We’re all tired of people who don’t know squat about probability calling out the AI for getting “too many” cascades and free turns. We’ve been over and over how it’s always the same, and there’s no such thing as the AI manipulating the RNG such that it will get multiple cascades to counteract your winning streak.
I think there’s a really easy way to put a stop to those threads. In the next update, implement this rule:
“Every time the CPU gets more than 8 free turns after a move, the player is awarded a mythic ingot post-game. Every time the CPU makes at least 10 cascades after one move, a Treasure Gnome replaces one of its troops.”
This is easy and harmless. By the numbers, this should only happen once every few thousand games, and thus only introduces 2 or 3 mythic ingots to the economy daily to really unlucky players, who will suddenly be happy for their misfortune. This will also help combat the stale Divines meta by causing misguided fools to set Goblins as their PvP defense.
But we all know this never happens. I certainly haven’t seen Goblins get 14 turns followed by 9 turns twice today. I definitely didn’t see an Elemaugrim team manage to pull off 9 turns while I was typing this up. And I certainly didn’t lose a GW match today to a King Bloodhammer team that, with a Bonestorm active, managed to get 6 turns + 12 skull cascades off of a 3-match while, on 5 consecutive turns, me making 5-matches didn’t drop a single skull. These are all impossible, so there’s no way I would’ve made any mythic ingots today.
Which is why this proves it’s fair: after a few weeks of this rule, players will realize they aren’t getting any mythic ingots or treasures, which must mean the low-probability events aren’t ever happening.
Any storm active with explosions tends to create cascades, Doomskulls were just the missing variable to make Bonestorms more propense to cascades.
I get it. You had a bad day, so far, in these specific aspects of the game more people also suffered with it, other people had bad luck with chests¹, orbs and so on…
¹ 40.000+ of glory for event keys (plus 230+ keys i already had) to finally ascend Pride to Mythic.
Yes, I know how bonestorm works. I also know if I have a bonestorm team I really don’t want to pass the turn, because then my opponent might cause explosions that cause me to get hit by doomskulls.
So in my hypothetical that totally didn’t happen:
CPU killed one of my dudes and summons a doomstorm. (Sorry, I meant doomstorm all along.)
I cackle and fire off Azura. Surely exploding 8 blue gems will cause some doomskull cascades.
0 skulls fall. However, 12 blue gems fall in such a way that 5 of them are touching a skull.
The opponent fires off king Bloodhammer, gets a 5-match, makes a 4-match, makes a 4-match that cascades into 3 doomskull cascades, makes a 4-match that cascades into 2 doomskull matches, makes a 4-match, makes a 5-match, makes a 4-match.
I have a 4-match with doomskulls available and make it. No skulls fall. I make a 3-match.
The opponent makes a 3 match that cascades into a 3 match into a 3 match into a 3 match into a doomskull match into a 4 match into a doomskull match. They make a match 4, a match 4, a match 4…
It’s bad luck. 3 out of 5 matches. For 5 days. Hypothetically. If I weren’t so rational, I’d say something like, "I’ve been playing with these teams for weeks, and I am spending an eye-rolling amount of time waiting for my turn.’
But that’s hypothetical. Of course I’d have zero mythic ingots if my rule were enacted. I get as many free turns as the CPU.
Far as free turns go, there is no stopping the combo of Titan Hero + Jarl Firemantle. Got into a match with him in guild war and barely survived thanks to managing to cast Weaver first. It was 20-30 turns to every one turn I had. “4-match, explode, 4-match, explode, 4-match, explode, 4-match, explode, cast Jarl’s spell, 4-match, explode, 4-match, explode…” with the only times I got to take a turn was when the AI would cast one of the other troop’s spells. If they hadn’t been webbed they would have wrecked me. As it was, I still almost lost to getting hit for 3-8 damage at a time, and having to wait a few minutes between every singular turn I got to have.
There are some combinations out there that can certainly pull such tricks, exploded gems could generate zero mana and still refill the mana of an entire team, simply because the board is “renewed” under certain storms and also lucky factors/drops. Some defenses are just better in such things. You probably learned a valuable lesson about checking out the enemies and consider if you should bring Mab or Mana Drainers into these battles.
This topic about the unfair/cheating A.I will be revived many and many times, the A.I doesn’t cares when the player is steamrolling it 1.315.578 times a day with such tactics, but when the A.I does something similar the loss will be engraved in the player’s mind forever.
Mana drain would have lost that fight hard, actually. Mass web was definitely the way to go, at least it kept their damage minimal. If I had used mana drain, I would have been wrecked utterly the first time the enemy managed to get Jarl filled. Freeze is sometimes good for these kinds of fights, but it can be hit or miss. Like, if I didn’t get a 4/5 match early on, or if Mab froze the wrong troop I’d still have been stuck in that loop.
The difference though, between the AI caring and the player caring? Is that this game is not designed to entertain the AI. In addition to the question of “Is it fair?” you also have to ask “Is it fun?” and no, sitting and waiting for a match to be over because only the AI gets to take a turn is not fun.
I can’t honestly elaborate an answer more honest and short as this.
The game will have it’s ups and downs, you must enter any competition knowing full well that there is a chance that you’ll lose, that your opponent is better than you due many reasonable reasons, like mutant powers or wishes granted by djinns and demonic pacts or alien augmentation.
Jokes aside, if it was a matter of discussing balance in general, as in arguing if Explosions and Storms really did any good to this game i would say “No, it didn’t.”
The players have all advantages they could have by being the first to play, critical thinking, cognitive processes to make decisions and luck, while the A.I have just some few parameters to guide not-so-random actions and luck.
There are mechanics present in this game that allows lucky rolls, but these can be used by both the players and the A.I and just like i said, only the players come here to complain, and they don’t even complain about the right issue, which in my opinion are the explosions and storms and some new skulls that now also explodes.
Now… would the players agree that these should go, or they would want an “improved” A.I prefectly depicted by this image?
I’ve always had the feeling that RNG in this game tends to get streaky, like Fizz exploding 4-5 times in a row (two days ago) or IK resurrecting 5 times in a row (yesterday), but I don’t think it’s inherently biased against the player in general… Just that defense teams in general tend to be designed to exploit it (because the IA can’t really win otherwise), and it really sucks when you’re on the receiving end of such a streak
I have a fever and I’m here, I promise I am here. If I need to I am happy to get all the old testing threads back here for you all, wherein our programmers did rigorous testing to show that the AI does not get luckier than the player. In fact, it can’t.
There are outliers where the AI can get cascades, just like the player, but the AI won’t be luckier over time than the player. (I’ll get the threads, but please don’t make me, I’m very sick. )
When i see a troop with 7 attack and an harmless spell killing 2 and half of my troops with 30+ hp and armor in a single turn (that was the 2 clock minutes turn i talked about before) there’s no way i trust them.
Leaving aside cascade let’s talk about devour maybe, i used Cerberus in dungeon for a “long” time and despite the 40% chance had always to restart the matches several times (up to15, ok, sometimes was right at start cuz no mana for him on board).
Then (unless the troops in boss raid a got bosted chances but never read anything about that) i get a stupid land shark that devour 10 times in a row without any fail (at the point i had to use the monster muncher to pass a lvl where the shark was the only monster) or an ai cerberus that not only never fail devour but also spawn a warg every time, wargs that when they die all spawn a wolf too.
And i should buy that ai use same tables? maybe i should aswell subcribe every marvelous offer i get daily on phone from call centers.
Said that i agree about the ai need to cheat seen even with those cascades and “lucky chances on skills/traits” it lose 95% of times.
Tbh isnt even such a big issue, give some challenge, is just annoying see that when the ai decide it’s time for you to lose there’s basically nothing you can do (like in a match i had a famine that was refilling and casting every turn).
There is a thread on Reddit today complaining about this too. Humans see patterns where they want to see patterns. This conversation will never die, even though it should. Devs have addressed.
You don’t have to fetch the threads @Saltypatra. The main intent is a form of entrapment: “If this is so very rare, then there’s no harm in paying players who have it happen handsomely. But if you’re hesitant to provide those jackpot rewards to the one in a million who hit it very rarely, surely you believe it is more common than it seems?”
This is somewhat half-facetious, but at the core I have a few arguments I’ve made before, one or more I see in @Razzagor’s responses:
Explosions, storms, and generators can create board situations conducive to very long cascades and strings of free turns. (Goblins exacerbate this.)
Players don’t get super excited when they get a lot of cascades because, in general, players expect to win most of the time anyway.
Players do get super upset when the CPU gets “an abnormal” amount of cascades, because it generally means they lose a game that they expected to win.
Overall, free turns make me angry far more than they make me happy.
I might win a losing game because of a lucky free move once or twice a week. I forget them almost instantly in a sea of games I’m grinding. I don’t even notice the time lucky moves help me win a winning game. I was already winning.
But every. Single. Time. I have to spend more than my accustomed 45 seconds because Goblins is taking 9 turns sticks out in my head. I can generally tell you how many times it’s happened in the last 4 days. (5 this week.) It’s very rare for it to happen. So rare that, if I get frustrated and switch to Goblins myself, I might play for an entire day without getting the satisfaction of completely locking my opponent out the way I remember being locked out.
This is confirmation bias, yes, but it feels frustrating nonetheless. It’s why I’ve always argued for caps on free turns, or a game mode where you simply don’t get free turns. They make players more angry than happy.
I sure would love for the game to start designing towards me losing “because I was overpowered” instead of “because when Zuul’goth can activate 4 times in 3 turns you cannot build a team that will win”.
I don’t think it gets luckier, I do wonder about bursts of luck though. Using the coin flip example, the number of heads and tails seem to be the same in the long run, sometimes the coin sure feels like ending up on the same side for an extraordinary number of times in a row. Has there been any testing in this regard?
Examples of streakiness that I couldn’t help but notice this week:
I’ve opened 250 event chests and received 17 Barghast, 0 Pride. Both troops are epic, the only epic troops in the pool.
While farming for pet gnomes I first found four treasure gnomes. Each of them gave me 100 souls.
I’m not sure about the probabilities involved, I suspect it’s somewhere in exceedingly unlikely territory. Somehow lightning appears to like striking the same spot multiple times.
It was proven in a very interesting thread regarding TDS’s old resurrection trait that the RNG has a strong bias towards streakiness.
That means if you observe 100,000 trials and expect 1%, you’ll see something around 1,000 hits as expected. But if you narrow that to bands of 100, you won’t observe something approaching an average of 1 hit per 100. Instead, the analysis suggested you’ll find a remarkably high percentage of bands with 3 per 100 or 0 per 100, and much lower than expected instances of 1 or 2.
I feel like this results in the RNG creating “very bad days”, where it’s just going to give Fizzbang and Nobend free reign to spend 18 turns doing nothing. In theory this also means sometimes you’ll have “really good days”, but as I pointed out before my best day in GoW doesn’t make me as happy as my worst day makes me angry.
This one’s not as unlikely as you might expect. It’s a weird vagary of event chests that a glory shop troop has 3 times the normal chance of being pulled. So this week, Barghast has 3 times the chance of Pride, as they’re both Epic. So you were unlucky, but not surprisingly so. The odds of pulling no Pride from 250 event keys is around 1 in 500. 18 Barghasts are exactly the average number expected from 250 keys.
I hate to say it, but…um…this premise already shows that you’re misunderstanding the statistics. Irony!
If something “should only happen every few thousand games”, and thousands of people are each playing several games every day, then that “rare” event will happen all the time, overall. “Every few thousand games” is more common than you think - again, overall - when there are “many, many thousands of games” happening.
)
Irony!