1 2 > 
Successful 0% shuttle missions: Failed at failing
Posted: 13 August 2017 12:42 AM   [ Ignore ]  
Avatar
Cadet
Rank
Total Posts:  20
Joined  2017-04-22

I need 2* vulcan meditation lamps and during all previous faction event (federation) not a single one dropped. So I continued running federation missions, with all the lower than expected successes that come with the honored reputation.
But reading everywhere to just give up and accept that I will lose some, that probably I should lose some on purpose to make it easier in the future is just not good enough of an explanation.

At a 1600, then 1700 difficulty level I ran 29 shuttle missions, with an average of 88,5% success rate displayed, I succeded 21 times, for a 72,4% success rate, and no 2* vulcan meditation lamps (is there a drop rate page for shuttle missions?).

So I decided to lower my difficulty level by assigning unleveled crew. I sent 3 shuttles at 29%, 33%, 20%. First 2 failed, the third succeded, highly improbable, but that happened.
I opened 3 new missions and the difficulty level was still the same, again at 29%, 26%, 19% I failed the first 2 and succeded the third one. ok, highly improbable but not impossible.
I opened again 3 missions, this time my difficulty level dropped from 1700 to 1600, not much to have some improvement in the following missions. So I lowered my bar even more. Unleveled crew was not enough, I put some crew with wrong skill to get a perfect 0% success rate, so I should have failed for sure, and maybe drop my difficulty level faster in the process.

The first shuttle that came back failed, the last two, succeded!

the pictures are from the “rescue citizen captives” (DIP or CMD: klingon torres - DIP: Selar - DIP: drone seven) and “measure temporal anomaly” (ENG:saavik - ENG or SCI: chakotay- SCI: klingon quark)

I didn’t take screenshots before sending them because there was no reason to. So I thought.

I’ll pause here to explain the difficulty level I’m ranting about: I adapted the probability success rate formula that can be found here: https://stt.wiki/wiki/Faction_Missions

Success rate probability = (1/(1+(EXP(3,5 * (0,5-(AvgSkill/Difficulty)))))) * 100

AvgSkill = sum(max_base_stat1 + 0,25 min_base_stat1 + skill_boost) * Event_bonus + .. + max_base_statn + 0,25 min_base_statn + skill_boost) * Event_bonus)
skill_boost is applied wether the crew has the skill or not, it is applied to the slot

Difficulty = 2/(vp_level * n_crew), where
  n_crew is the number of crew required for the mission
  vp_level is exactly the victory point you’re playing for in faction events, and an unknown integer in regular missions that have to be calculated by approximation.

The formula range from 14,8% circa for AvgSkill = 0, tending to 100 for AvgSkill that tend to infinite.
When the crew is unleveled the minimum is 14,8, when wrong skill crew is selected it shows a perfect 0%, probably to avoid a division by zero in the coding.

So, how did I succeded with a 0% success rate shuttle, probably the 14,8% minimum is still in place, probably rngesus is testing my skepticism.

So, my difficulty level was still 1600 and I wanted to repeat the failed failure situation. I ran other 30 shuttles, failing all but one:

and in the process I dropped to 900 difficulty level and I just sent my shuttles with the right crew with the success rate of 99%, 96% and 97%, will see if rngesus approves of my doing and how many shuttles should I run before getting back at a higher difficulty level. and if I’ll get a 2* vulcan meditation lamp faster with a shuttle mission, the faction store, the replicator, or by exchange: https://forums.disruptorbeam.com/stt/viewthread/67277/#365441

To those who’ll imply that the pictures are fake, it’s not impossible.
Edited as this forum doesn’t agree with not spam.

[I try not to spam more than usual]

Profile
 
 
Posted: 13 August 2017 02:40 AM   [ Ignore ]   [ # 1 ]  
Avatar
Ensign
RankRankRank
Total Posts:  118
Joined  2013-04-28

you must be new around here, if you actually expect the RNG to work smile))

I just got my streak broken by a 3/3 crit from a 5%...that’s at least the 5th time i have encountered this situation since i began playing…around 2500 rounds…and the odds of that happening is 1 in 8000 for one occurrence

And yet…i have never won a guinan, locutus or a top row award on dabo smile)

Profile
 
 
Posted: 13 August 2017 04:07 AM   [ Ignore ]   [ # 2 ]  
Avatar
Cadet
Rank
Total Posts:  20
Joined  2017-04-22
She wants the… - 13 August 2017 02:40 AM

you must be new around here, if you actually expect the RNG to work smile))

I just got my streak broken by a 3/3 crit from a 5%...that’s at least the 5th time i have encountered this situation since i began playing…around 2500 rounds…and the odds of that happening is 1 in 8000 for one occurrence

And yet…i have never won a guinan, locutus or a top row award on dabo smile)

Nothing to do with gauntlet.

I’m reporting an extreme case scenario, where rng may or may not be involved, depending if the server side formula differs from the displayed formula. Still, a perfect 0% is different than a low percentage > 0.
I already gave some data that outline that displayed success rate differs from the actual success rate, but as for the faction I’m monitoring in this thread, the dataset is just too little to imply anything.
the data I collected in the last factions events reflects (except the last buggy one) the displayed rate.
But this is just part of the reporting. I also gave some data on how many shuttles to run and fail to lower from a starting difficult level to another.

 

Profile
 
 
Posted: 13 August 2017 04:14 AM   [ Ignore ]   [ # 3 ]  
Avatar
Ensign
RankRankRank
Total Posts:  118
Joined  2013-04-28
Capt. Reynolds - 13 August 2017 04:07 AM
She wants the… - 13 August 2017 02:40 AM

you must be new around here, if you actually expect the RNG to work smile))

I just got my streak broken by a 3/3 crit from a 5%...that’s at least the 5th time i have encountered this situation since i began playing…around 2500 rounds…and the odds of that happening is 1 in 8000 for one occurrence

And yet…i have never won a guinan, locutus or a top row award on dabo smile)

Nothing to do with gauntlet.

I’m reporting an extreme case scenario, where rng may or may not be involved, depending if the server side formula differs from the displayed formula. Still, a perfect 0% is different than a low percentage > 0.
I already gave some data that outline that displayed success rate differs from the actual success rate, but as for the faction I’m monitoring in this thread, the dataset is just too little to imply anything.
the data I collected in the last factions events reflects (except the last buggy one) the displayed rate.
But this is just part of the reporting. I also gave some data on how many shuttles to run and fail to lower from a starting difficult level to another.

 

Yes it does! They didn’t create separate algorithms…one for crits, one for proficency, one for shuttles, one for pips, one for dabo etcetera…it’s the same program working for all of them!

Profile
 
 
Posted: 13 August 2017 01:54 PM   [ Ignore ]   [ # 4 ]  
Avatar
Ensign
RankRankRank
Total Posts:  220
Joined  2017-02-16

With the bug from the last shuttle event, I tried to send out all 4 shuttles with a 0% chance of success and two of them succeeded. It is never really a full 0% chance because of hidden skills. Also I have failed 100% success chance before.

Profile
 
 
Posted: 13 August 2017 04:13 PM   [ Ignore ]   [ # 5 ]  
Avatar
Cadet
Rank
Total Posts:  20
Joined  2017-04-22
Dinoss[CSF] - 13 August 2017 01:54 PM

With the bug from the last shuttle event, I tried to send out all 4 shuttles with a 0% chance of success and two of them succeeded. It is never really a full 0% chance because of hidden skills. Also I have failed 100% success chance before.

While a 100% success rate is a round up number, because the formula used is a curve presumably to prevent an over than 100% success rate with high AvgSkill, from the mathematical point of view it’s not impossible to fail a 100% shuttle mission.

But when the formula result of a wrong skill crew (14,8%) is substituted with a 0%, it’s clearly not a round of a percentage. So when an exception is elaborated it is expected to be implementend on every part of the coding.

About the unlisted skills, I thought of that, but didn’t look much in the wiki before. One of my fleetmates also suggested the same thing.
From the wiki Sito at level 100 has an unlisted skill of 25
Running an away mission I see that GI Chakotay at level 31 has 8 (2-6)
and Rogue Kim at level 5 has 5 (1-2)

for a total of 38 for AvgSkill, that would have brought the success rate from 14.8% to 15.96% at the time I ran that mission at a difficulty level of 1000. Still a really low percentage, and still not the point.

This example clearly shows that display rate differs from the one calculated for the actual result of the shuttle mission, because the display rate is a number placed and not calculated.
While in other cases we can just collect data and speculate that the displayed rate is not faithful to the actual results.
And to be pedantic, and avoid irrational responses: still this example does not prove all the other cases where display rate differs from the actual rate.

Also, we have nothing to show to prove that unlisted skills are used or not in the final count. But if that was the case, the display rate should be equal or lower than the actual rate as it would effect the AND slots where we’re using crew that has only one of the skills required. Also, an unlisted skill calculated as a second skill in the AND function count only for 0,25. Not much of a dent to make speculation in that direction. (and we should be more successful than we actually are)

So for me, I can only speculate that all missions have a minimum success rate of 14.8% whether or not it is displayed as such or not.
I can’t speculate on the actual rate of success formula, even with data from a few hundreads event and regular shuttle missions.

I started collecting data because during the events my success rate was way lower than the display rate, like it was stated many times in colourful ways in this forum. During the klingon mega event I believe the formula was adjusted at some point and the display rate began to be faithful to the actual rate, or maybe I was just lucky throughout the events.
With last event, they changed the formula again , the lucky buggy one, and all the missions were more difficult by around 5%, and that was clearly noticeable at 2750 and 3500 vp, that looked like a way to prevent to low level players to overcome those thresholds. By the time I reached 4000vp the buggy formula was reversed to the previous event, so I don’t know if in the beginning of the event the 4000vp missions were more difficult or not.

 

Profile
 
 
Posted: 13 August 2017 07:35 PM   [ Ignore ]   [ # 6 ]  
Avatar
Commander
RankRankRankRankRankRank
Total Posts:  1058
Joined  2016-03-16
She wants the… - 13 August 2017 04:14 AM

Yes it does! They didn’t create separate algorithms…one for crits, one for proficency, one for shuttles, one for pips, one for dabo etcetera…it’s the same program working for all of them!

Source?  Do you work for DB?

Profile
 
 
Posted: 14 August 2017 02:07 AM   [ Ignore ]   [ # 7 ]  
Avatar
Ensign
RankRankRank
Total Posts:  118
Joined  2013-04-28
milamber42 - 13 August 2017 07:35 PM
She wants the… - 13 August 2017 04:14 AM

Yes it does! They didn’t create separate algorithms…one for crits, one for proficency, one for shuttles, one for pips, one for dabo etcetera…it’s the same program working for all of them!

Source?  Do you work for DB?

It’s called common sense! It would be far too expensive to create more than one algorithm program to do the same task. Plus, it would be overkill.

But I love the way apologists need proof for the arguments against, but not for the ones that support their own view smile)

LE. And you already know that if I knew something from DB, from support, I would not be able to say it here because of the rules! But, I understand your position….I’m used to it when debating religion wink

Profile
 
 
Posted: 14 August 2017 03:56 AM   [ Ignore ]   [ # 8 ]  
Avatar
Cadet
Rank
Total Posts:  20
Joined  2017-04-22

Let’s just keep this thread civilized, productive, on topic, boring and a little bit dry with real, provable data. Please.

Profile
 
 
Posted: 14 August 2017 06:59 AM   [ Ignore ]   [ # 9 ]  
Avatar
Commander
RankRankRankRankRankRank
Total Posts:  1058
Joined  2016-03-16
She wants the… - 14 August 2017 02:07 AM
milamber42 - 13 August 2017 07:35 PM
She wants the… - 13 August 2017 04:14 AM

Yes it does! They didn’t create separate algorithms…one for crits, one for proficency, one for shuttles, one for pips, one for dabo etcetera…it’s the same program working for all of them!

Source?  Do you work for DB?

It’s called common sense! It would be far too expensive to create more than one algorithm program to do the same task. Plus, it would be overkill.

But I love the way apologists need proof for the arguments against, but not for the ones that support their own view smile)

LE. And you already know that if I knew something from DB, from support, I would not be able to say it here because of the rules! But, I understand your position….I’m used to it when debating religion wink

Oh.  I’m far from being a DB apologist, but shuttles, dabo, and mission proficiency rolls are different concepts and most likely have different formulas / algorithms behind them. 

And algorithms and programs are two separate things.  For example, the algorithm / formula to determine the shuttle success percentage is most likely the same between the iOS, Android, Steam, and Facebook game clients.  But those game clients are on separate platforms and as such have separate programs / code bases which implement the one algorithm.

And I don’t think it is against forum rules to mention information provided in support tickets, but it is against forum rules to directly post a support reply without the support reps permission.  In a recent support ticket I have on a stuck shuttle, my rep did suggest that the code bases are different between the different platforms.

Profile
 
 
Posted: 14 August 2017 10:05 AM   [ Ignore ]   [ # 10 ]  
Avatar
Ensign
RankRankRank
Total Posts:  118
Joined  2013-04-28

ok…i’ll just say that support told me that the percentages are not accurate.

Profile
 
 
Posted: 14 August 2017 10:44 AM   [ Ignore ]   [ # 11 ]  
Avatar
Lieutenant
RankRankRankRank
Total Posts:  393
Joined  2016-02-16

 Signature 

172 Immortalized

Profile
 
 
Posted: 14 August 2017 10:47 AM   [ Ignore ]   [ # 12 ]  
Avatar
Lieutenant
RankRankRankRank
Total Posts:  393
Joined  2016-02-16

DB’s logic. “Lets give them a percentage to show what their chances are, but make sure that it is still completely random. We don’t want them to expect anything from us.”

Error Message:  You are only allowed to post every 60 seconds

 Signature 

172 Immortalized

Profile
 
 
Posted: 14 August 2017 11:04 AM   [ Ignore ]   [ # 13 ]  
Avatar
Commander
RankRankRankRankRankRank
Total Posts:  1594
Joined  2016-09-02
milamber42 - 14 August 2017 06:59 AM
She wants the… - 14 August 2017 02:07 AM
milamber42 - 13 August 2017 07:35 PM
She wants the… - 13 August 2017 04:14 AM

Yes it does! They didn’t create separate algorithms…one for crits, one for proficency, one for shuttles, one for pips, one for dabo etcetera…it’s the same program working for all of them!

Source?  Do you work for DB?

It’s called common sense! It would be far too expensive to create more than one algorithm program to do the same task. Plus, it would be overkill.

But I love the way apologists need proof for the arguments against, but not for the ones that support their own view smile)

LE. And you already know that if I knew something from DB, from support, I would not be able to say it here because of the rules! But, I understand your position….I’m used to it when debating religion wink

Oh.  I’m far from being a DB apologist, but shuttles, dabo, and mission proficiency rolls are different concepts and most likely have different formulas / algorithms behind them. 

And algorithms and programs are two separate things.  For example, the algorithm / formula to determine the shuttle success percentage is most likely the same between the iOS, Android, Steam, and Facebook game clients.  But those game clients are on separate platforms and as such have separate programs / code bases which implement the one algorithm.

And I don’t think it is against forum rules to mention information provided in support tickets, but it is against forum rules to directly post a support reply without the support reps permission.  In a recent support ticket I have on a stuck shuttle, my rep did suggest that the code bases are different between the different platforms.

the post you quoted is not wrong. it all comes down to selecting a random integer from a range of available values. with portal pulls each crew and schematic are assigned a value and the RNG pulls from a hat. the gauntlet is self explanatory, and the frequency of wild outcomes is a prime example of said RNG’s flaws. the same goes for mission drops, scans, space battle crits and evades, etc. it’s just a number that gets picked from a range of numbers. that’s it. there’s no need to reinvent the wheel fro every little feature in the game.

 

 Signature 

“the village is strong!”

Profile
 
 
Posted: 14 August 2017 12:16 PM   [ Ignore ]   [ # 14 ]  
Avatar
Cadet
Rank
Total Posts:  20
Joined  2017-04-22

eh… perle ai porci…

All I can read right now is speculation if it’s not sustained by any proof. Even empirical data would be better than nothing.

The Random Number Generator is used to weigh another number generated by a formula. So you can discuss about the weight of the RNG, the formula, or the Random number generator itself, or all the above.
But saying as an absolute that the same rng is used without giving any information in support of that assumption doesn’t prove anything.
Also saying that something is self explanatory doesn’t prove anything.

That said, all programs are platform dependent, so even if the code is the same for every platform, the compiler for every platform may differ, giving different results/bugs in different platforms. But here we’re not discussing a platform dependent bug or behaviour.

Profile
 
 
Posted: 14 August 2017 12:43 PM   [ Ignore ]   [ # 15 ]  
Avatar
Ensign
RankRankRank
Total Posts:  118
Joined  2013-04-28
Capt. Reynolds - 14 August 2017 12:16 PM

eh… perle ai porci…

Capt. Reynolds, who are the pigs in this context??? I think you owe a few apologies!

Profile
 
 
   
 1 2 > 
1 of 2