Explaining Battle Calculations

[LichKing]

Since you are actually rerolling with battles where the 90% loses, on average, you end up getting BETTER results when you have a 90% battle than you would with just random results.

If I'm not mistaken, in a post was explained that the 90% rule is applied to all units. That is, if a certain unit has >90% chances of surviving the battle, it will. In this sense the results are yes better when the battle is rerolled, but in a predictable way. It's a bit less about luck and a bit more about arranging strong stacks. To lose with 89% is still a big setback, and it still happens with a frequency that wouldn't probably happen in a real world battle, so this 'unrealistic' surety when you have >90% chances at most compensates this aspect. If you compare it to reality, luck is still a big element. If you played the original Warlords 2 you probably remember how scouts could kill dragons with a certain frequency , I'm glad it doesn't happen here.

I would like for example if heroes had a skill like "Field Charisma" or "Divine Inspiration" or something like that, that could lower the threshold to 85,80, maybe up to 75 (still retaining the 90% for the single units, or not, I don't know )

[KGB]

Ams16,

90% should mean that you lose 10% of the time. It happens. Luck happens.

Then you are saying you prefer luck over strategy? Because that's what you are arguing for. A good strategy game tries to minimize the effect of luck where ever possible.

The 90% rule is designed to reward good strategy of combing units and bonus's etc. Without that rule in place 10% of the time pure luck will decide important battles which removes a fair bit of the strategy from the game.

It is MUCH better to go in with 90% than it is to go in with 89%, and that's just wrong.

That is the point of the rule. So players know specifically that if they reach a certain threshold they are guaranteed to win. I don't see any problem with that at all since it encourages strategic combining of units/bonus with the knowledge you aren't going to get screwed by bad dice. 90% is much harder to achieve than you think and it doesn't mean you get to wipe the other side clean with no losses and both players get the same benefit from the rule so it doesn't favor anyone.

KGB

[kenc80]

I think the problem there was just that it left a bunch of players and especially new players really really frustrated with the game and the results of those 90-100% battles. In Beta2 there were A LOT of players complaining about the bad results so not to speak for the designers but I think that was the main thing.

Ken

[Yahtzee]

I'm not really strongly on either side of this to be clear - I can see the benefits to having the rule and I hesitate to disagree with KGB give that its.. KGB.

But outside of talking about anything specific to warlords/warbarons, that logic seems wholly flawed. Arguing for 90% meaning 90% isn't arguing to have a game that's less strategic. By that logic, the game is 10% more strategic if we make it into a 80% rule.

[KGB]

Yahtzee,

But outside of talking about anything specific to warlords/warbarons, that logic seems wholly flawed. Arguing for 90% meaning 90% isn't arguing to have a game that's less strategic. By that logic, the game is 10% more strategic if we make it into a 80% rule.

Your statement is 100% true, not flawed at all.

Taken to it's logical conclusion, the most strategic the game could be is if there were NO random elements at all (Chess like, in that there would be fixed losses on each side for every battle based on well set rules). I've played similar style board games with armies and no dice, Diplomacy being the prime example of such a multi-player conflict game with armies, no dice and fixed rules for losses.

Most players however like some random element to the game (hero offers, allies, what you find in ruins etc) including in battle as long as the random element doesn't overwhelming affect the game result. The problem in Beta2 (Beta3 also had the 90% rule) was as KenC said, very crazy dice rolls happened too often because there aren't enough samples in combat (maybe only 20 rolls) to even things out in important battles. This led to frustration when a strong hero stack would lose a battle despite overwhelming odds and you'd games you should not have because there isn't any mechanism to even things out (ie have the other player lose a specific low odds hero battle).

KGB

[Yahtzee]

Well, taking it even halfway to its logical conclusion proves that it is flawed though, I mean having a 50% rule would not make the game more strategic. You'd simply be dealing with less calculations and less risk - managing risk and luck is strategy. That said, your last response has me won over anyhow.

This:"there aren't enough samples in combat (maybe only 20 rolls) to even things out in important battles" makes a lot of sense to me. Sorry if someone already said that and I missed it.

[KGB]

Yahtzee,

managing risk and luck is strategy.

Managing risk is strategy. Managing luck is just asking to be lucky. I mean if we played a game where the object was to guess a coin flip (the ultimate luck game) how can you manage anything strategic wise there?

Looking at it another way, imagine you want to kill 8 Lt Infantry with your own Lt Infantry (no bonus's either side). How many do you need/should you send to manage that risk? Well in theory if you send 8 vs 8 and everything works out right, there will be only 1 man left for either side and you'd need to potentially send 2 more to beat that last one if you lost the 8v8. For a total of 10 men. However without the 90% rule in effect your first 8 could lose without killing a man (1/10000) or just kill 1 man (1/5000). So you might need to send 16 or 20 or even more as there is unlimited downside (you might never kill them). How do you manage that? At least with the 90% rule you now know that the worst that can happen in the 8v8 is that the opponent will win with 4 men left. So you can then work out how to deal with the worst case and decide if that's an acceptable risk and how many men you'll need to kill those last 4 (another 6).

That said, your last response has me won over anyhow.

This:"there aren't enough samples in combat (maybe only 20 rolls) to even things out in important battles" makes a lot of sense to me. Sorry if someone already said that and I missed it.

Excellent. Glad to hear it. Since this was implemented in Beta3 battles and games have been much more fair for all involved and even though I still yell when I lose an 89.9% battle I know that it's only my own fault for not making my stack just tiny bit stronger.

KGB

[ams16]

Then we should make this game so that people automatically win a battle when the probability is anything above 50%. Then it will be more like chess, and much less luck.

[magian]

I suspect the 90% rule makes the small ambush ability of orcs relatively useless.

What about running battles through twice and keeping the more likely result? Then you would have the possibility of extreme outcomes, but they would be much less likely.

The 90% rule really seems to benefit those who have the mathematics of the game all worked out. This will always be an advantage mind, but now they can just calculate and be assured of victory. I think lazy players (like me) would be better served by nixing the 90% cap.

I often seem to be losing battles that I have a 12-13% chance of losing.

[KGB]

Magican,

I suspect the 90% rule makes the small ambush ability of orcs relatively useless.

It doesn't at all. That's why 1000 simulations are run of each battle to see what the effect of the 6% is in battle

What about running battles through twice and keeping the more likely result? Then you would have the possibility of extreme outcomes, but they would be much less likely.

I don't think you'd like this result. For example in a 75-25 battle, if you are the 25, then to win, you'd have to win twice (the unlikely outcome twice). Your winning chance is then 1/4*1/4=1/16=4% to win instead of 25%.

The 90% rule really seems to benefit those who have the mathematics of the game all worked out. This will always be an advantage mind, but now they can just calculate and be assured of victory.

I doubt anyone has it all worked out unless you know all the units/bonus's in the enemy stack (requires testing it first). I can usually guess within 10% pretty reliably on most common stacks (not a lot of bonus's) but even 10% still means I might guess 99 when it's only 89 and potentially lose.

KGB