Ludia Forums

What actually needs to be balanced (statistically speaking)?

There’s a lot of debate on the forums about which dinos need to be buffed or nerfed, and by how much. These arguments often come down to certain matchups or traits, but I wanted to examine this topic from a more objective standpoint. What does the balance curve for JWA look like, and based on that curve, which dinos need rebalancing?

I started by listing out every dino, along with it’s rarity and tier (based on the Gamepress 2.2 tier list). I chose rarity and not “rarity + hybrid level” or “rarity + exclusivity” to keep things simple, though these factors could easily be incorporated into a similar analysis. Using the Gamepress tier list seems to be the most objective way of ranking dinos without access to internal usage statistics, but does have some subjectivity. It’s also the 2.2 tier list, so some dinos will have likely changed in rank since then (e.g. Thor, Grypo). I plan on re-doing this analysis with a more up to date tier list.

I plotted rarity against tier, with rarity on the x axis and tier on the y axis, for each dino. This generated a balance curve for the game, showing how strength increases with rarity. This curve is shown below. Keep in mind that for rarity 1 is common, 2 is rare, 3 is epic, etc. Tier is similar, with 1 corresponding to the “Hatchling” tier, and 10 the “Tyrant” tier. The orange line is the average tier for each rarity. I plotted error bars around this line, equivalent to 1.5x the standard deviation for each rarity. Why 1.5x? Frankly, 1x standard deviation seemed too restrictive, but it could also be used (for example, it would mean that Beta tier commons would need Minor nerfs. Although maybe that’s a fair assessment?). Anyway, here is the curve that I found. I apologize for the quality.

Based on this curve, any tier that fell above or below the error bars was determined to statistically need rebalancing. Tiers 1 tier above or below the error bars were determined to need “minor” nerfs/buffs, and those 2 tiers above or below the error bars were determined to need “Major” buffs. Keep in mind that this is relative; needing a major buff doesn’t mean something needs to be nerfed into the ground, it just means that it’s current strength is significantly stronger than the rest of it’s tier. It also doesn’t mean that so-called “balanced” dinos wouldn’t benefit from getting changes to make them better/better to play against. This translates to the following:

Common: Balanced = Hatchling -> Beta; Minor Nerf = Alpha Low

Rare: Balanced = Survivor -> Alpha High; Major Buff = Hatchling; Minor Buff = Scavenger; Minor Nerf = Elite Low; Major Nerf = Elite Mid

Epic: Balanced = Beta -> Elite Low; Minor Buff = Survivor; Minor Nerf = Elite Mid; Major Nerf = Elite High

Legendary: Balanced = Alpha High -> Elite Mid; Major Buff = Beta; Minor Buff = Alpha Low; Minor Nerf = Elite High

Unique: Balanced = Elite Mid -> Tyrant; Minor Buff = Elite Low

Apex: Balanced = Elite High -> Tyrant

Applying this to the dinos, I was able to determine which dinos needed which balance changes. Keep in mind that I personally don’t believe that all these dinos need the level of balancing determined statistically, or even that they all need balancing at all. This is just what the results showed.

Before I start, since the Tier List is based on PVP, some raid dinos were determined to need buffs that might actually be fine, although they could be made more competitive in PVP. These dinos were Bajatodon, Dsungaia, Entelolania, Irritator, Ouranosaurus, Skoolasaurus, Tenontosaurus, and Tragodistis. You might still think some of these need buffs anyway, and that’s perfectly fine.


For commons, the only dino that needed rebalancing was Glyptodon with a minor nerf.

For rares, Irritator needed a major buff. Tenontosaurus needed a minor buff. Andrewsarchus and Purrolyth needed Minor nerfs. And Marsupial Lion and Suchotator needed major nerfs.

For epics, Koolasuchus and Ouranosaurus needed Minor buffs. Brontolasmus, Diplodocus, Entelochops, Indominus Rex G2, and Thylacotator needed Minor nerfs. Sarcorixis needed a major nerf.

For legendaries, Bajatonodon and Skoolasaurus needed major buffs. Darwezopteryx, Dsungaia, and Tragodistis needed Minor buffs. And Allosinosaurus, Dracoceratops, and Monolometerodon needed Minor nerfs.

For uniqes, almost all of them were balanced. The only one that fell below the curve was Entelolania needing a minor buff. You could also make an argument for some of the “Elite Mid” uniques like Erlidominus, Indoraptor, and Deinorachierus needing buffs, but they were within the curve, and someone has to be on the bottom. It’s just more noticeable with how few uniques there are compared to other classes.

For Apexes, all of the ones released as of 2.2 were balanced.

So with that all in mind, what isn’t on the list? One notable one was Woolly Rhino: despite how annoying swap in attacks can be, it’s overall strength was still within reason, at least according to the tier list. I also already mentioned the “Elite Mid” uniques. Are there any others you can think of?

So what do you all think of these results? Do you agree that these are the dinos that need rebalancing, or are some of them completely off? Personally Sarcorixis, Marsupial Lion, and Suchotator seemed like the biggest outliers to me as far as being above the curve goes, but I’m curious what your takeaways are.

Statistical Balance 2: Hybrid Boogaloo

As promised, I conducted the same analysis again, this time separating hybrids from non-hybrids. I chose this level of discrimination, and not for example including superhybrids or event exclusives, because hybrid vs non-hybrid is the most Ludia tends to distinguish groups by within a rarity. Plus, if I did include these other qualifiers, some of the groups would end up too small to yield meaningful results. And I person feel like even event exclusives should still be balanced within their rarity: they should be on the high end, but not balanced to an entire tier above everything else in their rarity. With that out of the way, here’s the analysis:

I used the same methods as before. Here is the curve that I generated:

It overall appears generally the same, though it was interesting to see that rare hybrids were on average stronger than non-hybrid epics. That makes sense to me though since rare hybrids can sometimes be harder to obtain, so I don’t have much of an issue with it. Plus, they aren’t even a full tier stronger, so they’re functionally the same overall, which again makes a lot of sense.

As far as balance goes, Commons, Legendaries, Uniques, and Apexes were all the same, since they only have either hybrids or non-hybrids in their rarity, so I won’t spend much time on them (Parasaurolophus Lux couldn’t be included, since it hasn’t been placed into a tier yet).

For commons, Hatchling through Beta tiers were balanced. Glyptodon again came out as needing a minor nerf (though speaking from my own personal opinion, nerfing a common at the Alpha Low tier isn’t really neccesary, though I would warn Ludia to watch put for power creep).

For rares, things start to get interesting. Rare non-hybrids showed pretty much the same range as before, with Survivor through Alpha High being balanced, Irritator and Tenontosaurus needing buffs in the major and minor nerf categories (but as raid dinos, this rating is still questionable), Andrewsarchus needing a minor nerf, and Marsupial Lion needing a major nerf (now is probably a good time to remind folks that these aren’t my own opinions on what needs to be balanced, this is just the results of an objective statistical analysis on a comprehensive and generally respected tier list).

For rare hybrids though, everything fell within the balanced range! I think this is because the rare hybrids were quite variable in strength, but are also small in number, so there is no strong statistical average of what a “standard” rare hybrid would be. If we had more rare hybrids, this might change, though I personally agree with this result. For posterity though, if I had applied a 1x standard deviation instead as my error bars (which may be more appropriate for the small sample size), Inostherium comes out as needing a minor buff, and Suchotator comes out as needing a minor nerf.

Epic non-hybrids kept the same range as before, with Beta through Elite Low being balanced. Koolasuchus and Ouranosaurus again were categorized as needing minor buffs. And Diplodocus, as the strongest non-hybrid Epic, needed a minor nerf.

Epic hybrids are perhaps where we saw the most change. The balanced range was Alpha Low through Elite Mid, with the high end being one tier higher than before. This means that on the low end, Edmontoguanodon needed a minor buff. But on the high end, Brontolasmus, Entelochops, Indominus Rex G2, and Thylacotator no longer needed Minor nerfs. However, Sarcorixis was still statistically determined to be in the “minor nerf” category (though this is still one step down from major nerf, and more accurate in my own personal opinion).

I mentioned before that the Legendary tier remained unchanged, and that’s true. It is worth mentioning though that with Skoolasaurus and Bajatonodon needing major buffs, Darwezopteryx, Dsungaia, and Tragodistis needing minor buffs, and Allosinosaurus, Draceratops, and Monolometerodon needing minor nerfs, that Legendary is currently the “least balanced” tier, at least in terms of the number of dinos outside of the balanced range (though the rare tier is more widespread, but with fewer dinos outside of the balanced range). There are likely a few reasons for this. One is that there are actually quite a few legendaries. Most of them are quite concentrated in the Elite Mid and Elite Low tiers though, so it’s basically the opposite of the rare hybrids, where the standard deviation is quite small. But given the large number of legendaries contributing to a larger sample size, this assessment is more statistically accurate than the one for the rare hybrids.

Uniques and Apexes are still unchanged. Entelolania still needs (at least) a minor buff.

Tldr (though I reccomend that you do before you comment), here is a summary of the results:


What do you think? Does this second analysis seem more accurate than the first? What do you think of the balance rankings? I found it interesting that Marsupial Lion is now the only dino rated as needing a major nerf, though I personally don’t feel like it particularly does (maybe a minor nerf, if anything). Its also worth noting that out of the 226 dinos included in this analysis, only 18 were statistically determined to need some kind of rebalancing. That’s only about 8%! And it’s even less when you exclude raid dinos and others that probably don’t actually need to be rebalanced (this analysis and even the tier list it’s based on don’t capture the entirety of strength in JWA). I know we as a community give Ludia a lot of flack (and oftentimes rightly so), but to me this is pretty dang impressive. I’m sure you’ll let me know your thoughts down below.

22 Likes

Going off of this curve, it makes sense. However, I feel balancing is better done by looking at important matchups that the questionable creature has rather than just using a curve, as while ML, Andrew, and Allosino stick out, they are well balanced within their own rarity.

Take Andrew for example, it crushes its resilient opposition, does well against it’s fellow fierce (Loses to Rex2 if it crits and I think ML), and gets wrecked by the cunnings in its rarity. Dodge does play a role as the cunning in question may survived by a lower margin, specifically Andrew Critting on CS and hitting through dodge and leaving them vulnerable to a swap in (The only cunning actually hurt by that stream of events are Oviraptor, which I think is fine, and the raptor squad, which sadly is irrelevant).

This thinking is how I feel most balancing should be done under the current system, it’s seems inefficient, but by playing different tourneys you really get a feel for what OP/UP, and that data can be used to balance in the future

5 Likes

Yeah I definitely agree that there’s more to balance than the curve suggests by itself. That’s why I put the disclaimer that this analysis doesn’t perfectly translate to what should be balanced all the time.

4 Likes

I think a lot of it has to do with the nature of the tier list too. Taking Andrewsarchus for example, while it punches far above it’s weight, there are some dinos of the same rarity that can punch it right back.

2 Likes

Don’t you dare suggest a brontolasmus nerf. Also how does suchotater need a “major nerf” it can be speed decreased, stunned, bled, etc. Monolometrodon is fine, rixis is fine, marsupial I don’t know because most creatures have rend resistance. Dracoceratops doesn’t need a nerf??? It’s not even that good. Thylaco is fine, just use a creature with rend resistance and it’s over.

Did you actually read the post? I specifically say that 1) I don’t personally agree with all of these, its just objective statistics and 2) that balance is more complicated and 3) that “Major nerf” and “Minor nerf” are more measures of relative strength than whether they ACTUALLY need a major or minor nerf.

Heres what I actually think. Brontolasmus is mostly fine. So is Suchotator. Monolometerodon could use a small nerf, but not much. Rixis needs a slight damage nerf, it’s way too bulky and stalls too well for how much damage it does. But I myself wouldn’t call that a “Major nerf”, it just means that Rixis is far better than most epics. Marsupial is fine, maybe lowering the rend % on its main attack if anything? Dracoceratops probably does need a slight nerf, maybe something to do with cleansing impact. And Thylacotator is fine. I would add that Diplodocus should swap Decelerating Rampage for Decelerating Impact, so it can’t 2-shot chompers anymore.

12 Likes

Great analysis and the results do line up fairly what the community has said and shown needs fixing while its not completely accurate to all types of creatures it’s still very well done. You should honestly tag the mods like @Ned so the developers can see it.

4 Likes

Lol thanks! I also like that it highlighted some buffs that don’t get discussed much, like Koolasuchus. Maybe I will tag the devs. But I would hope they’ve done their own analysis, since they (hopefully) have better data anyway.

3 Likes

Well like @Mudkipz says not everything lines up but there is also arguments to be made that they are too good compared to others of there class. Brontolasmus for example is honestly the fact it has double rampages and very high attack and most fierce creatures besides thyla, acrocanto, baryonyx, edapho and secodonto. The rest lose in its own rarity at least; and while some resilient creatures like rixs can also beat it most also get just destroyed by it. Suchotator is honestly one I have been trying to get the courage to say is broken cause well it kinda is. Think about it a rare easy to make hybrid is in elite mid which supers and unique creatures and it as good and if not better than it’s own hybrid. And its not really the kit that much it does have big moves although instant distraction every other turn is maybe too much I think it’s just more stats I mean compare it to other rare hybrid and there is a big big difference.


Plus their RARE HYBRIDS of course they shouldn’t be good cause they are the first hybrids you can get the game their your starters they should be they endgame creatures. Monolometrodon, sarcorixs, and Draco don’t need big nerf like many and the graph suggests. Monolometrodon just need to lose complete distraction resistance and lower Rend to 50%. Sarcorixs just need to be able to be slowed and have less damage, and Draco just needs to be less survivable or not be able to cleanse distraction or swap prevention. Thyla and lion are fine but I would say thyla doesn’t need superiority strike since speedsters should kinda counter it, and same goes for lion and prowl but idk what could be done about that (but for lion the counter should probably go)

1 Like

With glyptodon and andrew and ML, there are major drawbacks to each creature that doesn’t make them OP. Also, many creatures place high because they work well in the meta, like suchotator and marsupial lion, and they shouldn’t be punished because the meta is in their favor. Also no nerf glyptodon please Ludia

2 Likes

That will kinda ruin it

2 Likes

I think just remove the counter attack so it has to Crit to 2 shot a creature. But mostly cunnings keep it in its place.

1 Like

The counter is an important part of the kit with prowl. You take away so many matchups when you remove the counter. ML is fine

7 Likes

Yeah thats definitely true. Like I said in another comment, this comes from my data being derived from a tier list. These guys get a lot of credit for having good matchups against higher rarities, and that shows in their placement, but that doesn’t mean they’re op.

At the same time though, part of the tier list is based on the gamepress battle simulators, so the number of good matchups these dinos have are significantly better than others in their rarity. Does that make them op? Probably not by itself, but it’s something to think about.

1 Like

Yeah I like @Funtime_dino s suggestion better. And maybe have it so prowl doesn’t dodge 100% of the time. I can’t remember if they fixed that or not.

It isn’t. If anything, evasive strike should be buffed as ML takes advantage of it with the counter

Yeah, it can just spam prowl otherwise. It’s definitely quite counterable otherwise, and a lot of it’s strength is due to the meta. Plus if rend resistance was more common, especially in other rares, it would be taken down a peg. Theres so much rend in the rare tournament this weekend (ML, Andrew, Purrolyth), but nothing really resists it. Maybe Dimetrodon and some rare cunnings should get rend resistance?

2 Likes

Monolometrodon, rixis, and draco don’t need a nerf at all. Brontolasmus can only use its 2nd rampage on turn 2 and its deceleration rampage is not available for 2 turns. Brontolasmus is fine.

Also about suchotater, it has no resistances at all. I mean just use something that is immune to dot and suchotater is basically done. Sucho can be bled, speed reduced, stunned, etc. It barely lives against a lot of resilients and is vulnerable due to being speed reduced.

It’s the perfect resilient, who would want to change that?

1 Like