Heroes V: A Gamble?

[sqeecoo]

As a highly strategic game, I think HV might involve too much luck for its own good. Quite a few features are luck oriented: the clearest example are the necropolis' ghosts- sometimes they get hit and die, sometimes they make the opponent waste an attack. I would greatly prefer it they had a set percentage (and quite high at that) of physical damage reduction.
The luck and morale statistic is also very luck oriented. Morale makes a unit act faster, so it gets much harder to calculate when your troops will act, and plan your actions accordingly. While morale in H3 gave your unit an extra turn, it was right after its first one, but with the new initiative feature, morale also prevents your troops from acting, in adition to giving the unit an extra action.
Luck gives you double damage, making it much harder to judge just how much troops an attack or retaliation will kill. It is also quite easy to get a smallish amount of magic resistance, which will make your opponents reluctant to cast spells, because the valuble hero turn could be wasted if some or all of the enemy's creatures get lucky and resist your spells. Clearly, this is more of a problem with dungeon and necropolis, that rely on hurting their opponents with spells. This could be a potential balance problem. Also, being lucky with artifact finding could be more of a boon in HV that in previous games.
While I understand that events based on chance force you to be more adaptive with your tactics, and that properly calculating risks is an improtant part of playing skill, I feel HV might involve a bit too much gambling.
With the new initative bar and more moral and luck modifiers then before, I believe that an (un)lucky roll or two could change the outcome of a battle, creating trial and error situations in single player, and frustratingly one-sided anticlimactic battles in MP, which is never a good thing.

While I am by no means an expert in balance, and I realize that I haven't backed any of my claims with hard evidence, I would like to see my opinions discussed and criticised.

Thanks for reading, have fun

[BenchBreaker]

heroes games have always involved luck, the ghost ability is the same as the block ability of minotaurs in h4, and h4 moral also changed the order of creatures. in h3 gaining extra action for moral has a huge impact on battles and almost all creature abilites that activate on attack are % based (in h4 only the powerful ones like blind are % based) and in h3 whether netural stack joins a hero is random, also which spells you get in the mage guilds are random. generally speaking h4 has less luck factor, but it's still significant, whether h5 has too much, i don't know.

[Blue_Camel]

there's a somewhat similar thing posted at the very bottom of the first page of this HC thread.. concerning creature damage http://heroescommunity.com/viewthread.php3?TID=17725

[HodgePodge]

Good to see you have made it here Blue Camel. Welcome!

[Blue_Camel]

thanks much. it's altogether surprising that anyone remembered me

[Hambone]

it is important to differentiate "luck" (which is purely random) from "probability".

the ironic thing about your concern is that the combat events you mention are based on probabilities rather than pure chance. and so as probabilties, you should be taking them into account in your strategic decisions. it is, in essence, an elevation of precisely the kind of statistics you are talking about since in the long run the creature on the whole will end up doing the same amount of damage.

now, you might not like that, but it is a subjective matter of taste. personally, i get extremely bored of games where outcomes can be pre-calculated with certainty. its like playing a spreadsheet.

[BenchBreaker]

it is important to differentiate "luck" (which is purely random) from "probability".

the ironic thing about your concern is that the combat events you mention are based on probabilities rather than pure chance. and so as probabilties, you should be taking them into account in your strategic decisions. it is, in essence, an elevation of precisely the kind of statistics you are talking about since in the long run the creature on the whole will end up doing the same amount of damage.

now, you might not like that, but it is a subjective matter of taste. personally, i get extremely bored of games where outcomes can be pre-calculated with certainty. its like playing a spreadsheet.

well said Hambone, to add to that the damage range of the creatures are fixed, so you could calculate min/max dmg and use that to make your strategic decisions, e.g. if you want make sure you kill those few remainig black dragons without taking loss with your peasants, then you better check that the minimum damage they'll do is greater than the BD stack's remainig HP, if it's not but close, then you'll have to make a decision whether you want to take the risk, the decision is all yours, there's no luck/randomness involved.

only if the damage range of the all the creatures are huge, like 1-10, can it be effectively considered pure luck.

[DaemianLucifer]

only if the damage range of the all the creatures are huge, like 1-10, can it be effectively considered pure luck.

Well since these are stacks,the ranges do get pretty huge.Something like 100-800.

[BenchBreaker]

only if the damage range of the all the creatures are huge, like 1-10, can it be effectively considered pure luck.

Well since these are stacks,the ranges do get pretty huge.Something like 100-800.

sorry, i am confused. to get 100-800, you'll need a creature who's max damage is 8 times its min damage, which creature has that? the balor has a pretty big damage range at 13-31, but that's only 2.4 times. the biggest range i can think of is the horned demon's 1-4 (h5)

[Gaidal Cain]

it is important to differentiate "luck" (which is purely random) from "probability".

And what distribution would "luck" have? "Luck" is just the fortune of having the random draws work in your favour.

the ironic thing about your concern is that the combat events you mention are based on probabilities rather than pure chance. and so as probabilties, you should be taking them into account in your strategic decisions. it is, in essence, an elevation of precisely the kind of statistics you are talking about since in the long run the creature on the whole will end up doing the same amount of damage.

What, pray tell, makes "pure chance" differ from "probability"? And it's kind of hard to work in an ability that has a 50% chance of making every offensive action you make go to waste...

[Qurqirish Dragon]

only if the damage range of the all the creatures are huge, like 1-10, can it be effectively considered pure luck.

Well since these are stacks,the ranges do get pretty huge.Something like 100-800.

If they do something stupid with stack calculations (say, random damage of 1 cerature times the stack size), then this could happen. However, if they do something intelligent (such as a random damage for each creature, then add the numbers), then the probability of a large stack getting significantly far from average damage is tiny.
For example, take a creature with a damage range of 1-4. There is a 50% chance that it will do an extreme value (1 or 4).
Now take a stack of 100 of them (damage range 100-400). If the damage is calculated individually and then added, then the average damage is 250, and the probability of an extreme amount (under 175 or over 325 to kept the same idea of "middle 50%" from the single creature case)is about 0.1% (just doing rough calculations in my head- based of mean and standard deviation from being inside or out of the 50% range)
the 50% range of value (i.e. the damage done 50% of the time) is from about 236-264 (again, I used rough calculation in my head- this time with a random-walk approach).

Luck due to randomness of damage range should only have an impact on very small groups of creatures that also have large ranges. (such as 5 or fewer level 6 or 7 creatures), and if damage is calculated individually, this is indeed what happens.

I remember doing tests on heroes 3 stacks, and I found that they did damage individually for up to 100 creatures, after which they did 100 individuals, took the average, and then multiplied by the stack size. with 100 numbers, the the distribution of values is so close to normal that this was a perfectly valid way of doing it. This is all based on empirical tests (I used stacks of sprites to check the damage, since they had a small range and no retaliation -so I could have them beat up on stacks of 10000 archangels to get lots of data quickly in one battle- and a hero with attack skill made to give no attack bonus)

This main thing, however, is that if a reasonable model for damage of a stack is used, then that bit of luck is really minimized. Yes, 100 creatures witha 1-4 damage range can do only 100 damage, but only 1 time every 10^60 tries. If you play long enough to have 10^60 attacks, I'd be amazed (for a matter of scale, one year is about 3*10^7 seconds, so you can guess how big 10^60 is!)

And, of course, this makes spells like bless and curse (or whatever they are called now) that much more powerful, since even only boosting the minimum damage (or reducing maximum damage) by a small amount will make the stack do, on average, damage that was well beyond the likely range otherwise! This is also why a spell like the old bless and curse were so overpowered (in H2, give me a barbarian with mass bless and practically any troops, and I've won )

[BenchBreaker]

nice explaination, Qurqirish Dragon, i always woundered how stack damage really works, i imagined it would be 1 of 3 ways:

1) throw the die for 1 creature then times it by stack number (very unlikely)
2) throwing lots of dices and add them
3) calculate the total stack damage and throw a "really big" die

now i know it's number 2), which means with large stacks, everything tends to the normal-distribution and there's really very little randomness involved.

personally i perfer 3) more, with 2) large stacks's damage will be within 3 standard deivations of the mean 99.9% of the times and like you said, dmg ranges close to the min/max stack damage will never happen (unless you have bless/curse )

about that 10^60 figure you mentioned, even the universe is no where near 10^60 seconds old, if we had a billion billion identical universes and add them together, it would still be no where near 10^60 seconds old