Jump to content
You need to play a total of 5 battles to post in this section.
Counter_Gambit

Resource Container update (personal data collection)

27 comments in this topic

Recommended Posts

808
[NWNG]
Members
3,070 posts
4,925 battles

I personally have now collected 500 resource containers, since I started recording the information. I am still trying to find a way to create a data collection sheet, that can collect data from all containers, including super containers, but googlesheets simply does not play nice with dependent-drop-down-lists.

Anyway: Counter_Gambit's personal data collection
Here is the information provided by my own data, at the 500 mark, since the link provided above, updates as I put in new information.
 

@iDuckman is also at 500 containers collected, and if he doesn't mind (I'll wait until he says it's a go, or he shares it himself), I could also present that data, to see how ours differ at this point. But for now, you're all stuck with the data I collected.

Quick Key for that which is abbreviated:

Spoiler

FXP=Free Experience
ESCL=Equal Speed Charlie London
HY=Hotel Yankee
NF=November Foxtrot
PP=Papa Papa
JC=Juliet Charlie
ZH=Zulu Hotel

ALL AVERAGES AND PERCENTAGES ROUNDED TO NEAREST HUNDREDTH

Item|Total|Average
Coal|260800|521.60
FXP|151000|302.00
ESCL|513|1.03
HY|78|0.16
NF|228|0.46
PP|207|0.41
Zulu|198|0.40
JC|195|0.39
ZH|219|0.44


Slot 2 Appearance Rate:
Item|Count|Percent
Coal|124|24.80%
FXP|205|41.00%
ESCL|171|34.20%

Slot 3 Appearance Rate:
Item|Count|Percent

Coal|28|5.60%
FXP|97|19.40%
HY|26|5.20%
NF|76|15.20%
PP|69|13.80%
Zulu|66|13.20%
JC|65|13.00%
ZH|73|14.60%

Slot 3 - Signal Flag Total: 375|75.00%

Henceforth, all signal flags from slot 3 will be grouped together as simply "Flag."

Slot 3 Occurrence - based on Slot 2 Outcome
Slot 2 = Coal (124)

Item|Count|Percent

Coal|28|22.58%
FXP|21|16.94%
Flag|75|60.48%

Slot 2 = FXP (205)
Item|Count|Percent
Coal|0|0.00%
FXP|57|27.80%
Flag|148|72.20%

    Slot 2 = ESCL (171)
Item|Count|Percent
Coal|0|0.00%
FXP|19|11.11%
Flag|152|88.89%

Edited by Counter_Gambit
  • Cool 4

Share this post


Link to post
Share on other sites
808
[NWNG]
Members
3,070 posts
4,925 battles

 @iDuckman's collected data

Check previous post for what each abbreviation means, and like last post, all averages and percentages are rounded to the nearest hundredth

Item|Total|Average
Coal|261600|523.20
FXP|139500|279.00
ESCL|516|1.03
HY|96|0.19
NF|210|0.42
PP|231|0.46
Zulu|222|0.44
JC|204|0.41
ZH|222|0.44

Slot 2 Appearance Rate
Item|Count|Percent

Coal|135|27.00%
FXP|193|38.60%
ESCL|172|34.40%

Slot 3 Appearance Rate
Item|Count|Percent

Coal|19|3.80%
FXP|86|17.20%
HY|32|6.40%
NF|70|14.00%
PP|77|15.40%
Zulu|74|14.80%
JC|68|13.60%
ZH|74|14.80%

Slot 3 - Signal Flag Total: 395|79.00%

Henceforth, all signal flags from slot 3 will be grouped together as simply "Flag."

Slot 3 Occurrence - Based on Slot 2 Outcome
Slot 2 = Coal (135)
Item|Count|Percent

Coal|19|14.07%
FXP|22|16.30%
Flag|94|69.63%

Slot 2 = FXP (193)
Item|Count|Percent
Coal|0|0.00%
FXP|34|17.62%
Flag|159|82.38%

Slot 2 = ESCL (172)
Item|Count|Percent

Coal|0|0.00%
FXP|30|17.44%
Flag|142|82.56%


There is something here, that I do find interesting.

Gambit's Coal total: 260800
iDuckman's Coal Total: 261600

He has collected more coal than I have, despite the fact that I had coal show up in slot 3 more times than he did. However, coal showed up enough times in Slot 2 for iDuckman, to counter the difference of our Slot3-coal output.

 

Spoiler

RNG is a fickle thing, but comparing these two data sets, I am almost certain, that the RNG generates the crates using a pseudorandom generator, and uses the time variable(Hours minutes, seconds, possibly milliseconds) as the seed, for the generation of the outcome, due to the fact that it is the only thing guaranteed to be different every time it is called upon. It likely takes the date (Year Month Day) in as part of the seed as well. When the game is asked to deliver (aka generate) a crate, it grabs the Date and Time at the moment it was asked, to then run the pseudorandom number generator, thus producing the outcome. Hypothesis, this would mean, if timed correctly, and on the same server (because it would likely use the server time), two people can produce a container, with the exact same output, consistently. Again, that's a hypothesis, and if anyone is willing to attempt it, please let me know the results (again, it likely goes down to the milliseconds). When purchasing multiple crates from the online store, the Pseudorandom Number Generator, likely produces a long list, to account for multiple containers being generated at once.

Further hypothesis on this pseudorandom generator: When generating the daily containers, the first number generated, is likely what determines if it is a Super Container or not. The second number, is what gets compared to an "output list." We only ever see the output, not the actual numbers generated to come to the output.

 

Edited by Counter_Gambit

Share this post


Link to post
Share on other sites
3,425
[WOLF9]
Privateers
12,618 posts
4,602 battles
6 minutes ago, Counter_Gambit said:

I'll wait until he says it's a go,

Do what you want with it.  It's already public.  I'l interested to see what you find.  I already know about the large discrepancy in multi-Coal drop rate.

I got something new today - a Twitch Supercontainer.

 

Share this post


Link to post
Share on other sites
808
[NWNG]
Members
3,070 posts
4,925 battles
39 minutes ago, iDuckman said:

Do what you want with it.  It's already public.  I'l interested to see what you find.  I already know about the large discrepancy in multi-Coal drop rate.

I got something new today - a Twitch Supercontainer.

 

Didn't know if you wanted to do it yourself, that's all

Share this post


Link to post
Share on other sites
3,425
[WOLF9]
Privateers
12,618 posts
4,602 battles
3 minutes ago, Counter_Gambit said:

Didn't know if you wanted to do it yourself, that's all

Oh, go ahead.  I'm up to my eyeballs in this 狗操的 quest for ammo costs.

 

Share this post


Link to post
Share on other sites
5,549
[PVE]
Members
19,835 posts
11,995 battles
1 hour ago, iDuckman said:

Do what you want with it.  It's already public.  I'l interested to see what you find.  I already know about the large discrepancy in multi-Coal drop rate.

I got something new today - a Twitch Supercontainer.

 

Since @iDuckman is busy, would you be able to elaborate on the "...discrepancy in multi-Coal drop rate"please?

 

Also curious as to what his data shows please.

Share this post


Link to post
Share on other sites
808
[NWNG]
Members
3,070 posts
4,925 battles

Okay, second post edited to hold iDuckman's data. along with a hypothesis behind why are numbers vary.

15 minutes ago, Kizarvexis said:

Since @iDuckman is busy, would you be able to elaborate on the "...discrepancy in multi-Coal drop rate"please?

 

Also curious as to what his data shows please.

The "multi Coal Discrepancy" is what iDuckman and I are calling it.

Because Coal is 100% guaranteed in the first slot, we turn to slots 2 and 3.

Some of the data I haven't shared, can help paint the picture, What I call "Triple Coal" which should be obvious as "400coal/400coal/400coal" he and I have two differing numbers. I have 28, while he has only 19. This gives me a triple coal drop rate of 5.59%, and he a drop rate of 3.80%, when comparing to the total containers collected.

However, due to the fact that we only ever see coal appear in slot 3, when it also appears in slot 2, we have come to the conclusion that slot 3's outcome is linked directly to slot 2. Which is part of the data that I did share, which intensifies the difference of the triple coal drop.

I have a slot 3 coal drop rate of 22.58%, while iDuckman has a slot 3 coal drop rate of 14.07%, when compared to the total coal that appeared in slot 2.

This discrepancy is further compounded on, by the fact that I only have coal showing up in slot 2 124 times, while he has it 135.

I have a 24.75% slot 2 coal drop rate, while he has a 27.00% slot 2 coal drop rate.

Each of these data points on their own, is quite interesting, and where the whole "discrepancy" thing comes from. But all this shows, is that one must analyze the data very carefully. The fact that he has 135 coal in slot 2, and 19 coal in slot 3, gives him a total 154 extra coal drops (I say extra, because slot 1 is always guaranteed coal). Simultaneously, add my 124 slot 2 coal, and my 28 slot 3 coal, I only have 152 extra coal drops. By analyzing the data like this, the discrepancy completely falls apart, putting both of us well within the margin of error, that is expected, when dealing with near true randomness. So much so, that I call it a discrepancy, more for fun, than serious, because it is still interesting that this difference in triple coal, can be so largely different, yet still be a non-issue. And when I say "non-issue" I mean "no bias to one player."


 

Share this post


Link to post
Share on other sites
10,421
[ARGSY]
Members
18,199 posts
12,673 battles
2 hours ago, Counter_Gambit said:

There is something here, that I do find interesting.

Gambit's Coal total: 260800
iDuckman's Coal Total: 261600

He has collected more coal than I have, despite the fact that I had coal show up in slot 3 more times than he did. However, coal showed up enough times in Slot 2 for iDuckman, to counter the difference of our Slot3-coal output.

The difference is 800 coal, i.e. the minimum guaranteed output of two resources containers. I think this falls within the limits of normal variation and is not the basis for any reliable conclusion. I have tracked results from 227 containers so far this year, but my spreadsheet also tracks event and twitch container contents (not premium containers), so there is a fair bit of dross to wade through.

ETA: Ninja'd by @Counter_Gambit, who did indeed draw a conclusion - a correct null hypothesis IMO.

Edited by Ensign_Cthulhu

Share this post


Link to post
Share on other sites
808
[NWNG]
Members
3,070 posts
4,925 battles
Just now, Ensign_Cthulhu said:

The difference is 800 coal, i.e. the minimum guaranteed output of two resources containers. I think this falls within the limits of normal variation and is not the basis for any reliable conclusion. I have tracked results from 227 containers so far this year, but my spreadsheet also tracks event and twitch container contents (not premium containers), so there is a fair bit of dross to wade through.

I likely just posted my response to kiz, when you posted this. Which is funny. Go through my response to Kiz, and you will see that I do make mention of where the discrepancy appears, and why it really isn't one.

Share this post


Link to post
Share on other sites
10,421
[ARGSY]
Members
18,199 posts
12,673 battles
2 minutes ago, Counter_Gambit said:

and why it really isn't one.

Yep, like you said - ninja'd. If I'd read your post before hitting post on my post, I most likely would not have posted. :Smile_teethhappy:

Share this post


Link to post
Share on other sites
808
[NWNG]
Members
3,070 posts
4,925 battles
1 minute ago, Ensign_Cthulhu said:

Yep, like you said - ninja'd. If I'd read your post before hitting post on my post, I most likely would not have posted. :Smile_teethhappy:

you and I were likely typing at the same time, and very well could have posted within milliseconds of each other. So I don't blame you for not reading before posting.


I should mention though:

Because of the large differences in these numbers, Common Sense screams "there's a discrepancy," even though the data itself, shows that there is no discrepancy.

So me calling it a discrepancy, is more "discrepancy in common sense."

Share this post


Link to post
Share on other sites
5,549
[PVE]
Members
19,835 posts
11,995 battles
30 minutes ago, Counter_Gambit said:

Okay, second post edited to hold iDuckman's data. along with a hypothesis behind why are numbers vary.

The "multi Coal Discrepancy" is what iDuckman and I are calling it.

Because Coal is 100% guaranteed in the first slot, we turn to slots 2 and 3.

Some of the data I haven't shared, can help paint the picture, What I call "Triple Coal" which should be obvious as "400coal/400coal/400coal" he and I have two differing numbers. I have 28, while he has only 19. This gives me a triple coal drop rate of 5.59%, and he a drop rate of 3.80%, when comparing to the total containers collected.

However, due to the fact that we only ever see coal appear in slot 3, when it also appears in slot 2, we have come to the conclusion that slot 3's outcome is linked directly to slot 2. Which is part of the data that I did share, which intensifies the difference of the triple coal drop.

I have a slot 3 coal drop rate of 22.58%, while iDuckman has a slot 3 coal drop rate of 14.07%, when compared to the total coal that appeared in slot 2.

This discrepancy is further compounded on, by the fact that I only have coal showing up in slot 2 124 times, while he has it 135.

I have a 24.75% slot 2 coal drop rate, while he has a 27.00% slot 2 coal drop rate.

Each of these data points on their own, is quite interesting, and where the whole "discrepancy" thing comes from. But all this shows, is that one must analyze the data very carefully. The fact that he has 135 coal in slot 2, and 19 coal in slot 3, gives him a total 154 extra coal drops (I say extra, because slot 1 is always guaranteed coal). Simultaneously, add my 124 slot 2 coal, and my 28 slot 3 coal, I only have 152 extra coal drops. By analyzing the data like this, the discrepancy completely falls apart, putting both of us well within the margin of error, that is expected, when dealing with near true randomness. So much so, that I call it a discrepancy, more for fun, than serious, because it is still interesting that this difference in triple coal, can be so largely different, yet still be a non-issue. And when I say "non-issue" I mean "no bias to one player."


 

 

18 minutes ago, Counter_Gambit said:

you and I were likely typing at the same time, and very well could have posted within milliseconds of each other. So I don't blame you for not reading before posting.


I should mention though:

Because of the large differences in these numbers, Common Sense screams "there's a discrepancy," even though the data itself, shows that there is no discrepancy.

So me calling it a discrepancy, is more "discrepancy in common sense."

That is interesting. Thanks for the explanation. :Smile_honoring:

 

I would guess that it is like getting a streak of heads when flipping a coin, but since slot 3 depends on slot 2, you are right that it's not really a discrepancy at all.

Share this post


Link to post
Share on other sites
3,425
[WOLF9]
Privateers
12,618 posts
4,602 battles
54 minutes ago, Counter_Gambit said:

Each of these data points on their own, is quite interesting, and where the whole "discrepancy" thing comes from. But all this shows, is that one must analyze the data very carefully. The fact that he has 135 coal in slot 2, and 19 coal in slot 3, gives him a total 154 extra coal drops (I say extra, because slot 1 is always guaranteed coal). Simultaneously, add my 124 slot 2 coal, and my 28 slot 3 coal, I only have 152 extra coal drops. By analyzing the data like this, the discrepancy completely falls apart, putting both of us well within the margin of error, that is expected, when dealing with near true randomness. So much so, that I call it a discrepancy, more for fun, than serious, because it is still interesting that this difference in triple coal, can be so largely different, yet still be a non-issue. And when I say "non-issue" I mean "no bias to one player."

Excellent analysis.  I han't thought about it that way.  Thanks!

 

Share this post


Link to post
Share on other sites
808
[NWNG]
Members
3,070 posts
4,925 battles
1 hour ago, Kizarvexis said:

 

That is interesting. Thanks for the explanation. :Smile_honoring:

 

I would guess that it is like getting a streak of heads when flipping a coin, but since slot 3 depends on slot 2, you are right that it's not really a discrepancy at all.

Glad I could be of help.

 

1 hour ago, iDuckman said:

Excellent analysis.  I han't thought about it that way.  Thanks!

 

No problem. The interesting part about comparing our data, is seeing the large differences in our numbers, and trying to ignore our common sense screaming "there's a discrepancy." And I am positive, if we reduced our data collection, to match that of another, we would still see large variations, in the numbers, that screams "discrepancy" but actually isn't... in fact, I think I'll do just that for the sake of it, just to see what I find.

Share this post


Link to post
Share on other sites
3,425
[WOLF9]
Privateers
12,618 posts
4,602 battles
2 hours ago, Ensign_Cthulhu said:

I have tracked results from 227 containers so far this year, but my spreadsheet also tracks event and twitch container contents (not premium containers), so there is a fair bit of dross to wade through.

As does mine.  If you have a header row (column labels), highlight it and select Data->Filter->AutoFilter.  Then select a button and filter on the contents of the column you're interested in, in this case container type.

850 entries and I can pull out the two Steam containers with the click of a button.

 

Share this post


Link to post
Share on other sites
808
[NWNG]
Members
3,070 posts
4,925 battles

For the sake of it, I dropped the count temporarily, so that my data, @iDuckman's data, and @LT_Rusty_SWO's data are equal. And though looking at the data as a whole, there isn't a discrepancy, but common sense, to a degree is screaming that there is one.

Slot 3 Coal result, compared to slot 2 coal output, at 333 total containers collected

My Data:
Slot 2 Coal=81
Slot 3 coal=17

Percentage 20.99%

iDuckman's Data:
Slot 2 Coal=95
Slot 3 Coal=13

Percentage 13.68%

LT_Rusty_SWO's Data:
Slot 2 Coal=80
Slot 3 Coal=18

Percentage 22.50%

Mine and LT's are relatively equal to one another, while there appears to be discrepancy with iDuckman's.

But that would be cherry picking (which common sense attempts to do, because it's easier). Looking at it a bit more closely,

LT and I both have 98 extra coal drop, while iDuckman has 108 extra coal drop.

That's 10 extra coal drops, in favor of iDuckman. Which on the surface, looks like it's huge, but really isn't. All this really shows, is why one needs enough data points to compare, before coming to conclusion.

That was a fun little side project.

Share this post


Link to post
Share on other sites
1,257
[XBRTC]
Members
3,154 posts
10,011 battles

Might be worth looking at this not only in aggregate numbers, but also in terms of date and containers per day. Does the max-coal chance change on the third container vs the first one per day, etc.

Share this post


Link to post
Share on other sites
3,425
[WOLF9]
Privateers
12,618 posts
4,602 battles

Empirically, it seems that the drop rate of various things, notably ESCL, does seem to change over time.  But I bet it doesn't really.  I'll bring my data up to date.

Updated.  Three more sets of triple coal to bring my rate up to .. tada .. 4.10%.

Edited by iDuckman

Share this post


Link to post
Share on other sites
5,549
[PVE]
Members
19,835 posts
11,995 battles
1 hour ago, Counter_Gambit said:

For the sake of it, I dropped the count temporarily, so that my data, @iDuckman's data, and @LT_Rusty_SWO's data are equal. And though looking at the data as a whole, there isn't a discrepancy, but common sense, to a degree is screaming that there is one.

Slot 3 Coal result, compared to slot 2 coal output, at 333 total containers collected

My Data:
Slot 2 Coal=81
Slot 3 coal=17

Percentage 20.99%

iDuckman's Data:
Slot 2 Coal=95
Slot 3 Coal=13

Percentage 13.68%

LT_Rusty_SWO's Data:
Slot 2 Coal=80
Slot 3 Coal=18

Percentage 22.50%

Mine and LT's are relatively equal to one another, while there appears to be discrepancy with iDuckman's.

But that would be cherry picking (which common sense attempts to do, because it's easier). Looking at it a bit more closely,

LT and I both have 98 extra coal drop, while iDuckman has 108 extra coal drop.

That's 10 extra coal drops, in favor of iDuckman. Which on the surface, looks like it's huge, but really isn't. All this really shows, is why one needs enough data points to compare, before coming to conclusion.

That was a fun little side project.

 

1 minute ago, iDuckman said:

Empirically, it seems that the drop rate of various things, notably ESCL, does seem to change over time.  But I bet it doesn't really.  I'll bring my data up to date.

 

@iDuckman is secretly sending the bots Java through the PTS, so he gets better numbers. :Smile_teethhappy:

 

Seriously, I am interested in the results.

Share this post


Link to post
Share on other sites
178
[-RNG-]
Members
527 posts
12,321 battles

What was your SuperContainer drop rate?  I did a tracking earlier this year and over a couple hundred TYL drops, I averaged a 5% SC rate.

Share this post


Link to post
Share on other sites
3,425
[WOLF9]
Privateers
12,618 posts
4,602 battles
29 minutes ago, TheCompGuy25 said:

What was your SuperContainer drop rate?  I did a tracking earlier this year and over a couple hundred TYL drops, I averaged a 5% SC rate.

We weren't tracking SCs.  That analysis has been done elsewhere, and your 5% rate is a mite high.  For all-TYL it runs about 3.5% long-term.

But I for one don't do TYL because while your SC rate is higher, the total items received is much lower.

 

Share this post


Link to post
Share on other sites
2,669
[TARK]
Members
5,174 posts
2,126 battles
32 minutes ago, iDuckman said:

We weren't tracking SCs.  That analysis has been done elsewhere, and your 5% rate is a mite high.  For all-TYL it runs about 3.5% long-term.

But I for one don't do TYL because while your SC rate is higher, the total items received is much lower.

 

How much was the SC drop rate on resource container selection?

Share this post


Link to post
Share on other sites
3,425
[WOLF9]
Privateers
12,618 posts
4,602 battles
1 minute ago, Daniel_Allan_Clark said:

How much was the SC drop rate on resource container selection?

Half of TYL, or 1.75%. 

Which I'm starting to doubt because I haven't gotten one since Jan.

 

Edited by iDuckman

Share this post


Link to post
Share on other sites

  • Recently Browsing   0 members

    No registered users viewing this page.

×