The Point of Goblins?

[Tonepoet]

I'm talking mostly general utility when I talk about the Goblin's purpose; the strikes certainly are nice but I generally am forced to pick between the two. Tactically speaking which unit is better relies strictly on the exact situation of course but when I see a unit with 1hp as opposed to 9hp, I can't suddenly change my mind and reinvest my gold into changing what units I already have, trading grunts for goblins or visa versa. There's nothing stopping me from having both but looking at it as a strategic viewpoint, why would I want both? What long term goals can't the grunt achieve, that the goblin can? Not many unfortunately: once you don't judge these two units based upon the damage types of their weapons, the goblins lose most positive distinction.

Ghosts vs. Bats could be said to be very similar when compared to each other too but I can see different points for both of them existing and not just in the proportionally similar price gap. A ghost has much stronger resistances than a bat and the higher base damage means more kills and more drain, so it's much more well suited to combat. A bat however has an extra movement point and a slightly better move-type, making it even faster and more nimble. This means if I needed a speedy unit for combat, I'd pick the Ghost because it's better equipped to deal with it, where as if I wanted to do stealth ops. unit for fog recon, village theft or just something to go through water quickly I'd more often pick a bat.

With the Grunt and the Goblin it's very different when you compare their stats: Each are about as fast, their attacks are more or less equivalent in damage, their resistances are the same, their terrain defense is the same, their movetype is the same and they each have no abilities. Also I've double checked: the grunt really does get 3 times the H.P. while a goblin only gets double before resistances come into play. This means for each goldpoint spent, you get 50% more H.P. and the Goblin's economy isn't looking so great anymore: He has to survive at least five turns to match this raw economic potential, which is a very, very long time for a unit who can't take a hit.

Only once I've strategically recruited my unit can its tactical specific come into play, which is where considerations like "What's my chance to kill with each unit?" or "I'll level up this unit if I use my grunt first so I'll use my goblin instead?" or even "Oh hey, if my low H.P. goblin dies first, it'll free up another attacking space so I can more feasibly kill their leader!" come into play. So the first step is to pick out the Northerner: When do I say "I've picked the goblin to do [what job] because it's better suited at it than any other unit I have in my arsenal."

How much reasons do you need to recruit a unit that only costs 9 gold? Does it have to fetch the paper in the morning?

Not necessarily: It's just that at the moment, two contenders seem to've applied for the exact same position of C.H.Y.M.A. and one of them appears to have handed me a better resume: As an employer, it's my responsibility to pick the best and most effective candidate to do all of the responsibilities I need for the job. If that means the Grunt will do the same thing as the Goblin and fetch my Sunday paper too, well, that service might be worth the extra cost if I suddenly find I need my papers fetched.

[Eskon]

"I've picked the goblin to do [what job] because it's better suited at it than any other unit I have in my arsenal."

This will lead to misleading results simply because counted individually, it isn't better suited at anything, except maybe for raw damage against cavalry. (Isn't it remarkable enough that a level 0 unit can end up beating all the level 1s of a faction in attack power, even if only against a certain opponent?)

Goblins have grunts beat when it comes to "bang for the buck". Grunts win out, if they do, because they can tank and ZoC, neither of which goblins can do. Having all your melee damage units be capable of tanking at the same time is sweet, and a perfectly viable strategic choice to make. As with everything, it has strengths and weaknesses. Having more flexible main forces is a strength. It being expensive in upkeep is a weakness.

Goblins can also most certainly live for five turns and longer, depending on the flow of the match - a slow match, which usually also has you go well over your upkeep threshold, can easily contain 10 or more turns with few battling. Against a Northerner opponent, you often want your opponent to attack, to have them skewer themselves on your defenses. If you get lots of grunts and trolls and no goblins, you lose strategic flexibility since you already sort of have to attack in order to not fall behind economically. By recruiting some goblins instead of some grunts, you gain almost as much offensive power as the grunts would have given you, lose some potential tanking units, but also keep a lot more gold over time.

I do have to agree that in the more fast-paced and action-packed matches in which both sides lose and kill units early, grunts are near 100% the better choice. In those kinds of matches, upkeep doesn't really matter and solid lines that allow you to protect your glass cannon-type units don't really exist, so units gain a lot of worth simply from being able to defend themselves, and if fewer units are on the field the inability of the goblin to project a ZoC is much more painful.

[Tonepoet]

Good point: Goblins often do survive more than five turns more often than not; those five turns can easily translate into lost opportunity though, so there's a cost to be had for it still. Also if the goblin lacks the individuality of a distinguishing purpose, how much flexibility can it contribute? Probably not a lot, even if there are some things it can do that others can't. Surly nothing necessary and I still have the whole rest of the faction to work with too.

Besides, I don't believe you always "have" to attack in those situations or face a disadvantage. In slow matches, the players acknowledge that they believe attacking is more difficult and might not be in their best interest, simply by not taking advantage of the opportunity. This often leaves you with the alternative option of slipping up on your recruits for a few turns to hold your gold and not pay the upkeep, giving you the potential to reinvest the gold that you would've spent on a goblin on something that might serve you better in the long run. Once you think he's getting ready for actual combat or if you're particularly clever, just a touch before, you can unleash your hidden assets and present something seriously threatening. Afterall, if your opponent's just going to shy away the whole time, what's stopping you from utilizing this strategy? Not much honestly, so long as you're paying less upkeep to begin with. During this time, the leader can even be used to reenforce the battlefront, if you so please and are very, very careful about the manner, although it's not necessary.

[darkside]

goblins have meele pierce damage - helps a lot vs drakes if you use them clever.

[Dodgy Tactician]

Non-Dim Goblins require only 18 XP to level into Goblin Impalers, which gain the Firststrike ability on an 8-3 melee attack. By comparison, Non-Intelligent Grunts require 42 XP and Non-Intelligent Whelps require 36 XP to level up.

If you get an Impaler, then they become the bane of all mounted units and can carry a significant threat to mounted units at more than just night-time, being able to ZOC means that they can be used to trap enemy units prior to killing them and the attack is far better than that of a Grunt or a Whelp. They still can't exactly tank, but it's a lucky enemy that can take them out with a single unit and they can often dish out enough retaliation damage in melee to be a worthy sacrifice for the prospect of killing a couple of enemy melee units.

Also, placing them on a village that you don't want the enemy to steal but you don't have the resources to defend (eg. because you're busy crushing him in another part of the battlefield) will prevent the enemy from stealing it. Against a determined attack, you will not be able to hold it, but keeping it there will not cost you anything in upkeep whereas any other unit on the Northeners' roster will cost a lot more. Consider holding a village two turns away from your keep for a full day-night cycle. The Goblin will cost you 9 gold, a Grunt will cost you 12 Gold to recruit plus 2 Gold in upkeep getting to the village and a further 6 gold in upkeep to hold it - making 20 gold in all. That becomes 26 gold for a further day-night cycle and so on.

Goblins don't have an automatic place in any army, but as a specialist unit they have specialist uses that they shine at.

[Tonepoet]

Just to reitterate, the argument being proposed is that because their HP is so low the use of their specialist role is near inadvisable. All of the pierce vulnerable level 1s in default era have moderately strong to very strong melee attacks that highly threaten the advisability of using the unit to attack because if you have something else, you're more prone to gain solid means sometimes even if its fully capped off.

In its current state, it can't really be used to a defensive extent either because it's simply too flimsy to rely upon: one lucky attack and you may've very well been better off simply leaving yourself completely unguarded because now you've not only lost the village but you've lost 9 gold to go with it. Consider for a moment that this is done in order to gain what I'd perceive as an 8 gold advantage (20 gold investment -12 because you could still easily have the grunt in this scenario) you're now at least 1g loss over that and every turn thereafter a six gold per round swing. Gryphons, Drake Fighters/Burners and Horsemen can gain this result rather easily; Wolves and Ghosts somewhat less so but its not entirely outside the realm of possibility; you might need more than Elvish Scouts to pull it off. I suppose you could say you have the advantage of diverting the opponent elsewhere but those units are often bought for the specific purpose of village stealing (well okay, maybe not the Drakes but due to mobility they can repurpose themselves easily,) so it's not much sweat off of the opponent's back to try for this

As far as exp. management goes, I believe their quicker deaths mean no advantage overall either way. Low exp. doesn't necessarily mean a unit is quick to level, take the Elvish Shaman or Vampire Bat for instance, they may only need two kills to level but they're usually lucky if they can even get one because of their low attack. I'd imagine it's just as hard to level a Goblin Spearman, except for the opposite reason; it's too easy to kill off before it even gets there. Even if you do get the 1 kill, a return kill means you've lost 8 exp. where as a Grunt is more likely to retain it. Similarly, in terms of preventing your opponent from gaining exp. the difference may be similarly moot: your opponent might gain less experience per kill, but they'll get more kills.

Why get a unit that's just a specialist when a generalist can gain the same if not better results and has added versatility to boot? Keep in mind that I've been talking purely about its comparison to the grunt so far, so I haven't even had much of a chance to bring it up but do in mind that for fighting cavalry an Orcish Archer is a highly necessitated unit for the faction. You'll already just have a lot of innate pierce damage on the ranged, without even needing to bother investing your gold into specifically.

[Dodgy Tactician]

Ok, let's run some numbers.

Horseman is Lawful, has 38 HP and 9-2 Pierce Attack (Resistances: 30% Impact, 20% Blade, -20% Pierce).
Goblin Spearman is Chaotic, has 18 HP and 6-3 Pierce Attack.
Goblin Impaler is Chaotic, has 26 HP and 8-3 Pierce Attack with Firststrike.
Orcish Grunt is Chaotic, has 38 HP and 9-2 Bladed Attack.

Goblin Spearman attacking a Horseman at night (assuming both on 40% 60% CTH). Using notation HP (Probability):
Horseman: 38 HP, 7-2 Attack. 38 (6.4%), 29 (28.8%), 20 (43.2 %), 11 (21.6 %).
Goblin Spearman: 18 HP, 9-3 Attack. 18 (16%), 11 (48%), 4 (36%).

Bring in a second goblin and you get a 23.3 % CTK:
Horseman: 38 HP, 7-2 Attack. 38 (0.4%), 29 (5.0%), 20 (13.8 %), 11 (27.6 %), 2 (31.1%), 0 (23.3%).
Goblin Spearman: 18 HP, 9-3 Attack. 18 (16%), 11 (48%), 4 (36%).
Goblin Spearman: 18 HP, 9-3 Attack. 18 (16%), 11 (48%), 4 (36%).

Using an Impaler instead:
Horseman: 38 HP, 7-2 Attack. 38 (6.4%), 26 (28.8%), 14 (43.2 %), 2 (21.6 %).
Goblin Impaler: 26 HP, 12-3 Attack. 26 (16%), 19 (48%), 12 (36%).

Using a Grunt:
Horseman: 38 HP, 7-2 Attack. 38 (16%), 29 (48%), 20 (36 %).
Orcish Grunt: 38 HP, 9-2 Attack. 38 (16%), 31 (48%), 24 (36%).

The Grunt has little more than parity with a Horseman, even at night since the Horseman's resistance to blade all but cancels out the Grunt's night-time advantage. A cheap Goblin however has much better chance of playing havoc with the Horseman's health insurance. Personally, I'd jump at the chance to sacrifice a Goblin to kill a Horseman, since I'd be 14 Gold up, plus one for every turn since the Horseman was recruited. Even if the Horseman has just killed an Archer, then I'd still be at parity.

Another example is if you have an Elvish Archer on a village that you must attack from Flat. It has 1 HP left, but you want to kill it off. Assuming that your assassin supporting the move has already attacked it, who has the best CTK?

Grunts or Trolls have two strikes, therefore they have a 64% CTK.
Goblins or Archers have three strikes, therefore they have a 78.4% CTK.
However, the Goblin attacks in melee and the Archer at range, so the Goblin will only take melee retaliation. He is therefore the best unit to attack the enemy Archer to finish it off.

Yes, you have to be careful using Goblins because you are using a level 0 unit against level 1 enemies. That's like when you use level 1 units against level 2 enemies in campaign (or indeed MP). You can't fight on equal terms or you will be annihilated, so you tip the battle in your favour. Against higher level units, even if you do the damage type that they are weak to, you have to fight at your time of day (unless it is also his time of day, but that doesn't apply to Goblin Spearmen).

In its current state, it can't really be used to a defensive extent either because it's simply too flimsy to rely upon: one lucky attack and you may've very well been better off simply leaving yourself completely unguarded because now you've not only lost the village but you've lost 9 gold to go with it. Consider for a moment that this is done in order to gain what I'd perceive as an 8 gold advantage (20 gold investment -12 because you could still easily have the grunt in this scenario) you're now at least 1g loss over that and every turn thereafter a six gold per round swing.

A Goblin will never stand up to a full-on assault from decent combat units, that is true. If it ends up facing one, then the enemy has out-manoeuvred you and you'll have to pay the consequences. What I am talking about is one or possibly two fast-moving scouts grabbing villages. In general, fast-moving scouts do not have particularly powerful attacks (Horsemen and Gryphons excepted). A Gryphon still needs to get two hits on you and your retaliation can kill a Horseman outright from Dusk until Dawn (which may give him pause for thought), but both of these are capable fighters as well as being capable scouts. Elvish Scouts, Footpads, Gliders, Bats and Ghosts can't kill you outright in one turn, even at their time of day where appropriate. Cavalrymen and Wolf Riders need three hits at 60% defence (16% chance) and even then, they can't occupy the village until their next turn. There is therefore a 16% chance of spending 9 gold to delay a 6 gold per turn swing by one turn (ie. you lose three gold). There is also an 84% chance of delaying it by two turns or more (ie. you gain 3 or more gold).

You also gain information as to your opponent's intentions and if he has other slower-moving troops threatening your village, then you have a pretty good chance of having a turn's warning before they attack. If you see five Dwarves coming at your one Goblin for example, then the Goblin is going to die, whether he retreats or tries to hold the village. They have less movement though, so you get a turn's reaction time to send troops to defend whatever object of strategic value is behind that village (eg. chokepoints, villages or your leader). You lose nine gold that way, but your opponent has committed a significant force to a surprise attack that you have detected. That sort of information is worth the nine gold you've just lost.

The other fairly obvious use is as a hex-filler. If you out the Goblin in a Hex the enemy wants to occupy (eg. Forest for Elves), then the enemy must first kill it before occupying that hex. That uses up a minimum of one unit's turn. They can also stand in front of an injured unit to reduce the number of enemies that can attack it. These are "fodder" uses, but given their low upkeep and high attack, they can make the enemy pay for killing them a lot more efficiently than most other fodder units.

Edit - corrected CTH value.

[Huumy]

Sorry but I read only to the end of page 2. But some people made good points:
Goblin cost 9gold with no upkeep most of times it can't as good as grunt... but sometimes it's better than grunt.

IMO goblins are most useful when it's past the early game rushes, when theres group of units vs group of units. Still theres no use to have too many goblins.
So instead of making mathmetical arguments or giving situations where goblin is better:
How about any1 who doubts the usefulness of goblin go play few games with northeners and buy just 1 or 2 goblins to go with your larger group of army and see how it works.

Results:
1. Wohoo there was actually situation where it made more sense to use goblin than grunt or troll.
-there you go, wasn't it nice to find something new in wesnoth?

2. The goblins were useless they only made my army weak.
-it seems your playstyle don't work with goblins. Stick with your old style or figure out how to use goblins, anyway you want.

3. I didn't see any difference, it was same like having 1 more grunt.
-Next time you play northeners try goblins again you can only get better at using them.

[Euthanatos93]

goblins are not entirely useless. One with an initial recruit will keep upkeep down and cheapen an already very cheap faction (which...by such nature, gives it a high upkeep it's a practical balance). Once you know your enemies' recruit one or two more may be called for. Their high damage makes them nice to put a killshot in and make them easy to level.

That's gobos in a nutshell, y'all give it too much damn thought.

[pauxlo]

Ok, let's run some numbers.
[...]
Goblin Spearman attacking a Horseman at night (assuming both on 40% CTH).

You mean "Chance to not hit" aka "Defense", I think.
40 % CTH is 60% defense and this does not occur for a human horseman. (Your numbers also were for 40% defense, not CTH.)

[Dodgy Tactician]

You're right, I really should proof-read my posts better.

[Sorrow]

Running numbers in itself is rarely useful in the real world of wesnoth where tactics, gold management and the like do not produce ideal battlefields for either player.

Goblins have their uses for their sheer offensive power and the fact that they have 3 strikes. They are useful in situations against high defense targets that are too powerful in the ranged department of either archers or assassins to try and tackle.
Making 1 per turn when you are at upkeep can be an effective plan if you want to play defensive/economy game.
They are very nice for filling holes in your ZoC.
You can use them like ulfs to try and control what your opponent does to some degree.
Another move would be to use them as attack absorbers.
Kind of an abstract of a concept, but you can use goblins to limit the number of good choices your opponent can make. WCs can be used this way too.
My favorite reason to buy them is simply their 3 strikes pierce melee. Rounds out the orcish offense (close to home) nicely.

But honestly, I usually adhere to the good old "hit enemy with grunts until it works" strategy.

[Chris7mas]

I think goblins can be of great use to counter-attack Drakes in a Drakes vs. Northerners match. They are cheap and a pack of such guys backed up by some grunts will do serious damage to Drake Fighters and Drake Burners, since those have a -10% vulnerability to pierce attacks. I know it's a particular situation, but a good example in my opinion. Putting up against Drakes at day can be very hard considering their tremendous melee attack, so buying 3-4 goblins can be a good temporary solution, especially if they survive to get their night turn too. On the other hand, they are weak and will die for almost nothing if the enemy recruited 2-3 Saurian Skirmishers. Besides the fact that they are cheap, their (relatively) powerful melee attack compared to their cost is another good advantage. And with appropriate leadership from an Orcish Rouser they can get even more useful.

[Tonepoet]

True Sorrow but there are some basic assumptions that can be made. For instance, the full health of your standard traitless goblin is 18hp so barring slow, that number's a fairly safe constant, I think. Negative traits kinda bend my mind in uncomfortable ways so perhaps slow is likely. At any rate your safest bets are probably relying on the lowest numbers possible, so assuming you get an 18hp goblin, this means a regular horseman can attack it during the day for what'd likely be a cost free O.H.K.O. Also keep in mind that I find it highly unlikely for a horseman to attack a cheaply produced fully chaotic faction at night. There might be some instances where the death of a village holder is game breaking but I'd want a beefier unit guarding vital ground in such cases if not only for security purposes. The original presumption was that we had a full ToD cycle wasn't it? If so, I'd presume rather scrutinous horseman use, since they're such high cost units with such a stake raising special.

I'd go on further but I'd presume the ad nauseam nature of the discussion means the use is obvious enough to the average player that much further commentary by me would amount to my beating of a dead horse. That is unless somebody reeaaaaaallly wants to hear all of the contrived little considerations I have on mind for why another unit might be equally suitable for attacking a 1hp Elvish Archer. I kinda doubt it but if so, just ask and I'd be glad to give it a once over.

[Sorrow]

Ah, I did not mean to come across as bashing using math in wesnoth at all. I just meant the kind of number crunching people usually do on the forums has 0 meaning, doesn't take situations or traits into consideration and is just plain fluff. Play 40 games with northies straight and you will know the niche of the goblin well enough, no need to theorize on forums :\

Math for use in the game is super useful though. Like only use strong grunts to hit the 24hp mages, only non-strong ones to hit 33hp elvish fighters, and so forth. Very useful for maximizing the number of kills you can get during your favorable ToD.

Also 18 is an annoying number for HP being a multiple of two very common damage values. Makes it very easy to finish goblins off perfectly with no wasted damage.