frontier

Velensk
Multiplayer Contributor
Posts: 3983
Joined: January 24th, 2007, 12:56 am

Re: Multiplayer Replay Analysis

Post by Velensk » February 22nd, 2014, 7:00 pm

Turn1-6: No Comment.

Turn 7: I like your play on this turn but I feel like it'll be inefficient except as a distraction from elsewhere. You're investing pretty heavily in your scout units which will cut into your offensive power later unless you can make something happen. I think when things get intense you'll find yourself more wanting hitters.

Turn 8: I think that it would be worth letting him use the mage on the ghost to make him have to use the mage on the ghost. You're just biding for time and money there.

I think that even had you known the wose was there, it was a decent choice to push forward though you are running into the thing I was worried about: Not having the killing power to topple the wose which allows one unit to delay your entire push effectively.

Turn 9: I think you might as well have started working the wose down with your ranged attacks and done what you could to shield adepts from elves with the skeletons. It's not like you're going to be able to leave it alone (it'll be behind you blocking your retreat and doing tons of damage) so you might as well start trying to kill it.

Turn 10: Might have been better to let the ghost hold the village (something ghosts are good at) and have the skeleton attack something skeletons do better).

Turn 11: No Comment

Turn 12: It worked out for you the way you did it but if you had softened the mage with the bat you could have significantly reduced the risk of losing the skeleton to the mages counter attack.

Turn 13: In a situation like the one you were in, it's frequently best to emphasis kills over efficient unit use. The reason being that kills turn into less damage he can inflict immediately which snowballs rapidly (also kills can't be healed). This is especially so for undead because all undead units (except ghosts) are either extremely melee or extremely range specialized and so almost dead units can reliably attack at full potential (note this is less of a problem for evles because all their units fight at both ranges so to them a skeleton and an adept with one health left are much less of a threat). In that case I believe it would be more worthwhile to target the fighter in the open with the adept because if you can get even one hit you can finish off one of your most threatening foes with a ghosts whereas with your admittedly bad luck you failed to kill the other fighter and it will now be able to slice up an adept as as though it were at full health.

Turn 14: Your moves on the right this turn were probably the best available but that also kind of shows how bad a situation you're in. On the left you could have moved your bat more inward to threaten the far village. It's fairly safe to leave it in the range of the fighter (if he attacks you he leaves the village open and has to hit thrice to kill you).

Turn 15: If you were going to attack with the skeleton, at least leave it on the mountain, it's not like he's going to let you escape. Also, no reason to retreat the skeleton on the left. In fact, as long as you can avoid being trapped there's no reason to not move it forward. One fighter cannot defend 4 villages at once without killing you and he cannot kill you quickly by himself. If nothing else, threatening that area will force him to divide his attention.

Turn 16/17/18: No Comment

Turn 19: The skeletons flank could have been screened to avoid being pinned (of course that would leave another target open to attack. Alternatively you could have pulled back further and then pulled back again the next turn (taking a mauling in the process likely) in preparation for one big counter attack. Either option I think would give you better odds though neither is really good.

Turn 20: No Comment

Turn 21: You got really good luck on his turn. Enough to turn a game normally.

Turn 22: No Comment

Turn 23: Game over

Overall: The critical point here was the battle over on the right from turn 8 to turn 13. At first you didn't have the stuff there to even hurt the wose much at day let alone kill it (though if he hadn't gotten one of your adepts you'd have been in better shape, that did make a huge difference) and then you were embattled in a nasty melee with elven fighters against adepts. Different choices and luck throughout could have made a major difference but I'd put a lot of it down to you not getting kills, or not getting kills without trading. I do have to say though, your opponent played a fine match and although your opening gambit gave you a good advantage I don't think you'd have had an easy time in any case. Other small things could have been done to force your enemy to deal with more but these are relatively small details.
To Emphasise: although better harrassing could have helped you, you spent quite a bit of money on bats and a couple ghosts, the reason that the wose was able to attack you so freely was because you had one less adept due to that, and one less adept because he killed one. If you'd had one more adept because of your choice, and one more adept due to better fortune (and the first adept may have been enough to convince him to not attack) that offense would have been much scarier and his best option may have been to retreat and surrender to you the village for free.
"There are two kinds of old men in the world. The kind who didn't go to war and who say that they should have lived fast died young and left a handsome corpse and the old men who did go to war and who say that there is no such thing as a handsome corpse."

abhijit
Posts: 96
Joined: October 4th, 2011, 1:27 am
Contact:

Re: Multiplayer Replay Analysis

Post by abhijit » February 22nd, 2014, 7:44 pm

Thanks for the reply, Velensk. Again, nice inputs.

I can see what you mean by lack of killing power on the right. I carried around an adept on the left flank doing nothing with it, for almost 10 turns. Probably, that's what a 'passive' unit is? :)

About retreating the skelli on t15, I was trying to trick him into thinking I have nothing there, and make the steal with bat. The plan almost worked... until the stupid bat ate 3 hits from the fighter and died and I got nothing. :oops:

In retrospect, yeah, the bats didn't pay off.
Lesson noted:- Trickery is only the means, battle is still won by 'killing' the enemy.

regards,
abhijit
Winner of The Alternate Frontier

User avatar
max_torch
Posts: 329
Joined: July 31st, 2011, 5:54 pm
Location: Cavite, Philippines

Re: Multiplayer Replay Analysis

Post by max_torch » April 14th, 2014, 7:33 am

I don't know how to penetrate enemy defense. I know his leader is very vulnerable to assassination but I cannot reach. Or did i miss a good leader kill opportunity?
2p_—_Caves_of_the_Basilisk_replay.gz
(27.66 KiB) Downloaded 257 times
Oh by the way this is a mirror match of knalgans

Velensk
Multiplayer Contributor
Posts: 3983
Joined: January 24th, 2007, 12:56 am

Re: Multiplayer Replay Analysis

Post by Velensk » April 15th, 2014, 10:58 am

In general I consider a knalgan mirror to be worthy of immediate restart because all it takes to create a stalemate situation is for one player to go dwarf heavy and for nobody to mount an effective attack early.

Turn 1: Village grabbing scheme could have been better put together.

Turn 2/3/4: No Comment.

Turn 5: I'm curious why you didn't capture he village he abandoned, it would have helped alleviate some of the massive gold advantage he's built over you.

Turn 6/7/8: No Comment.

Turn 9: I think a fourth griffon might be a bit excessive. Remember, each griffon although a great effectiveness increaser is also a loss to efficiency. I'm also not sure what you're doing on the right flank as even though you hit you don't seem to have anything in place to take advantage of it either now or next turn.

Turn 10: No Comment.

Turn 11: I don't think you actually needed to surrender on the left. He may have an impressive number of chaotic units but what he doesn't have is any way to do impressive amounts of damage to a dwarf on a village (once you've blocked the footpad flank with griffon). With your more mobile forces you could threaten a deadly counter attack to destroy him if he commits to it and as long as thieves don't get to use their backstab, dwarves toughness with terrain advantage will be king.

At this point: I watched the rest of the replay and I honestly don't have a lot to say about it in the format I typically do (mostly it seemed to me that it followed the natural course of the advantages he's gathered), so instead I'll address your question about assassination directly.

Your opponent was actually doing pretty good at making it so that his leader was never really at risk of being picked off. One griffon cannot outright kill a thief and two griffons odds of doing so are slim. As such, there was never a moment where you could have taken a do or die risk to win the game. However, what you could have done is chosen to prioritize weakening his leader so he'd have to either keep it back more or risk assassination but that was one tactical choice out of many options which may have been better.
"There are two kinds of old men in the world. The kind who didn't go to war and who say that they should have lived fast died young and left a handsome corpse and the old men who did go to war and who say that there is no such thing as a handsome corpse."

User avatar
Practice
Posts: 12
Joined: April 20th, 2011, 8:15 pm

Re: Multiplayer Replay Analysis

Post by Practice » June 14th, 2014, 11:49 pm

Hello,
So i started playing ladder after the 2 year break and I'm losing my nerves already.
I was playing Knaglan vs UD on Basilisk.

My enemy didn't make a offensive move through the whole game, which ended in almost 2h of stalmate. Mass ghouls, bats, wc and some adepts.
I understand I made some mistakes, but the other guy wasn't almost playing at all which is just annoying.
Is there any tactic I don't know, how to break campy opponent or is it just a nature of Wesnoth/matchup? That was so anti-fun, I'm not sure If I want to still play ladder in the future.

The game ended because I didn't have more time, energy and patience to continue this marathon of boredom. I simply threw everything i got at him and it didnt work.
Attachments
ZZZZZfest.gz
(30.91 KiB) Downloaded 251 times

Velensk
Multiplayer Contributor
Posts: 3983
Joined: January 24th, 2007, 12:56 am

Re: Multiplayer Replay Analysis

Post by Velensk » June 15th, 2014, 11:51 am

Preface: When I read your description, I was thinking the match would be what I'd consider a stalemate (all flanks packed with lines of units, too many defensive units not enough offensive, little movement on both sides, neither able to capture enemy villages). This game was nothing of the sort. The replay was only 22 turns long and was effectively over on the forth day. If this game took two hours, it was only because someone was taking their with their turns not because it was in any danger of becoming a stalemate. Now that can be plenty frustrating, but don't think that just because your enemy doesn't attack you that you're running into a passive foe that will inevitably dissolve into a stalemate. About the closest I'd say about the match was that it wasn't a match determined by the initial gambit, it was still relatively eventful.

I'm going to give a turn by turn here, but the overall is the part that's really going to matter for you.

Turn 1/2/3: No Comment

Turn 4: Maybe would have wanted to nudge the righter on the left up a hex. Your defense doesn't matter if he isn't going to attack you and if he does attack you it works in your favor anyway so you might as well have the fighter in reach to threaten the village.

Turn 5/6/7/8: No Comment

Turn 9: Although the ghost is a good pick off, you could end up losing far more there than it was worth. You cannot save the ulfserker from the bats so don't lose the griffon trying to save the ulfserker.

Turn 10: No Comment

Turn 11: You're focusing all your reinforcements on the front that's getting crowded. The right is guarded only by bats. The army he has isn't going to break you offensively very easily so you don't need it for defense. I also find the choice of yet another griffon to be an interesting one. It can help you get a flank but it's not efficient as a fighter by itself. I think you'd be able to get a lot more threat power out of another ulfserker or even a thunderer (which are decent against ghouls, corpses, and adepts).

Turn 12/13: No Comment

Turn 14: I generally like to have the ulfserker commit first as losing him to the ghoul in similar situations means that you know how hard you have to commit with other units and this way he cannot be zombieified.

Turn 15: In this situation, you might want to consider a poacher or thunderer over a footpad. Footpad wreck skeletons and can poke at ghouls but right now what you need is something that can block dark adepts and kill ghosts.

Turn 16: No Comment

Turn 17: You are getting far too reckless here. Your enemy has enough firepower on the left flank to destroy whatever you keep there, in the meanwhile you have ways of making things come out in your favor elsewhere. You can lose several units in a probably futile attempt to save one of yours and kill maybe a couple of his or you can cut your loses and settle for a good board position.

Turn 18: No Comment

Turn 19: One ghost at day, a strong ulfserker might take but probably not two and certainly not three. He has a ghost that's about to level up which is something that if at all possible you want to prevent (gryphons and thunderers being your best bet for that).

Turn 20: Guess it worked out for you. Were I him, I would have gladly sacrificed a ghost to get a wraith in this position.

Turn 21: No Comment

Turn 22: Replay Ends

Overall:
Momentum: You talked about how your enemy never attacked you not even at night. This is true, it's also primarily because there was never a good opportunity for him to attack without it being suicide. He was on the back foot pretty much the entire game (and would have been even more so if you hadn't kept being careless with your griffons). When you are able to take an enemies villages regularly that generally means the momentum is in your favor. If you had been more patient you would have been able to keep rocking back and forth taking that village, squeezing him for income until you would have been able to dominate him on both flanks or crush him when the engagement happens. Now, this isn't a fast and furious way to win (and if your opponent plays slowly I can see why it wouldn't be appealing) but it is reliable. Now, you lost the momentum when you assaulted him and it didn't work out tactically. There are a few reasons why it didn't work out tactically but until that point you were in no danger of losing strategically and this despite having taken considerably more attrition that he had up to till that point (two griffons and an ulfserker and some village stealing of his own).

I'll take apart the push you made here just because that was the turning point in the game:
The Situation:
-It's the bottom of the morning turn. You have 3 footpads, 3 fighters, a thief, a ulfserker, and a griffon with a thief on the swing and another ulfserker a couple turns behind the front and a fighter as distant reinforcement.
-Your enemy has 4 adepts, 3 ghouls , 4 WCs a vampire bat, a blood bat, a ghost, and his leader, and whatever reinforcements he recruits and has a decent phalanx weakened by the fact that much of it is lvl0. He has a ghost and potentially a ghoul on the swing.

Options I see for you:
-Assault: You're actually running into a ton more firepower than you have however it is day and you have much more durability. If you can manage to kill enough enemies you can make it worthwhile on village income (though pretty much no chance of really breaking him unless he exposes his leader). In this case, your priority is to destroy his adepts. Aside his leader, they're the only thing he has that can kill you with any kind of speed.
-Attrition if you can maintain a solid formation with your mobile pieces on you could hit him for the morning turn and then retreat as the afternoon wears on. This is unlikely to net you any pieces you want to kill and will cost you a ton in poison damage but you could pick off some walking corpses and possibly damage his critical units enough that he won't be able to push you too hard as you meet up with your reinforcements and get further away from his.
-Hold: This is the option I would have picked. Just as it would be hard for you to attack him, it'd be stupid for him to attack you. If you can just not be attackable for a turn or two you can keep the gold retreat as night is coming and then the minute ToD starts shifting back in your favor push back to the village again with a much scarier onslaught (and by then, maybe you'll have picked up the units you need to break him).

Unit composition: The units you have aren't horrible but they aren't really ideal. The footpads you picked up near the beginning are normally very good against undead but much less useful against the unit composition he has picked (I notice you haunt gotten any since the start which is a good read on your part). The best thing you have to take down ghosts are your ulfserker (who'll nearly inevitably die in the assault) the griffon and the thief if it can get a backstab [unlikely]). This means that each ghost he has will be a strong point that you probably won't be able to break which he can use to protect the units you want to kill and poke at you with a ranged attack safely. As I mentioned in the Turn by Turn, some thunderers or poachers would help a lot in that department, they could also help attack ghouls and block adepts (who're his main damage dealer).

Tactics:
-Morning: Like I said in the Turn by Turn, I find it best to let the ulfserker go first. If it succeeds you've cleared the village without blocking any hexes that might prevent you from building your support base there and if it fails you'll know how many hp it failed by and be able to choose your finisher appropriately (and this way you don't have to worry about zombies) After that point, I think you did the best you could with what you had but with an extra hexside to push from you might have been able to capture another hex next to the village and make it that much harder for him to mass the firepower to push you off.
-Afternoon: Might have been a good idea to try to finish the dwarf zombie off with a footpad first, this way you'd be committing far less damage to the task and could have reserved another fighter to hack into something juicer. Other than that, again good for what you had but it's a hairy situation.
-Dusk: Good for what you had. I'd have gone for the adept with the ulfserker but I don't think that the choice is necessarily better, it's just a matter of tactical priorities.
—At this point the fight was already decided.

Conclusion: You might not like it that there wasn't a good set up for either of you to break the enemies army right away but you weren't in a stalemate and your opponent was playing pretty well given the situation he was in. If you'd been patient you could have used the momentum you had to force him to do something risky but instead you tried a gambit of your own and it didn't work out. If you wanted a quicker solution there was an entire front you could have pressured to force him to spread out a bit. Still, you'd been playing pretty well up till that point but you might want to watch to see what kinds of decisions you make on reflex (for example the griffon the block the ulfserker when the bats can just fly around the other way).
"There are two kinds of old men in the world. The kind who didn't go to war and who say that they should have lived fast died young and left a handsome corpse and the old men who did go to war and who say that there is no such thing as a handsome corpse."

User avatar
Practice
Posts: 12
Joined: April 20th, 2011, 8:15 pm

Re: Multiplayer Replay Analysis

Post by Practice » June 15th, 2014, 9:59 pm

Thank you.
I see a lot of things I can improve on and I really appreciate your analysis. The situation wasn't as hopeless as it seemed to me during the game.

Velensk
Multiplayer Contributor
Posts: 3983
Joined: January 24th, 2007, 12:56 am

Re: Multiplayer Replay Analysis

Post by Velensk » June 16th, 2014, 12:59 am

As a note: I've updated the origional post to contain Beetlenauts version of the posting guidelines and I'd say this thread is ready to be sticked now like it'd been origionally intended.
"There are two kinds of old men in the world. The kind who didn't go to war and who say that they should have lived fast died young and left a handsome corpse and the old men who did go to war and who say that there is no such thing as a handsome corpse."

Post Reply