Dinosaurs and stat distribution trends, what's your experience?

[Azmeaiel]

d calculator for tamed dinos. It's flat out wrong. 

This is on Dododex: "Dododex's stat calculator will calculate how a wild creature's stats were distributed in Ark: Survival Evolved."

This is on the wiki; "Note that after the creature is tamed it gets bonuses on some stats depending on the taming effectiveness. This makes it hard to retrieve the levels on a tamed creature, so this tool is only for wild ones, but gives a first impression, how well the stats are distributed."

And yet so many people keep using it for dinos it's not meant for and then claiming massive stat numbers that are wrong.  I do a lot of breeding so it's a continual frustration sorting fact from fiction when talking to people about their dinos.

Typically something like this; "I have a great Rex with 42 points in melee", which is great for a few minutes until you realise they used the wild calc on a tamed dino and it's only really 31 points.  Sigh.  

 

which is why i clearly defined wether the dino is post-tame, pre-tame or raised...obviously. Noone here is debating, we know..

[WontonFreak]

6 hours ago, Thalantyr said:

Yer unfortunately they treat the figures completely different. DodoDex will say L70, ASB will show L55. Even DodoDex comparing HP to DMG is incorrect. I'd like to see this fixed but I'm sure the authors believe they're doing it correct. Shrug. Either way, getting 75 levels into something is ridiculous. Even 71 levels is right out there. Something along the lines of 52 levels I can imagine as I've seen 45 levels before, but it'd still be a very small chance.

What you could do is to just use the spawn command and instead of using 150, you'd use 147, it should then end up as 221 or so. The base level is lower but the stat budget in the end result should be the same.

[cad]

22 hours ago, WontonFreak said:

Does still not explain the fact that people claim to have dinos with 60+ points when the average shows the exact opposite, i've been using the past 4 hours spawning rexes and forcetaming them, and not once have i seen one with more than around 40 to 42 in other than melee damage. If only we could get the raw formula and intended behaviour from the devs it would be great

also for @Thalantyr (thanks for the ping)

The stats are very probable just random distributed, so the theoretical maximum (with a wildMaxLevel of 150) for a stat is 149 for a wild dino (every creature has level 1, then for each levelup in a stat it gains one total level), 224 for a tamed creature (with 100% TE, which is impossible without cheating) and 1569 for a bred creature (without mutations) (taking the base level of 1, and 224 levels for each of the 7 stats).

The probability to see such a max-dino is practically 0. The probability to have a dino with 50 or more levels in one certain stat is around 1:1250 (https://www.wolframalpha.com/input/?i=probability+of+50+successes+in+224+trials+with+p%3D.14285714285) so I don't believe that people have a dino with 50 or more levels in melee. That said, it's not impossible to even have e.g. 200 levels in melee. BayesianJudo did a post on reddit where they put the probability of finding such a dino in comparison (it is from the time when the max wild level was 120). They write something like if every bacterium on earth would tame every second a max level dino, it takes longer than the universum existed to get such a dino. Give it a read, it gives you kinda an impression how rare such a dino would be.

 

[Thalantyr]

1 hour ago, cad said:

Explanation

Thanks Cad. Always appreciate you dropping in to clear some things up

I don't quite understand everything on that wolframalpha.com link, but get the general jist of it. With that in mind, is there somewhere that shows the chance of getting 49/48/47/46/45 levels in a single stat? Can that formula be changed / reinterpreted to get those probability figures? I'd be interested to know how much effort is involved. Thank you.

[cad]

19 hours ago, Thalantyr said:

Thanks Cad. Always appreciate you dropping in to clear some things up

I don't quite understand everything on that wolframalpha.com link, but get the general jist of it. With that in mind, is there somewhere that shows the chance of getting 49/48/47/46/45 levels in a single stat? Can that formula be changed / reinterpreted to get those probability figures? I'd be interested to know how much effort is involved. Thank you.

If you have Excel, you can use this formula, that gives you the probability of a creature to have at least X levels in a stat (multiply by 100 to get %):

=1-BINOM.DIST(X,224,1/7,TRUE)

If you want the probability to get exactly X levels, change the TRUE to FALSE and remove the 1-.

You also can use wolfram-alpha by changing the 50 in the link I posted before with the level you want to know. Then have a look at the table where it says "more than X successes".

[WontonFreak]

4 hours ago, cad said:

If you have Excel, you can use this formula, that gives you the probability of a creature to have at least X levels in a stat (multiply by 100 to get %):

=BINOM.DIST(X,224,1/7,TRUE)

If you want the probability to get exactly X levels, change the TRUE to FALSE.

You also can use wolfram-alpha by changing the 50 in the link I posted before with the level you want to know. Then have a look at the table where it says "more than X successes".

Ok, so what you're saying is that anything above 50 is "unlikely" in that it's possible but insanely rare. That would mean that something in the lines of 40-45 is a "high tier" stat and should be good to breed, am I right?

[Thalantyr]

4 hours ago, cad said:

If you have Excel, you can use this formula, that gives you the probability of a creature to have at least X levels in a stat (multiply by 100 to get %):

=BINOM.DIST(X,224,1/7,TRUE)

If you want the probability to get exactly X levels, change the TRUE to FALSE.

You also can use wolfram-alpha by changing the 50 in the link I posted before with the level you want to know. Then have a look at the table where it says "more than X successes".

Sorry Cad, I'm a bit fresh on this stuff, but can you please tell me if the green section in the image below is correct in terms of percentages to get a specific level?

YELLOW = Has the TRUE statement, GREEN = Has the FALSE statement.

[cad]

14 hours ago, Thalantyr said:

Sorry Cad, I'm a bit fresh on this stuff, but can you please tell me if the green section in the image below is correct in terms of percentages to get a specific level?

Hi, sorry, I forgot to add a 1- before that expression. The green rows (with FALSE) are correct without it. The correct expression is below. In Excel you then can just mark the results as % for an easier reading.

=1-BINOM.DIST(X,224,1/7,TRUE)

[cad]

14 hours ago, WontonFreak said:

Ok, so what you're saying is that anything above 50 is "unlikely" in that it's possible but insanely rare. That would mean that something in the lines of 40-45 is a "high tier" stat and should be good to breed, am I right?

Yes! Though I'd set the range of levels worth to breed from 35+, it depends how much time you have to look for good dinos.

[Thalantyr]

4 hours ago, cad said:

Hi, sorry, I forgot to add a 1- before that expression. The green rows (with FALSE) are correct without it. The correct expression is below. In Excel you then can just mark the results as % for an easier reading.

=1-BINOM.DIST(X,224,1/7,TRUE)

Thanks @cad. How does the below look?

[PlatinumCore16]

This is fascinating.  Now I'd already read about the astronomical probability of actually getting a dino like that, but it's nice to see an explanation like that one you posted from Reddit.

Now I understand that DodoDex is incorrect for things such as melee due to the post tame multiplier, BUT there is no multiplier for weight.  So using their calculator to track weight is not an issue?

Also I have used it for pretame mostly, so it's not a huge issue for me, also due to the fact I'm only looking into weight.

As far as how high that number is I think one of my thornies got 40 points into stam.

Hoping to find that 50 weight level tame some day...

[TotallyUncool]

On 16/01/2017 at 5:54 AM, Kallor said:

It's not really debatable, you cant use a wild calculator for tamed dinos. It's flat out wrong. 

This is on Dododex: "Dododex's stat calculator will calculate how a wild creature's stats were distributed in Ark: Survival Evolved."

This is on the wiki; "Note that after the creature is tamed it gets bonuses on some stats depending on the taming effectiveness. This makes it hard to retrieve the levels on a tamed creature, so this tool is only for wild ones, but gives a first impression, how well the stats are distributed."

And yet so many people keep using it for dinos it's not meant for and then claiming massive stat numbers that are wrong.  I do a lot of breeding so it's a continual frustration sorting fact from fiction when talking to people about their dinos.

Typically something like this; "I have a great Rex with 42 points in melee", which is great for a few minutes until you realise they used the wild calc on a tamed dino and it's only really 31 points.  Sigh.  

Unfortunately, not all of us can use the ARK: Smart Breeding, so we have to stick to Dododex etc and keep annoying you with our wrong numbers. I wish I could use it, but no luck so far.

[cad]

2 hours ago, Thalantyr said:

Thanks @cad. How does the below look?

Looks good now.

1 hour ago, PlatinumCore16 said:

Now I understand that DodoDex is incorrect for things such as melee due to the post tame multiplier, BUT there is no multiplier for weight.  So using their calculator to track weight is not an issue?

As long as you don't level up the creature it's fine for most of the stats. To be sure, check the wiki, e.g. https://ark.gamepedia.com/Ankylosaurus#Base_Stats_and_Growth, if there is no taming-bonus listed, it's probably fine.

1 hour ago, TotallyUncool said:

Unfortunately, not all of us can use the ARK: Smart Breeding, so we have to stick to Dododex etc and keep annoying you with our wrong numbers. I wish I could use it, but no luck so far.

If you're on linux, you could try mono, people reported it's running there. Not sure it's working on MacOS, though. If you don't have access to any of them, you could try it with a google-spreadsheet, but that's some work to do.

[PlatinumCore16]

Just now, cad said:

...As long as you don't level up the creature it's fine for most of the stats. To be sure, check the wiki, e.g. https://ark.gamepedia.com/Ankylosaurus#Base_Stats_and_Growth, if there is no taming-bonus listed, it's probably fine....

Thanks.  Accruing quite a few breeding tames, I think I'm gonna give a look to that tool @Thalantyr mentioned.  I mean I love making spreadsheets, but yea...

[Kallor]

1 hour ago, TotallyUncool said:

Unfortunately, not all of us can use the ARK: Smart Breeding, so we have to stick to Dododex etc and keep annoying you with our wrong numbers. I wish I could use it, but no luck so far.

Doesn't make sense, given you have a Melee% actual number that you know is correct and a stat point calculation from Dododex the you know is wrong, why would you choose to use the number that's wrong instead of the melee% that's right?

Just quote the melee% and if someone wants to know the point value you can say you don't know.

[cad]

15 hours ago, PlatinumCore16 said:

Thanks.  Accruing quite a few breeding tames, I think I'm gonna give a look to that tool @Thalantyr mentioned.  I mean I love making spreadsheets, but yea...

You can export the data from the tool into a spreadsheet, win-win

[PlatinumCore16]

Haha now now, I'm supposed to get work done today.

[TotallyUncool]

17 hours ago, cad said:

If you're on linux, you could try mono, people reported it's running there. Not sure it's working on MacOS, though. If you don't have access to any of them, you could try it with a google-spreadsheet, but that's some work to do.

I'm on windows and I can get the program running just fine, I just can't add creatures to the library because I get errors when trying to extract their levels/stats. I've already posted about this on Github, and you've replied. This was your last reply in that thread: "The tameAffinity-multiplier only affects melee and food, and for some species it's only melee. So it sounds as if this particular multiplier is wrong in the settings. Can you post a screenshot of a creature that failed to extract and has no domesticated levels yet? Maybe we can work it out with its data."

 

17 hours ago, Kallor said:

Doesn't make sense, given you have a Melee% actual number that you know is correct and a stat point calculation from Dododex the you know is wrong, why would you choose to use the number that's wrong instead of the melee% that's right?

Just quote the melee% and if someone wants to know the point value you can say you don't know.

Habit. We always refer to levels when we talk on the server I play on so that's why I do that, plus I didn't learn about this until recently. Most people seem to use Dododex/wiki so most people know if it's "good" or "bad" anyway if you write about it (although always write the entire stats and not only the levels).

[PlatinumCore16]

13 minutes ago, TotallyUncool said:

Habit. We always refer to levels when we talk on the server I play on so that's why I do that, plus I didn't learn about this until recently. Most people seem to use Dododex/wiki so most people know if it's "good" or "bad" anyway if you write about it (although always write the entire stats and not only the levels).

I would assume most people use DoDoDex/app, less people use SA/app, and even less people know about this external tool, let alone use it.  Maybe would be nice to see some sort of indicator in the UI to whether a stat is "good" or "bad", but at the same time, that may not be a great idea, let alone within scope.  Would also break the immersion.

[Shangalar]

From my own experience, the thing is that the rounded all around stats are mostly due to pre-tame stats. It is all in how the newly tamed stats are distributed that make all the difference after that

Currently, my rex breed got the following points in stats: Health 42.3 Stam 42 Oxy 35 Food 30 Weight 37 Damage 45.2

As you can see, i found so far better stats overall but you have found much higher damage. It's pretty much up to luck, really

And lastly, the highest single stat i have ever found on ANY dinos was a 92 points dumb on stamina on a itchy... Pointless yes, but this itchy got 3060 stam lol

[TotallyUncool]

Just now, PlatinumCore16 said:

I would assume most people use DoDoDex/app, less people use SA/app, and even less people know about this external tool, let alone use it.  Maybe would be nice to see some sort of indicator in the UI to whether a stat is "good" or "bad", but at the same time, that may not be a great idea, let alone within scope.  Would also break the immersion.

I assume this as well and I guess you hear about the tool when you start going insane due to too much breeding and googling on stats etc. in general.

[PlatinumCore16]

1 hour ago, Shangalar said:

From my own experience, the thing is that the rounded all around stats are mostly due to pre-tame stats. It is all in how the newly tamed stats are distributed that make all the difference after that

Currently, my rex breed got the following points in stats: Health 42.3 Stam 42 Oxy 35 Food 30 Weight 37 Damage 45.2

As you can see, i found so far better stats overall but you have found much higher damage. It's pretty much up to luck, really

And lastly, the highest single stat i have ever found on ANY dinos was a 92 points dumb on stamina on a itchy... Pointless yes, but this itchy got 3060 stam lol

You can't get partial levels, so are you saying those numbers are the multiplier?

 

1 hour ago, TotallyUncool said:

I assume this as well and I guess you hear about the tool when you start going insane due to too much breeding and googling on stats etc. in general.

Oh yea, already there.

[WontonFreak]

22 hours ago, PlatinumCore16 said:

Thanks.  Accruing quite a few breeding tames, I think I'm gonna give a look to that tool @Thalantyr mentioned.  I mean I love making spreadsheets, but yea...

Dododex isn't "wrong" if you're just comparing the stats within the same breed, as that's what we mostly use it for anyway.

[tristanl1997]

I know it seems that y'all are talkin bout rexes, but I just hatched about 13 wyverns and four of my 180 amd 175 eggs came out with roughly 5k hp, while the 160 I hatched came out with 9990 hp. The stats are really random so you might wanna check rexes that are even above 100, might have a lot of points in one stat

[PlatinumCore16]

6 minutes ago, tristanl1997 said:

I know it seems that y'all are talkin bout rexes, but I just hatched about 13 wyverns and four of my 180 amd 175 eggs came out with roughly 5k hp, while the 160 I hatched came out with 9990 hp. The stats are really random so you might wanna check rexes that are even above 100, might have a lot of points in one stat

Wyverns are a bit of a different animal (pardon my pun ).  Wyverns are unable to be bred and have a health and damage penalty as opposed to a multiplier as most other dinosaurs do.  You don't really get to look at a wyvern before you tame it.  It hatches and that's what you get.  I suppose if you trolled the scar for long enough you could hatch 50 lvl 150+ and just keep the ones that matched your criteria, but you would never get a 'bloodline' as it were, so like I said they're a bit different.