What actually needs to be balanced (statistically speaking)?

I won’t argue against that. This analysis definitely doesn’t show the whole picture, it’s just a new frame of analysis. And your definitely right about the power of using statistics to manipulate results. I tried to be objective here, but I think the most subjective I got (besides the tier list itself) was in how I chose the error bars. 1.5x standard deviation seemed to line up best with what the community says, but I could have chosen any number of metrics to tell different stories. That said, I tried my best not to just have this be my own opinions, instead more of a breakdown of the communities collective opinions.

I’m not sure how I would implement boosts though. They haven’t even made a boosted tier list yet. The best I can do is assume that everything can be boosted evenly, so boosts don’t really matter for understanding the base strength of a dino, but this isn’t strictly true.

2 Likes

Understood. Thanks

1 Like

It’s based on Gamepress’ tier list which is not at all only based on stats…

That’s cool. Understood. Still doesn’t change the overall thought.

tenor (23)

A good amount do.

The bottom line is this. These creatures have no business being as strong as they are. Easy to get epics and legendaries should not overwhelm uniques.

1 Like

Noone in here using any of that popcorn to feed the trolls right? I’m watching you guys… Agree with most of the things that have made it onto the list. Most are not problematic but I could see using this as support for the decision to nerf/buff. Do wonder what the decision making process at Ludia hq looks like. Usage vs win rate?

5 Likes

Doubt so. The balancing changes in 2.2 and 2.3 don’t correspond to usage or win rates at all. It seems random, frankly - why else would things like Apatosaurus, Scaphotator, and Carbotoceratops get nerfed?

2 Likes

I agree with some of your other points, but this isn’t really relevant here since the process by which the graphs were obtained is completely transparent in this case. You’re presented with the full context.

1 Like

What I have often found is that the dinosaurs people often want nerfed tend to be the ones beating their team.

2 Likes

Thats why I tried to be more objective here, looking at balance from a different argument besides “x dino needs a nerf because it beat my y dino”. Tier lists aren’t completely objective, but they can do a good job of combining thousands of matchup simulations with other benefits that simulations can’t really capture (like DCs swap in) and the “expert” opinions of some of the game’s best players (in the case of the Gamepress tier list), so I think they’re about as good as we can ask for without having access to all the internal data.

Again, understood as I’ve said earlier.

My statement as a whole still stands, viewpoints are always subjective, mine included and to the point of this reply, graphs. What can not be denied, is that the graph/viewpoints presented earlier presents one side of the facade and to Mudkipz credit and full realization of ability, limited or not, has not and possibly can not take into account. So how CAN it be completely transparent? Keyword, completely.

Heck, all you have to do is look at the game in real time to see how many factors are at work regarding that. I get that the viewpoint, is a work detailing BASE ground work at its starting point, but beyond that what is it? The starting point is lost literally, the moment the game is played. Sort of like buying a new car fresh out of the car lot. The moment it touches the street, it loses its value.

Flawed illustration(s)or not, it still stands. The reason for it was to try and point out its subjectivity, before it sparked an “outrage”, that had already started. To get ones to slow down and think instead of jumping on material just because it was tangible. Heck, even Ned responded.

Anyway Qaw, that’s that. I’ve said all I’ve could and wanted to on the matter. :tooth: :tooth: :tooth:

I disagree that the graph is subjective. Regardless of your opinions, Gamepress has put x creature in Tyrant, y in High Elite, etc. That’s the objective reality. The graph is just an alternative representation of that tier list, but it is wholly objective, even though your valuation of its accuracy or importance with regards to balancing may be subjective.

I’m not saying that it accounts for every variable. I’m just saying that we know everything about that graph that there is to be known, every bit of context and every step leading up to obtaining the final result. I think you have a different idea of transparency from mine; (quoting Wikipedia) “Transparency, as used in science, engineering, business, the humanities and in other social contexts, is operating in such a way that it is easy for others to see what actions are performed. Transparency implies openness, communication, and accountability.”
By that definition I think it’s fair to say that the graph is completely transparent.

Earlier you said

This is only the case when important information is withheld from the reader, which is obviously not the case here. You have the full context.

The graph has been presented to us as a representation of the variance in GP tier placement within each rarity. That variance is not subjective. What is subjective, again, is how relevant, accurate or important the values the graph represents are as far as balancing goes. We can all agree that it isn’t the be all end all when it comes to balancing, and no one is saying that it is. We’re all just pondering it’s importance here, but the data itself is what it is.

1 Like

Basically, what I’m trying to say is, the graph is an objective representation of a group’s opinions. There is no withholding of information and hence no toothpaste trickery going on here.

@Qaw and @Tim_G I agree with both of you guys. While the data does objectively represent the opinions of the group, these opinions themselves do not objectively represent the state of true balance in the game (though using Gampresses tier list is probably as close to objective as we’re going to get). In terms of this analysis in itself goes, I like to think I was quite transparent in my methods, even if the data it was performed on isn’t perfectly transparent in itself. While I hope people will take this analysis into account, I would never want Ludia (or anyone) to base their opinions on game balance entirely on my work here. It can be a good relatively unbiased starting point, but it should also be supported by one’s own observations, and more importantly by the internal statistics that Ludia (hopefully) keeps on the game.

1 Like

It is interesting to note that the 1.5 patch notes go into some detail on how balancing is done - Ludia keeps track using an estimated survival ratio and balancing relative to the global pool of creatures. It is also nice to see how detailed the explanation for changes used to be. All the inane balancing changes lately might have been better received if some explanation were offered - otherwise they just seem frivolous.

3 Likes

I think this is a good analysis and largely reflects the state of play on an individual level without boosts of course because we all know how that changes the game. But I think there are other aspects that also come into play around how effective something is, and a lot of it is in combinations.
For example a Sarcorixis/Stygidaryx/Swap in stun combo is disproportionately effective except against certain combinations because of the bleed/swap prevent/stun cycle. It can be countered, there’s no impossible to beat strategy - so long as you have the right combo in your pool needed at the time to beat it.

I find generally the creatures with a high number of big resistances also tend to be disproportionately powerful because not only do they resist practically everything, they tend to have abilities to counter everything. Yes they can individually all be taken out, but in combination end up severely weakening an opponent to the point where they can maintain an edge the entire match due to high resistance, high utility and no swap in disadvantages.

Once coupled with boosts, a bunch of these combinations become essentially mathematically predetermined outcomes because one stat pool and ability combination will consistently outperform almost everything else except very specific combinations that you can’t guarantee you’ll pull.

Once you also add imbalanced matchmaking most of the time (my opponents tend to have a combined health pool about triple the size of mine and about double the damage of mine) there’s a lot of very unrewarding matches.

2 Likes

Yeah, there definitely are some emergent traits that show up once you start combining different dinos. This is actually one advantage of using a tier list over something like win rates, because humans can manually account for these kinds of situations in their rankings, though it does introduce some subjectivity.

I do agree about resistances making some dinos very powerful. If something is going to have a lot of resistances, it’s attributes or abilities need to be weakened to compensate, but this isn’t always the case. As such, certain dinos become excellent targets for boosts, which makes these problems worse.

This is part of why I think it’s ok NOT to account for boosts here: the best dinos to boost generally tend to be the best dinos anyway. I can’t really think of any cases where an average or subpar dino becomes a monster with boosts. The closest might be Thor, but Thor isn’t even that bad (probably in Elite High now if I had to guess). No, the more frequently boosted dinos are the ones that were already at the high end of the tier list in the first place. So while boosts can certainly make some dinos stronger, it doesn’t necessarily change the order of dinos on the tier list. If anything, it just widens the gap between the higher tiers of a rarity and their lower tier counterparts, making their nerfs all the more deserved.

1 Like

I agree that this analysis does not need to factor in those things, because the performance of individual dinos still matters. Some dinos are never used because they just don’t provide any utility. So if a dino has stats that says it’s about right, but it isn’t being used, that tells you that there’s something else about it that’s getting it passed over.

There are certain dinos that gain specialised uses with boosts that they can’t do with their base stats. eg diplovenator is a really good counter to raw damage creatures with boosts because it needs them to actually survive and do enough damage to make it meaningful, whereas without boosts it’s slow, low hp and low damage. It’s also completely countered by resilient moves, the most common moveset in the game.

Thor is disproportionately a problem not because of it’s stats so much because it can be countered, but because you’ll routinely get a level 24ish team with an average speed of 120 and a combined healthpool of less than 12 000 hp facing a level 30 Thor with 130 speed dealing 5-7000 damage per hit

Even if the stats are a little more even, it’s the constant crits and the instant charge erasing any advantages you have due to the high damage output against teams that are just not at that level. Instant charge is the only ability it has that ever needs to be used to do 1x damage. If you reduce it’s damage with distraction (50%), it uses a 1.5x-2x damage ability and crits, still doing a minimum of 100% of it’s already high base damage. If you aren’t immune to stun, then instant abilities become a weakness, not a strength because it will hit you again and likely crit again.

But of course if you have a thor at roughly the same boost and level as the opponent, then the thor is easily countered in a number of ways.

1 Like