Which company had the most diverse 2021 lineup?

[Halichoeres]

People obviously have many different kinds of collections, but if you're like me and broadly interested in past biodiversity, you might wonder who's offering the widest variety of creatures. I've come up with a way to quantify it, introduced in late 2019: https://dinotoyblog.com/forum/index.php?topic=8305.0

The basic process is two steps:

Step 1: Put every taxon released by a company into a time-scaled cladogram, and then just add up the branch lengths in millions of years. This essentially measures the cumulative evolution required to produce every species in the tree starting from their common ancestor. The diagram below shows the 2020 releases from Creative Beast, with the red arrow indicating the (always hypothetical) common ancestor.



Step 2: Because Step 1 tends to disproportionately reward more recent taxa, we add a temporal bonus. If Company A made a gorgonopsid and a creodont, and Company B made a creodont and an amphicyonid, Company B would have a higher score by the method of Step 1, even though intuitively it seems like a company that makes a Paleogene animal and a Permian animal has produced a more diverse lineup. So we add a bonus for every period that a company represents, measured as the time from the period's end to the present:

0     Quaternary     
2.6  Neogene         
23   Paleogene
66   Cretaceous
145  Jurassic     
201.3 Triassic           
251.9 Permian         
298.9 Carboniferous   
358.9 Devonian         
419.2 Silurian         
443.8 Ordovician   
485.4 Cambrian   
541    Ediacaran   

So the Creative Beast lineup above would get a 66-million year bonus, which isn't much, unsurprisingly considering they have historically had a very narrow focus at any given time (that might be changing, though!).

You can read the 2019 and 2020 versions of this thread for detailed methodology or to calibrate your guesses, but I'm interested: which companies do you think had the most diverse lineups? I'm building the trees now, and I'll reveal the answers in a couple of days.

[Faelrin]

I guess it depends on the contenders. Is O @Oammararak 's stuff being counted for this year for example?

I think the only other ones with substantial releases have been CollectA, Creative Beast Studio (though all are ceratopsians), Eikoh, Mattel, and PNSO. Of those I think CollectA, PNSO, and Eikoh were the most diverse, unless we include O @Oammararak's miniatures, though while all are from the Paleozoic (Ediacaran-Devonian), there was definitely a great deal of first timers, including some from groups that never got any figures prior, particularly from several invertebrate groups (that I'm aware of anyways). Also the only one of those that has released plant figures for the year.

[Halichoeres]

I guess it depends on the contenders. Is O @Oammararak 's stuff being counted for this year for example?

Oh, that's an excellent question. In my head I've been thinking of them as 2022 items because I haven't received mine yet, but Europeans and Canadians have, so I guess they count as 2021. However, I don't plan to build a tree for the Life game miniatures for a couple of reasons. First, it's an enormous task, with 268 taxa. Even if the phylogenetic position of every species were well resolved, that would take a long time. But it's compounded by the fact that many, especially the Ediacaran biota, have very uncertain placement--it's hard to extract character data from films of snot on the seafloor. Moreover, whereas for any dinosaur you can go to Wikipedia and get a ton of references for taxonomy, including conflicting cladograms from different analyses, you generally don't have that rabid fandom for echinoderms, brachiopods, mollusks, and arthropods. So to build it I'd have to download dozens of primary sources and try to stitch them together. It's a geometrically more difficult task than, say, Mattel's tree.

That said, the sheer number of taxa, the fact that it covers the Ediacaran and entire Paleozoic, and the fact that it samples from about half of all known animal phyla in addition to several plants mean that the Life game would be the winner by a factor of several. There is no way that tree is less than 10 billion years in length. So I guess here we're trying to determine the first few runners-up.

Anyway, thanks for your guesses! I don't know all the answers yet but I think you've identified good contenders. Maybe some of the rubber blind bag manufacturers could give them a run for their money, but those figures are so half-assed I'm not sure they should get credit.

[Faelrin]

Thanks for the in depth response. Yeah I imagine that would require a ton of work for the Life game minis, so I definitely cannot blame you for leaving that one out, and like you said about the confusing placement for many things, lack of dedicated info, etc, which I came across back in 2020 I think when compiling lists of all those during or after the kickstarter.

Anyways I'm definitely eager to see who will come out on top in regards to the others.

[bmathison1972]

Oh this is fun!!! I think CollectA might score high as they have invertebrates, fish, and mammals. PNSO might be stronger than usual with more fish diversity this year.

[suspsy]

PNSO also had one prehistoric mammal in 2021.

[bmathison1972]

PNSO also had one prehistoric mammal in 2021.

Oh yeah! Livyatan! Forgot about that one (which is funny as it sits on my shelves LOL)

[Halichoeres]

There's a company that scored surprisingly well that nobody has mentioned. Any more guesses? I'm almost done working up the results.

[Lynx]

I think CollectA, Papo, or PNSO had the most diverse lineup. Papo because it released a shark and prehistoric mammal, but no dinosaurs. However, there were only two figures released, so then comes CollectA, who released a surprisingly wide range of prehistoric creatures! Then PNSO, who probably released the most figures of diversity. Both ocean and land creatures, though sadly not much else. Mostly Hadrosaurs, Therapods, and marine reptiles. So I personally think CollectA won as for mainstream companies.
I think Mattel also did fairly well with diversity!

[Halichoeres]

Both ocean and land creatures, though sadly not much else.

I feel like that covers most things! Well, except freshwater...


Edited to add: Papo is actually punching above their weight this year; despite only having two offerings that I don't find particularly compelling, there is an awful lot of phylogenetic distance between the two.

[Halichoeres]

Also, I have to thank @Faelrin for the excellent Mattel reference thread here: https://dinotoyblog.com/forum/index.php?topic=9974.0

Made this so much easier!

[Faelrin]

@Halichoeres I am glad that has helped you with this project! I hope my 2021 thread helped as well. That covers a bit more ground too, since the other thread linked was just focusing on mainline stuff, but the 2021 thread doesn't have a convenient text list summary of every species for the year just yet. I guess I'll go work on that, and add it here as well.

[bmathison1972]

There's a company that scored surprisingly well that nobody has mentioned. Any more guesses? I'm almost done working up the results.

Trilobiti Design?

[Halichoeres]

There's a company that scored surprisingly well that nobody has mentioned. Any more guesses? I'm almost done working up the results.

Trilobiti Design?

B @bmathison1972 no, but good guess! I actually hadn't calculated for Trilobiti, but I did so and the corrected length is 2.413 billion years, respectable for a six-taxon tree, but not in the top 3. I'm not too confident in that result, of course, because it's an Ediacaran-heavy lineup, which means the topology and scale of the tree are both highly uncertain, and so the estimate is sensitive to alternate placements.

[bmathison1972]

I was thinking more about this today and then it dawned on me that Eikoh must have done well, with a shark (albeit Cenozoic), turtle, Permian synapsids, birds, and mammals, including a primate!

[Halichoeres]

I was thinking more about this today and then it dawned on me that Eikoh must have done well, with a shark (albeit Cenozoic), turtle, Permian synapsids, birds, and mammals, including a primate!

B @bmathison1972 also a good guess! See below

OK, results time! Below each tree (cladogram) I'm going to post three metrics, all in Ma (millions of years). They're reported to the nearest million.
raw length: the actual length of the tree
adjusted length: the corrected length of the tree after bonuses are applied for every geological period represented
LCA age: the estimated age of the Last Common Ancestor of all taxa in the tree. This should help give you a sense of scale when comparing trees; it's how far back in time the earliest (leftmost) node in the tree is.

Papo:

raw length: 846
adjusted length: 849
LCA age: 430
OK, this looks kind of silly. I didn't find this an interesting set of releases, but interesting-ness is subjective. Tree length is objective, and this is objectively a pretty long tree for only two species, almost as long as you can get with two vertebrates!

Safari:

raw length: 160
adjusted length: 226
LCA age: 176
This year was a bummer from Safari from a biodiversity perspective. An understandable bummer, but a bummer. I didn't give credit for the Dino Dana repaints, which I estimate would roughly triple this.

Eikoh:

raw length: 1,639
adjusted length: 2,319
LCA age: 431
It's easily a member of the 1 billion club, but I was honestly expecting a bigger number here! It turns out that placental mammals are not very distantly related to each other, and over half this set was placental mammals.

Mattel:

raw length: 2,006
adjusted length: 2,418
LCA age: 265
This is an awful lot of taxa! But most of them are just variations on a handful of themes. They're literally all Mesozoic sauropsids. Also, Mattel is getting credit for repaints here because I just couldn't bear the thought of trying to figure out which were repaints and which were retools or resculpts. However, it doesn't include products like the minis, so it might be kind of a wash.

And now the top 3:

#3 Takara Tomy:

raw length: 2,210
adjusted length: 3,054
LCA age: 736
A real sleeper! Takara Tomy was one of only a couple of companies to offer an invertebrate this year. I haven't included licensed products, which would bump it up a bit but probably not much.

#2 CollectA:

raw length: 3,075
adjusted length: 3,286
LCA age: 736
Unsurprisingly, CollectA did quite well, thanks to an invertebrate, an actinopterygian, and a mammal. They're getting credit for their miniatures, on the logic that they are new molds and not just repaints. But because the miniatures are all dinosaurs and pterosaurs, excluding them would only knock CollectA down one place, behind Takara Tomy but still ahead of Mattel.

#1 PNSO:

raw length: 2,509
adjusted length: 3,535
LCA age:
The big winner! No invertebrates, but the vertebrate lineages that they sampled were so distinct from one another that it ended up being a really diverse lineup anyway. And of course they just released a ton of product last year, something for basically everyone, although I imagine few people buy everything they make. CollectA actually had a slightly longer tree because they had a protostome, but they sampled fewer time periods than PNSO. CollectA had zero Triassic or Paleozoic offerings this year, but PNSO surprised me by having both!


And tree length plotted as a function of the number of taxa produced by a given company (in the bottom left, EoFauna and Wild Past had identical lengths so they got smushed together). The farther a company is above the line, the more efficiently they're covering the tree of life: Takara Tomy looks really good by this metric. The farther below the line a company falls, the more same-y their releases are, phylogenetically speaking. Last year, PNSO was way below the line, and I said, "they're unlikely to be my favorite company any time soon. They make some pretty nice figures, but they rarely take a chance except for the miniatures, sticking to two periods (the Cretaceous and a sprinkling of the Jurassic) and overwhelmingly two major clades." Well, I have to say they've branched out quite a lot in the last year, making me eat my words. My hat is off to PNSO! I hope the trend continues and that lots of different organisms get the kind of loving attention they've historically given mostly to dinosaurs. 

Now, @Faelrin mentioned the Life game miniatures by O @Oammararak (https://www.kickstarter.com/projects/tragoonha/life-the-evolution-of-life-on-earth-series). No way am I going to build a 268-taxon tree, especially one in which many taxa have highly uncertain phylogenetic affinities. But I pulled 27 representatives and made a skeleton tree.

raw length: 7,272
adjusted length: 9,819
LCA age: 1,496
This is 10% of the taxa, but of course the full tree wouldn't be ten times as long because I deliberately tried to sample all the major clades. Nevertheless, the full tree would be several billion years longer: this game included a staggering variety of arthropods and vertebrates in particular. This one release is more diverse than most companies' lineups for the entirety of their existence.

If there's another company you'd like to see a tree for, let me know! Next time I do this I'm going to try to print the trees directly onto a geological time scale; might make it easier to interpret them.

[bmathison1972]

Enjoying every second of this. Nice to see my original thoughts of CollectA and PNSO panned out well (although I didn't expect PNSO to be first)

[Gwangi]

Thank you for doing these, quite fascinating!

[Faelrin]

Thanks for sharing the results of this. I appreciate the work that goes into these.

[Dusty Wren]

This is fascinating stuff! Thanks for putting it all together, and taking the time to explain it.

Consider me shocked that PNSO took the top spot. Thinking back at their 2021 releases, I think my mind goes to the glut of theropods they released early in the year and it skews my perception of the diversity of their lineup. I thought for sure that CollectA would take it, but that's why we need to crunch the data.