Bad dice

[Igor]

Need to do something with dice. Very annoying. May be to set auto-reloading all battles where advantage is 80+%, or something else. Games with huge bad dice could be interesting only in principal matches but never just for fun. Need to be changed I think.
Good luck and have fun.

[Moonknight]

If playing this game isn't just for fun for you Igor, I think an intervention needs to convene.

[Igor]

No fun is with mass of bad dice, Monknight
Very annoying and no fun.

[KGB]

We already have a mechanism for bad dice. It's the 90% rule where 90% or better can't lose (which means 10% or less can't win) and the dice are re-rolled.

A 9-1 advantage should be enough to guarantee victory. But a 4-1 advantage (80%) isn't that overwhelming as you should be losing 1 in 5 of those type battles. And to give you an idea of how easy it is to get 80%, consider 8 Lt Inf vs 8 Lt Inf. This is a 50/50 battle at 10 strength. But if you just replace 1 Lt inf with a L1 hero 25 str hero with just a +2 bonus (making all Lt Inf 12 str) from say Isadurs Ring, you easily surpass 80%. That isn't enough to be an overwhelming win. Or consider 8 giants vs 8 giants (all 25 str). If one group has a L1 Valk (+6 bonus) it will be over 80% to win. Once again I don't consider that overwhelming enough to be a guaranteed victory.

KGB

[piranha]

In fact there is something like this in place already. Should update the wiki about it. (Its been there since beta4)

If the total value of battle of both sides is higher than 60 the 90% rule turns into 85% rule. If its higher than 400 it turns into 80%. The reason is because big battles can be game changing so luck should be reduced even more. It would be possible to put more such levels in place so instead of having 3 steps it could be more gradual. The rule may secure a win but for a big battle it's quite important that your army deals some damage to the opponent even when you die. Otherwise one single battle can mean game over and that takes away skill.

[Igor]

How it's counted 60 and 400, from 1st turn to now or another way? Does it counted only side vs side battles or vs neutrals too?

[piranha]

Before the battle starts both armies total battle value when all bonuses are calculated the rule is set. So if 1 giant and a new paladin vs 2 light cav on open terrain it would be 29+29 + 18+18 = 94. 85% will then be used instead of 90%. So 90% is only applied to battles with few and/or weak units. Most battles probably fall within 85% and then the quite big battles within 80%.

The thing that I think could be worth doing is:
1 make it incremental with 1% steps instead of 5. That way it wont matter as much if there happen to be 1 more unit thrown in to reach the next level because the difference is gradual.
2 I could see it going all the way to 75% for the most massive battles. There is not just the problem that you might lose with 75% but if you lose and the cost of the other side is so small that the battle cost you the game. The possible result needs to be more narrow for bigger battles or the game can be decided by luck.

I think we've all seen it and been on both sides, being lucky or having unusual bad luck. Being really lucky in a important battle can ruin a game for me because if I was planing and trying to be smart and manage to get a offensive position and then in a important battle I just kill everything with hardly any loss it takes away the good feeling of having played well and beating my opponent with skill.

[Igor]

In gameid=48863 a battle was with 30+10+25=65 (>60), chance was 87%, the battle have to be won using 85% rule, but it was lost.

[KGB]

Piranha,

Using points is a bad idea. Especially 90 points for both sides because that's way way too low. It makes it harder than ever to kill a lone enemy Barb because technically a Wolfrider won't be able to use his ambush to kill because the % is less than 15%. Plus 85% is much easier to achieve than 90% in battles with larger numbers of units (8v8 battles) simply by having a small advantage over 8 units vs say over 2 units in a 2v2 battle.

The example Igor shows gives an idea of how easy it is to achieve 85% and that battle doesn't look anything close to being what is a 'sure thing'.

The other problem with reducing things to 85 and 80% is that you take away underdogs chances to take a shot at an upset to get back into a game. By that I mean if I am losing, I'll often risk battles where I am only 15-20% chance to win. That's because I have nothing to lose and I need a big upset to have a chance to get back into the game. But you are now taking that out of the game and I don't think that's a good idea at all as there should be a chance for someone who is losing to take a 'hail mary' chance.

I would suggest just leaving the entire thing at 90%. Then allowing heroes to buy 1% increments with their personal points (like intelligence, undead lore etc) as a Luck skill. 1% per point so a 5th level hero could change their stack by 4% to 85% (or 14% when an underdog). Some units could potentially have 1-2 points of Luck skill.

KGB

[Igor]

I'll often risk battles where I am only 15-20% chance to win

I thought only razurandal does so

I absolutely disagree with such method because it brakes the game. Weak player can win skillful player because of big luck. I think that Piranha makes great thing that luck doesn't work when 90-85-80% chance to win.
Only I can't understand why it didn't work in my case.

KGB, imagine that it's real ancient war. General sent a little group to attack big group. Little group knows that they all will die with 85% chance. Do you think they will go to attack?