# If team selection is random, why was my most powerful creature selected for my lineup only forty-eight times out of one-hundred twenty-five consecutive battles? There is a <0.6% chance of that happening at random

#1

i got the sense that my indominus (my strongest creature) wasn’t being selected for my lineup as often as it should have been, so i started counting. in the same trophy range (~2,200 to ~2,500) and with the same team, it was selected forty-eight times out of one-hundred twenty-five consecutive battles. the probability of any one creature being selected that infrequently or less out of eight creatures at random in one-hundred twenty-five consecutive battles is 0.6%. in other words, six out of one thousand one-hundred twenty-five consecutive battles will see the indominus selected for the lineup forty-eight times or less. meanwhile, my velociraptor (my weakest creature) was selected seventy-two times out of one-hundred twenty-five consecutive battles. the probability of any one creature being selected that frequently or more out of eight creatures at random in one-hundred twenty-five conseuutive battles is 3.7%.

why is the game preferentially not choosing my strongest creature and choosing my welaest creature for my lineup during team selection when team selection is supposed to be random?

(yes, i understand how probability works and i understand that what i observed, although unlikely, is within the realm of possibility, but what i observed is also below the 1% significance cutoff — it’s a significant effect and deserves an explanation.)

4 Likes
Chose random dinos for battle?
#2

I dont really have much to add to this topic other then do you really think the title needs to be that long… problem with random team selection… would have summarized your post nicely then that giant wall of text.

I mean thor is easily my least picked dino by the random team selection. While i get indo raptor almost every match. There the same level but i may keep track on how many they get selected. It wouldnt really surprise me if there was an algorithm at work to give priority to your weaker creature thus making you feel you need to spend cash.

2 Likes
#3

So, what unholy force compelled you to put your sanity through 125 consecutive battles? I can’t even get through 5 consecutive battles let alone 125.

4 Likes
#5

it’s an issue that must be addressed by ludia and, after putting in the effort to collect the data to call them out on it, i didn’t want the post getting buried.

1 Like
#6

I’m doing good if I can manage more then 8 -10 fights a week. I’ve been doing way less then that lately though. SIAs really killed my interest in Arena or Tourneys.

I like making oddball teams and doing friendlies though lol.

#7

it wasn’t easy, i’ll tell you that. i collected the data over a weekend. i had a hunch that something was off, but i needed the supporting evidence to make my claim — and i have every reason to believe that this outcome is reproducible.

2 Likes
#8

Not going to lie. U lost me at the title lol

6 Likes
#9

Selection is always random, but some dinos are obviously more likely to be choosen. I usually get choosen way more times Utasino, Tryo, Spinota, Indo or Gorgo.

#10

If certain dinos are more likely to get choosen. That is certainly not purely random… if there is a system th that gives preference to certain dinos that a system worth knowing and using to your advantage

Take my example where thor and indo are equal level and i seem to get indo on my team more then thor. Now if they were both ready to level but i had enough coins to level one. If Indo is going to show up in 50% of my matches and thor 25% … then it would be a better investment to level indo.

Honestly, though more data is needed. I could take this data and say it proves that the rng is streaky. Who knows if over the course of another 125 matches things dont even out.

2 Likes
#11

flipping a fair coin and getting seven heads in a row has a probability of 1/128 or ~0.8%.

if i gave you what i claimed to be a fair coin and you flipped it seven times and got heads all seven times, it should strike you as being a bit strange. not impossible, but improbable. if it kept happening, you’d begin to question the fairness od the coin.

i’m beginning to questions the fairness of team selection in this game.

5 Likes
#12

more data is not needed. the sample size was sufficient and the effect is already well beyond the point of significance — far enough to demand an explanation. not only is the effect beyond the point of significance, but the exact creature that i predicted would be benched the most was benched the most. the hypothesis (“if team selection is unfair, then the most powerful creature will be selected the least.”) was supported.

what you’re asking for is reproducibility. so, in the name of peer review, someone else should make a team with one power outlier on it and run the experiment again.

2 Likes
#13

Then why are you complaining. It’s random. And you accepted it’s already within the realm of possibility. Maybe the next 125 matches you’ll get indominus a lot more. Who knows. Let’s move on.

2 Likes
#14

I created Diloracheirus a few days ago and I also feel like it doesn’t get selected very often… I didn’t do the math, but I’m pretty sure something is not right there and it’s annoying… all the effort just to get snubbed most of the time

1 Like
#15

don’t just quote part of my statement out of its full context. being below the 1% significance cutoff is something to take note of. if you understood how probability and statistics work, you wouldn’t brush it off so easily.

it’s an exceptionally rare outcome that’s far rarer than one would expect if all creatures on a team had an equal probability of being selected for the lineup.

something is off with this game. it’s one thing to play against one another, but it’s quite another thing entirely to have to play against the game itself — especially when the game’s revenue is based on a pay-to-advance model.

i want someone from ludia to explain this. it is not an artifact of fair chance.

2 Likes
#16

without changing your team or leveling any of the team members up, record your lineup for your next fifty (or ideally more) tournament battles. if you can do that, i’ll do the math for you.

#17

I agree with you except on your point that no more data is needed.

I would love to see this reproduced by another player with say 300 battles. That’s called… Well I don’t remember but peer review and the ability to duplicate an experiment is essential to proving it’s validity.

I’m not saying I don’t trust you but I think that it would bolster your argument if a second voice agreed with hard data.

Good work man kudos

1 Like
#18

Your issue here is you think someone from Ludia is gonna respond here and engage you in any type of conversation… they wont… this company doesnt engage in those kind of talks at all. At best you might get someone to tell you to email support and support will tell you theyve forwarded your feedback to the devs. And thats the end of it.

There isnt really anything that says its 100% random. We assume it is, but there certainly isnt any gurantee that team selection is 100% random. If there is some kind of algorithm at play here… noone from Ludia is gonna come on here and say, yeah we do that to encourage iaps.

#19

Although you made a lot of effort, data probably isn’t correct. You should include battles from strike towers too. Eventhough you manualy chose lineup these fights are counted on the long run in probabilities data.

This should be repeated from other players, to get even more accurate data.

#20

In this forum it doesn’t matter if you have an idea about stochastics and samples that indicate that somethings wrong, people will always tell you you are wrong, because their opinion has a higher value than effort to prove something.

1 Like
#21

it doesn’t require that many to show the effect. even one-hundred twenty-five battles was generous. the next one-hundred twenty-five battles are just as likely to reverse what happened as what happened in the first place, and what happened then first time is already beyond rare. there’s reasonable evidence to support that not every creature has the same probability of selection for lineup. some should always be chosen a little more and a little less than average, but forty-eight out of one-hundred twenty-five is exceptionally low.