Whatever the dev used to randomize the dice roll, it's not working properly

[Quasibobo]

So I reached level 5 and that's when enemies start rallying. In the first battles the enemy always made their check with a discipline roll of 7. That should've been around 6 times the AI made this roll. There is a reason why units below let's say 15% of their starting strength cannot or have a harder time rallying in classical tabletop games. One battle took 17 rounds (!!!) where I could just stand there with my Dwarf Warriors and get charged by the enemy Ratkin units over and over again. They charge, loose combat, flee, rally next round, turn around and charge again. It was 4 units just constantly running laps around me like this. I was pressing space for like 10 minutes and just watch and hope for the best. That's not fun at all.
Additionally the AI averages 6/10 on a 4+ dice roll. I'm sure of it. This becomes extremely frustrating when a buffed up Dwarf Ranger unit with a hero in it just straight up loses to Mongrels. They fought for 3 rounds and the AI made 3 quarters of their 4+ defense saves whereas my Rangers constantly failed the majority of their 3+ defense saves. This is more than my bias or just pure coincidence. I almost feel tempted to save the replay and average out the AI rolls. Judging from the other posts I am not the only one who has reasonable doubts about the randomization.

When you grind your teeth thinking it's just your bias, reach the end and then see your units failing discipline rolls of 10, your Monsters rolling three 1's out of 5 on a 2+ roll and consequently dying, while the AI makes discipline rolls of 5 three times in a row or saves 6 times on a 4+, then it's just frustrating. It's a risk to have dice rolls in a game like this, but when the player get's the feeling the dice are being tampered with it sucks out all the fun. Especially because the AI is so dumb and getting flank or rear charges on it's units is easy, but in the end it doesn't matter if you outmaneuvered the AI, it just effortlessly holds a flank charge. I find it really hard to belive that it's just my perception that's wrong. I have another clue that leads me to believe randomization doesn't work properly. At the beginner levels the AI flees so easily even when charged in the front. On higher levels the AI's units are not improved stat-wise, they just get more units so they're not better or worse than before, but suddenly the individual units of the AI perform way better than before. Gobbos holding flank charges and such.

I wanted to post something like this before but didn't. The experience this morning broke the camels back for me.

[abyssalfury]

Humans are bad at recognising randomisation as actually being random, rather than fitting into specific probabilities (if I roll a d6 six times, I should roll each number once). We also tend to remember things that go against us while ignoring all the times things go our way.

'Other people complained about it' isn't evidence. There's plenty of other games with completely documented RNG systems that are correct, and people will still ***** about them being broken. My favourite is when the RNG is seeded and pulls the next number in line when it needs to roll a dice, yet people will *still* claim the AI gets advantaged rolls despite being able test and 100% prove how the system works.

That's not to say there isn't an issue with the RNG, or that there aren't hidden modifiers in the background, or something else impacting dice rolls. But it would require *a lot* more empirical evidence (or playing around in the back end) to confirm it, not just people 'having a feeling'.

[Quasibobo]

Originally posted by abyssalfury:

Humans are bad at recognising randomisation as actually being random, rather than fitting into specific probabilities (if I roll a d6 six times, I should roll each number once). We also tend to remember things that go against us while ignoring all the times things go our way.

'Other people complained about it' isn't evidence. There's plenty of other games with completely documented RNG systems that are correct, and people will still ***** about them being broken. My favourite is when the RNG is seeded and pulls the next number in line when it needs to roll a dice, yet people will *still* claim the AI gets advantaged rolls despite being able test and 100% prove how the system works.

That's not to say there isn't an issue with the RNG, or that there aren't hidden modifiers in the background, or something else impacting dice rolls. But it would require *a lot* more empirical evidence (or playing around in the back end) to confirm it, not just people 'having a feeling'.

That's definitely true. What my rant is really all about is saying *if* the randomization has hidden modifiers the dev should consider using other means to make the run harder. Being screwed over on random dice rolls feels a lot worse than just being unlucky or making bad decisions. I was salty and if it's just me getting things wrong then just ignore me.

[76561199100143607]

For me it is part of the charm that there is true random in this game - the dice can really be against you, and make your tactics fall to pieces. Just as it should be in something that is simulating the uglines of war. You need to take probablilitie into account, even very unlikely ones.
but yeah, it sucks when you play ranked multiplayer and you loose because a bad dice roll or an unlucky streak. I had my main undead knights with hero and everything, not being able to kill a war machine and loosing, due to undead crumbling, against them. In multiplayer that really really sucks...

[Ilostmycactus]

The dev said in one thread that the game is true random, it's not pseudo random or weighted in any way.

[Andra]

I've definitely had some nonsense rolls like enemy saving 16/16 attacks on a 4+. But then I've had some nonsense rolls in my favour as well. It just feels more punishing for the player because your job is to conserve units and momentum, while the AI can lose every army without consequence because it spawns a new one for the next fight.

If you play Bloodbowl, you know the true feeling of painful dice.

[Quasibobo]

Originally posted by Ilostmycactus:

The dev said in one thread that the game is true random, it's not pseudo random or weighted in any way.

I see. So it is my salty interpretation. Good to know

[Quasibobo]

Originally posted by Andra:

I've definitely had some nonsense rolls like enemy saving 16/16 attacks on a 4+. But then I've had some nonsense rolls in my favour as well. It just feels more punishing for the player because your job is to conserve units and momentum, while the AI can lose every army without consequence because it spawns a new one for the next fight.

If you play Bloodbowl, you know the true feeling of painful dice.

Yeah that's part of it. I used to play a lot of tabletop and know how sometimes you're very unlucky and sometimes the other way around. I think since you're having a lot of battles in a short time it feels different than in real life. In this game there isn't at least a week in between the experience.

[Quasibobo]

Originally posted by Bilstone:

For me it is part of the charm that there is true random in this game - the dice can really be against you, and make your tactics fall to pieces. Just as it should be in something that is simulating the uglines of war. You need to take probablilitie into account, even very unlikely ones.
but yeah, it sucks when you play ranked multiplayer and you loose because a bad dice roll or an unlucky streak. I had my main undead knights with hero and everything, not being able to kill a war machine and loosing, due to undead crumbling, against them. In multiplayer that really really sucks...

How is multiplayer? Are the units well balanced or are some way stronger? I've only played single player.