@[NDQ]Joker41NAM I started adding my data to yours with my screenname. I am also saving my runs in a spreadsheet. So far my sample size is too small to determine anything. Hopefully when combined with everyone's data we can start coming up with something.
If I have time I'll add stuff to the community wide data. This is mine so far. If I'm being spammy let me know. I can update a post vs a new one.
New update, new fails
I've had 5 total fails and 3 of them used this crew in some fashion on different missions
I sent out Lead By Example at 99%, very curious as to whether or not it would succeed. It didn't.
Distruptor Beam, what is going on? If shuttles sent initially in this event all pass at 99%, why doesn't this one?
as have many asked for the raw stats be posted.
i would more like to see their algorithm calculation so can verify how they are coming up with their calculations, and see for 100% if it is right, because the odds of success are from from right. some in our fleet sent out 3 0% shuttles and got back 3 success' off of a 0% chance, that just completely defies DB's claims that it is normal. 0% is strait up 0% meaning no chance ever.
so i say, show us the algorithm calculation in the raw code. and let us decide for our selves.
DB needs to fire the Ferrengi and higher more Engineers, Rom doesn't count. [FSC] Peace Keepers
Gryphon [****] Adm
What bothers me is have they fundamentally altered this particular mission? I will play aware the odds are up to RNG, but this seems deliberately skewed. Disruptor Beam, do you have an answer for this? What is the point of participating in an event when clearly something is amiss.
Missions (Hirogen): A Worthy Opponent (3-seat), Follow the Leader (4-seat), Trouble in the Dollhouse (4-seat), Consolation Prize (4-seat)
I have tracked all 10 waves of shuttles (40 total missions) since I finished my 'modified kick-start'. That includes 1 wave @ 2,250, 1 wave @ 2,750, 1 wave @ 3,500, and 7 waves (so far) @ 4,000.
All 40 missions were posted between 91% - 97%, with the 28 missions @ 4,000 ranging from 91% - 94%. The average posted success rate for all 40 missions is 93.55%.
In reality, I have only passed 33/40, for a true success rate of 82.5%. I understand standard deviation and luck, but a difference of over 11% just doesn't make sense.
I am still sitting just outside the top-200, so I am quite happy, yet still can't help but feel a little disappointed in success rates compared to the posted mission success %.
Observations that can't be quantified:
- the mission with a shared crew member seems to fail much more often, despite having a 92% posted success rate (usually not the lowest of each wave)
- a couple waves were run with a combination of 3* and 4* boosts, and the missions with the lesser boost seemed to fail more often
I sent out Lead By Example at 99%, very curious as to whether or not it would succeed. It didn't.
Distruptor Beam, what is going on? If shuttles sent initially in this event all pass at 99%, why doesn't this one?
with the crew you have previous 99% shuttles were actually 99.999...% and depending on the vp level, even by only filling one slot you had the display rate of 99%. Where I'm going with this, while it is statistically possible to fail a 99% shuttle, in reality it can be that one of your crew or one of the slots are not registering their stats. I doesn't fail at lower level but it may at 1800vp you were playing for. The other theory is that rng instead of going from 0 to 100 has a compressed range.
But I'm curious, why did you apply skill boost? what was the success rate before boost?
on the subject of the 99% shuttles that fail, keep in mind that in the past they were rounded to the nearest integer, so people complained when a 100% shuttle failed. Rounding by defect made it just a little bit more plausible, but not really.
Success rate before boost was 96 or 97%; I added the boost because I wanted to see just how high of a percentage I could get, hoping for 99%, which passes all initial shuttles sent out. About a year ago, give or take a few months, when sending the initial one minute shuttles, often the pass rate was 100%, then it was altered to 99%. So if those shuttles always pass, why does it change? I'm not talking statistically, but in the grand scheme. I feel as though Distruptor Beam has altered the rules, which doesn't surprise me, however I am weary of these deliberate manners of.... Let's just say after next week's event, where I will collect my fourth star on Cromwell, I will be altering my methods of play. No use in spending my time on a game when it is not appreciated.
I sent out Lead By Example at 99%, very curious as to whether or not it would succeed. It didn't.
Distruptor Beam, what is going on? If shuttles sent initially in this event all pass at 99%, why doesn't this one?
I would try someone with a sci primary in the first slot of that mission. I think one of the current theories is the on screen percent shows the high stat plus 1/4 the low stat, but the server percent is first stat plus 1/4 second stat.
You are probably correct. My point is that as someone who doesn't take into account the detailed stats, playing this solely for fun, where is the enjoyment of sending off a shuttle at 99% only to have it fail? Makes me feel as if I'm being jerked around, and that is not my reason for spending time on this game. I realize I'm beating a horse well past dead, but if a mission is hard-wired to be so flaky, why participate at all? I don't mean to ignore the stats so painstakingly put into this thread; without them, I wouldn't have been aware of the delicacy of this particular mission, and I am indeed grateful to all those who have contributed. But aside from the stats, there is the simple fact that if this is how Disruptor Beam wants to treat its customer base, and goodness knows that's the sentiment this player is feeling, don't be surprised if this customer doesn't renew their monthly card, refrains from additional spending, and eventually walks away. There is RNG, and then there is something else going on. (Don't mean to sound conspiratorially minded, but I've been playing this game for a while....)
At this point there is no longer any way that I can say that DB is NOT lying. These issues have persisted forever, they're acutely aware that we are not getting the results they're telling us. Be it intentional, or bad programming, the fact that they're aware and continue to present it as Gospel makes it a lie.
Then they tell CS all is good, dismiss any case our customers bring to you, this is active lying, even if the rep is just a poor toadie following orders in an overworked terrible job.
These shuttle events are just flat out unenjoyable when you have to deal with these antics and failure rates. I wish I could see their stats and see how many people this has driven and is driving away.
Here's a free clue, you don't even have to buy a vowel, people walk away from your game when it's a chore to play.
I haven’t kept track like many others but my gut told me the same thing within a few hours, something is very rotten in Denmark. When one has played this for a long time you can pick out things that are awry. I just hope the data everyone is painstakingly collecting actually does something and we don’t get the standard “you only have a small sample, nothing is wrong” when it is so obvious something is.
I am actually running a bit above expectation across 23 shuttles tracked in this event. Displayed ~94%, actual success average is 95.65%. Tiny sample size yet.
I just want it to reflect actual percentages. DB would get into trouble if Google and Apple have a rash of complaints requesting refunds because it was found that actual percentages are different than shown. I would like them to fix it before they lose a ton over this issue.
Missions (Hirogen): A Worthy Opponent (3-seat), Follow the Leader (4-seat), Trouble in the Dollhouse (4-seat), Consolation Prize (4-seat)
I have tracked all 10 waves of shuttles (40 total missions) since I finished my 'modified kick-start'. That includes 1 wave @ 2,250, 1 wave @ 2,750, 1 wave @ 3,500, and 7 waves (so far) @ 4,000.
All 40 missions were posted between 91% - 97%, with the 28 missions @ 4,000 ranging from 91% - 94%. The average posted success rate for all 40 missions is 93.55%.
In reality, I have only passed 33/40, for a true success rate of 82.5%. I understand standard deviation and luck, but a difference of over 11% just doesn't make sense.
I am still sitting just outside the top-200, so I am quite happy, yet still can't help but feel a little disappointed in success rates compared to the posted mission success %.
Observations that can't be quantified:
- the mission with a shared crew member seems to fail much more often, despite having a 92% posted success rate (usually not the lowest of each wave)
- a couple waves were run with a combination of 3* and 4* boosts, and the missions with the lesser boost seemed to fail more often
Updating with additional stats.
I have run 5 more waves (20 missions total), with consistently posted success rates of 91%, 92%, 94% and 94%, for an average of 92.75%
Actual results: 4/4, 2/4, 3/4, 3/4 and 2/4, for a total of 14/20 = 70.0%.
The actual results for my last 5 waves of 4,000 point missions was 22.75% less than what was posted! That is absolutely ridiculous!
In total, for 60 missions, posted % VS actual % = 93.3% VS 78.3% (difference of -15%)
If I include all 31 missions from my modified kick-start, which were all at 99%...
- total for 91 missions, posted % VS actual % = 95.2% VS 85.4% (difference of -9.8%)
Overall total is currently 91 shown to 88 actual. But I am having good luck post phase 1 shuttles and their shown percentages are much lower. Phase 1 was 94.4 shown to 88 actual. Phase 2 is 82 shown to 88 actual.
a little update while I still have leftover missions to run.
I didn't have the big outlier behavior with missions with shared crew like in the previous event and being only a hybrid event doesn't help to get a good number of mission stats.
I ran only the same SEC31 missions:
S Questionable Loyalty, with shared crew
S Identifiable Weaknesses
S Desperation
S Lead by Example
The mission with shared crew had a displayed rate of 94% and actual rate of 80%, while missions without shared had 92,62% display rate - post-boost and 95,5% actual rate.
Overall 93% display rate - post-boost vs 91,6% actual rate
So I had other event data laying on my computer and combined for 211 shuttles. The biggest difference I saw was in the 90 to 94 range. 6.5 point difference after 109 shuttles. When I get a chance I'll combine my data with hopes and it will be almost 1000 runs of data.
Alright last of my spam posts. This is data combined with hopes. 939 event shuttles over the last 6 months. Not every event was tracked but when it was it's almost to 1000. 7.7 overall point difference. 80-89 percent had the worst difference followed by 90-94 percent shown. So it's enough data for me to officially declare its broken and we need an official DB response or fix. (I know, going on a limb saying it's broken lol)
ok... I played a little bit with numbers. In the end I didn't have many shuttles stats to begin with, I'll call this a dress rehearsal for next faction only event.
stats varies a lot from player to player, from mission to mission, with or without shared crew.
Combining these 3 datasets:
and a little bit of statistics of the most run missions:
I think there's enough of a case to say that the displayed percentages are wrong. The question is if DB will just say that the sample size is too small.
All I need to know is this: I sent four shuttles at 0% at the end of the first phase of this event and two came back successful. That alone should tell you that something is wrong with the displayed percentages.
The most damming evidence in all this is that as an average, there should be some data points above, at, and below the displayed average. To date, there have been zero even approaching, let alone at, or above the displayed average. There is no possible way these averages can be correct.
Small sample size is also no longer a viable explanation. At this point, the only way small sample size could be viable is if there were 3x more sample points than the total sum collected here that were *all* at or above the posted average.
We have another faction event coming up. What, if anything, is our plan to track? Have we learned anything yet?
I don't know how much I learnt, other than it's a big mess... so, I picked my pile of events I tracked since mid July and put everything together, discarding data for which I wasn't sure about or didn't have the time to track, nonetheless, 1644 shuttle missions. So here they are:
then going event from event looking for patterns...
with this one there was a mission that stands out as really odd. Changing the bonus for the shared crew from 3 to 0 made the actual rate a lot more reasonable.
again here all missions with shared crew had really lower rate
and here...
another one with a mission to avoid... I'm masochist, I didn't:
another event where shared crew (which was newly paid and leveled) gave no love at all:
then another event with a mission to avoid:
and another one where shared crew probably was out buying cygarettes..
Seeing that there are funny things going on with single crew and/or missions, other than the compressed actual rate, I think it's better to keep into account acquiring more data than only the displayed rate vs. actual rate.
I'll keep tracking, again with a shared document, so if you're interested... message me here or on discord. I don't care much about this event, so I'm willing to fail on purpose if anyone finds oddities with a mission or a crew and wants to test it more...
Comments
New update, new fails
I've had 5 total fails and 3 of them used this crew in some fashion on different missions
I could hit top row every week!
Distruptor Beam, what is going on? If shuttles sent initially in this event all pass at 99%, why doesn't this one?
i would more like to see their algorithm calculation so can verify how they are coming up with their calculations, and see for 100% if it is right, because the odds of success are from from right. some in our fleet sent out 3 0% shuttles and got back 3 success' off of a 0% chance, that just completely defies DB's claims that it is normal. 0% is strait up 0% meaning no chance ever.
so i say, show us the algorithm calculation in the raw code. and let us decide for our selves.
[FSC] Peace Keepers
Gryphon [****] Adm
Missions (Hirogen): A Worthy Opponent (3-seat), Follow the Leader (4-seat), Trouble in the Dollhouse (4-seat), Consolation Prize (4-seat)
I have tracked all 10 waves of shuttles (40 total missions) since I finished my 'modified kick-start'. That includes 1 wave @ 2,250, 1 wave @ 2,750, 1 wave @ 3,500, and 7 waves (so far) @ 4,000.
All 40 missions were posted between 91% - 97%, with the 28 missions @ 4,000 ranging from 91% - 94%. The average posted success rate for all 40 missions is 93.55%.
In reality, I have only passed 33/40, for a true success rate of 82.5%. I understand standard deviation and luck, but a difference of over 11% just doesn't make sense.
I am still sitting just outside the top-200, so I am quite happy, yet still can't help but feel a little disappointed in success rates compared to the posted mission success %.
Observations that can't be quantified:
- the mission with a shared crew member seems to fail much more often, despite having a 92% posted success rate (usually not the lowest of each wave)
- a couple waves were run with a combination of 3* and 4* boosts, and the missions with the lesser boost seemed to fail more often
with the crew you have previous 99% shuttles were actually 99.999...% and depending on the vp level, even by only filling one slot you had the display rate of 99%. Where I'm going with this, while it is statistically possible to fail a 99% shuttle, in reality it can be that one of your crew or one of the slots are not registering their stats. I doesn't fail at lower level but it may at 1800vp you were playing for. The other theory is that rng instead of going from 0 to 100 has a compressed range.
But I'm curious, why did you apply skill boost? what was the success rate before boost?
on the subject of the 99% shuttles that fail, keep in mind that in the past they were rounded to the nearest integer, so people complained when a 100% shuttle failed. Rounding by defect made it just a little bit more plausible, but not really.
I would try someone with a sci primary in the first slot of that mission. I think one of the current theories is the on screen percent shows the high stat plus 1/4 the low stat, but the server percent is first stat plus 1/4 second stat.
Then they tell CS all is good, dismiss any case our customers bring to you, this is active lying, even if the rep is just a poor toadie following orders in an overworked terrible job.
These shuttle events are just flat out unenjoyable when you have to deal with these antics and failure rates. I wish I could see their stats and see how many people this has driven and is driving away.
Here's a free clue, you don't even have to buy a vowel, people walk away from your game when it's a chore to play.
Or at the very least, they threaten to on the forum.
I just want it to reflect actual percentages. DB would get into trouble if Google and Apple have a rash of complaints requesting refunds because it was found that actual percentages are different than shown. I would like them to fix it before they lose a ton over this issue.
Updating with additional stats.
I have run 5 more waves (20 missions total), with consistently posted success rates of 91%, 92%, 94% and 94%, for an average of 92.75%
Actual results: 4/4, 2/4, 3/4, 3/4 and 2/4, for a total of 14/20 = 70.0%.
The actual results for my last 5 waves of 4,000 point missions was 22.75% less than what was posted! That is absolutely ridiculous!
In total, for 60 missions, posted % VS actual % = 93.3% VS 78.3% (difference of -15%)
If I include all 31 missions from my modified kick-start, which were all at 99%...
- total for 91 missions, posted % VS actual % = 95.2% VS 85.4% (difference of -9.8%)
I didn't have the big outlier behavior with missions with shared crew like in the previous event and being only a hybrid event doesn't help to get a good number of mission stats.
I ran only the same SEC31 missions:
S Questionable Loyalty, with shared crew
S Identifiable Weaknesses
S Desperation
S Lead by Example
The mission with shared crew had a displayed rate of 94% and actual rate of 80%, while missions without shared had 92,62% display rate - post-boost and 95,5% actual rate.
Overall 93% display rate - post-boost vs 91,6% actual rate
stats varies a lot from player to player, from mission to mission, with or without shared crew.
Combining these 3 datasets:
and a little bit of statistics of the most run missions:
https://docs.google.com/spreadsheets/d/1lm2IsgidhAXtCfApSk8Z0Lyzscc0-_u_HtyDWe2WZUM/edit?usp=sharing
All I need to know is this: I sent four shuttles at 0% at the end of the first phase of this event and two came back successful. That alone should tell you that something is wrong with the displayed percentages.
Will DB actually do anything about it?
Small sample size is also no longer a viable explanation. At this point, the only way small sample size could be viable is if there were 3x more sample points than the total sum collected here that were *all* at or above the posted average.
I don't know how much I learnt, other than it's a big mess... so, I picked my pile of events I tracked since mid July and put everything together, discarding data for which I wasn't sure about or didn't have the time to track, nonetheless, 1644 shuttle missions. So here they are:
then going event from event looking for patterns...
with this one there was a mission that stands out as really odd. Changing the bonus for the shared crew from 3 to 0 made the actual rate a lot more reasonable.
again here all missions with shared crew had really lower rate
and here...
another one with a mission to avoid... I'm masochist, I didn't:
another event where shared crew (which was newly paid and leveled) gave no love at all:
then another event with a mission to avoid:
and another one where shared crew probably was out buying cygarettes..
Seeing that there are funny things going on with single crew and/or missions, other than the compressed actual rate, I think it's better to keep into account acquiring more data than only the displayed rate vs. actual rate.
I'll keep tracking, again with a shared document, so if you're interested... message me here or on discord. I don't care much about this event, so I'm willing to fail on purpose if anyone finds oddities with a mission or a crew and wants to test it more...
all stats can be found here... do whatever you want with it...
https://docs.google.com/spreadsheets/d/1QfEL-1z7Vtay86dB_E1qppp1hLooA6RBzggifkXwGrs/edit?usp=sharing