Shuttle mission success chances during events - Post your data here!
Hope_F
✭✭✭
I (and many of you) have long suspected that DB is giving us wrong information about success chances of shuttle missions during events. I started collecting data on my shuttles several weeks ago during the First Impressions hybrid event. I collected the success and fail data from all missions running 180 minutes (I go through the shorter missions within minutes with time boosts and do not bother putting down the numbers of those minute long missons…since each shuttle event is (supposed to be) an independent event, this should make no difference).
The success and fail rates you can find here:
https://docs.google.com/spreadsheets/d/1h3wLNzSDOiu-OIIUhP9OlaKPqRH2YKT83BOPHpFxFCg/edit#gid=0
I do not show all other collected data (like date of the shuttle, order of shuttles, boosts applied, etc.) in order to not compromise future data collection. But during events there is a boost (mostly lvl 3) applied to almost every shuttle, especially if I reach the 1800/2250 VP range.
I evaluated the overall success chance, the chance for missions less or equal 90%, and for missions between 40 and 60% (although for the latter the amount of collected data is very limited as of right now).
overall less or equal 90% between 40% and 60%
missions 286 156 31
successes 210 96 13
failures 76 60 18
theoretical success 83.27% 73.92% 51.23%
actual success 73.43% 61.54% 41.94%
total failure increase -9.85% -12.38% -9.29%
relative failure increase +58.86% +47.49% +19.05%
(You find those results in the link provided above, this forum is a horror when it comes to provide tables)
As you can see the overall success chance is around 10% lower than you would expect. This sounds not so bad, but what’s actually more interesting (and relevant) is the relative increase of failures. I fail around 60% more shuttles than you would expect. This is A LOT!
Further observations and comments:
1. Statistics usually work like this: You virtually never observe an exact behaviour, but your results circle around the average expected outcome. That is, you should observe deviations lower AND higher than the average. But no matter how hard I try, the results are always lower than expected. This is highly erratic behaviour from a statistical point of view. All three events separately show about the same deviation, whereas the first (hybrid event) was the worst of them.
It might also be interesting that the last three events I tracked had a lot better success chance than the two before that, which I did not track. It were those events that finally compelled me to collect data in order to get a better understanding about what’s going on.
2. The final push shuttles tend to have catastrophic outcomes: In the last two faction events (full faction) I was practically the whole time in exactly the reward bracket I wanted to be in (4 whole days). Both times towards the end of the event the success rates got worse, at first somehow explainable with “bad luck”. The final four shuttles I sent out should have left me easily in the bracket I was in the whole time. But then this happened: During the first event I had a 0.34% chance to fail all four shuttles (and that did not happen during the entire event). All four shuttles failed. The most recent event it was exactly the same. I was where I wanted to be the entire event. The final four shuttles had a 0.26% chance to fail. All four shuttles failed. Both events I was pushed out of the reward bracket I was in the entire event literally at the last minute. The chance of that happening two times in a row is 0.00088%. One could say that this is still larger than zero (and people will, there is a base of people that will not doubt DB on anything even if you bring them a billion data points), but this is hardly the point. It’s another example of “things are always worse than expected and never better). My take on why this is happening: If people get kicked out of the bracket they want to be in they either send more shuttles and bring them home quickly using dilithium or they try to get the crew with event packs (also dilithium). I myself do not pay anything on the game anymore. I got to VIP 4 with month cards alone in order to have 4 shuttles, a few extra crew slots, and two daily replicates. I have not paid anything on the game since then (6 months) and do not intend to. So buying event packs is never gonna happen for me.
My feeling (no proof, but it feels very much so) is that sometimes there is no RNG at work at all, but the system works toward preventing your progress. When you get close to a goal, the system actively prevents it: 4 shuttles in a row fail at exactly the right time, your win streak in the gauntlet gets sabotaged at a certain point no matter how ridiculously low the chance of your opponent is, etc.
3. In a recent post in this forum (and many before) (see https://forum.disruptorbeam.com/stt/discussion/1208/it-would-be-great-if-the-displayed-shuttle-success-percentages-got-fixed) the issue was discussed as well. The member (FF) This Sisko1 said there that he is tracking data for an entire year now and practically observed the same as me. It would be very helpful if (s)he would share this data. That should be more than sufficient proof that DB provides wrong numbers. Also, I encourage everybody else who collected data to post it in this thread. Also, in the thread https://forum.disruptorbeam.com/stt/discussion/476/statistical-improbabilities the member Ditto indicated (s)he had a year worth of data.
4. In the same post it was (as many times before) suggested that boosts might be applied wrongly but the RNG actually calculate the correct results. This would also explain why I do not observe this kind of erratic behaviour outside of events, since I virtually never use boosts there. If this is the case, then DB was told this eons ago and it would have been easy to correct the issue a long time ago. This should be very easy from a programming point of view. Just tell the program to display the exact same number as success chance that the RNG (correctly!) calculates. This should be easy! But we are talking about a company that for months could not fix the bug that constantly shows that you have an achievement reward ready to collect once you finished #1 in the Gauntlet, although you already collected it. So what do I know!?
5. Since the data out there (for example from (FF) This Sisko1 or Ditto) should easily prove that DB provides wrong data, doesn’t that amount to fraud? People are paying real money and make decisions of how many resources they spend based on the numbers DB provide for the shuttles, but they literally do not get what they paid for. A displayed chance of x% should represents a promise from the service provider that that is what you get. When you pay with real dollars in order to increase this chance, and you are not getting this chance, then, from my understanding, this amounts to fraud. If it were an honest mistake, that would be one thing, but DB has been made aware of this problems a thousand times, through the forums and through tickets. They know about the problem, they CHOSE to remain inactive about it. I am not a lawyer, but if there are lawyers out there, I am dying to hear them elaborate on this issue.
The success and fail rates you can find here:
https://docs.google.com/spreadsheets/d/1h3wLNzSDOiu-OIIUhP9OlaKPqRH2YKT83BOPHpFxFCg/edit#gid=0
I do not show all other collected data (like date of the shuttle, order of shuttles, boosts applied, etc.) in order to not compromise future data collection. But during events there is a boost (mostly lvl 3) applied to almost every shuttle, especially if I reach the 1800/2250 VP range.
I evaluated the overall success chance, the chance for missions less or equal 90%, and for missions between 40 and 60% (although for the latter the amount of collected data is very limited as of right now).
overall less or equal 90% between 40% and 60%
missions 286 156 31
successes 210 96 13
failures 76 60 18
theoretical success 83.27% 73.92% 51.23%
actual success 73.43% 61.54% 41.94%
total failure increase -9.85% -12.38% -9.29%
relative failure increase +58.86% +47.49% +19.05%
(You find those results in the link provided above, this forum is a horror when it comes to provide tables)
As you can see the overall success chance is around 10% lower than you would expect. This sounds not so bad, but what’s actually more interesting (and relevant) is the relative increase of failures. I fail around 60% more shuttles than you would expect. This is A LOT!
Further observations and comments:
1. Statistics usually work like this: You virtually never observe an exact behaviour, but your results circle around the average expected outcome. That is, you should observe deviations lower AND higher than the average. But no matter how hard I try, the results are always lower than expected. This is highly erratic behaviour from a statistical point of view. All three events separately show about the same deviation, whereas the first (hybrid event) was the worst of them.
It might also be interesting that the last three events I tracked had a lot better success chance than the two before that, which I did not track. It were those events that finally compelled me to collect data in order to get a better understanding about what’s going on.
2. The final push shuttles tend to have catastrophic outcomes: In the last two faction events (full faction) I was practically the whole time in exactly the reward bracket I wanted to be in (4 whole days). Both times towards the end of the event the success rates got worse, at first somehow explainable with “bad luck”. The final four shuttles I sent out should have left me easily in the bracket I was in the whole time. But then this happened: During the first event I had a 0.34% chance to fail all four shuttles (and that did not happen during the entire event). All four shuttles failed. The most recent event it was exactly the same. I was where I wanted to be the entire event. The final four shuttles had a 0.26% chance to fail. All four shuttles failed. Both events I was pushed out of the reward bracket I was in the entire event literally at the last minute. The chance of that happening two times in a row is 0.00088%. One could say that this is still larger than zero (and people will, there is a base of people that will not doubt DB on anything even if you bring them a billion data points), but this is hardly the point. It’s another example of “things are always worse than expected and never better). My take on why this is happening: If people get kicked out of the bracket they want to be in they either send more shuttles and bring them home quickly using dilithium or they try to get the crew with event packs (also dilithium). I myself do not pay anything on the game anymore. I got to VIP 4 with month cards alone in order to have 4 shuttles, a few extra crew slots, and two daily replicates. I have not paid anything on the game since then (6 months) and do not intend to. So buying event packs is never gonna happen for me.
My feeling (no proof, but it feels very much so) is that sometimes there is no RNG at work at all, but the system works toward preventing your progress. When you get close to a goal, the system actively prevents it: 4 shuttles in a row fail at exactly the right time, your win streak in the gauntlet gets sabotaged at a certain point no matter how ridiculously low the chance of your opponent is, etc.
3. In a recent post in this forum (and many before) (see https://forum.disruptorbeam.com/stt/discussion/1208/it-would-be-great-if-the-displayed-shuttle-success-percentages-got-fixed) the issue was discussed as well. The member (FF) This Sisko1 said there that he is tracking data for an entire year now and practically observed the same as me. It would be very helpful if (s)he would share this data. That should be more than sufficient proof that DB provides wrong numbers. Also, I encourage everybody else who collected data to post it in this thread. Also, in the thread https://forum.disruptorbeam.com/stt/discussion/476/statistical-improbabilities the member Ditto indicated (s)he had a year worth of data.
4. In the same post it was (as many times before) suggested that boosts might be applied wrongly but the RNG actually calculate the correct results. This would also explain why I do not observe this kind of erratic behaviour outside of events, since I virtually never use boosts there. If this is the case, then DB was told this eons ago and it would have been easy to correct the issue a long time ago. This should be very easy from a programming point of view. Just tell the program to display the exact same number as success chance that the RNG (correctly!) calculates. This should be easy! But we are talking about a company that for months could not fix the bug that constantly shows that you have an achievement reward ready to collect once you finished #1 in the Gauntlet, although you already collected it. So what do I know!?
5. Since the data out there (for example from (FF) This Sisko1 or Ditto) should easily prove that DB provides wrong data, doesn’t that amount to fraud? People are paying real money and make decisions of how many resources they spend based on the numbers DB provide for the shuttles, but they literally do not get what they paid for. A displayed chance of x% should represents a promise from the service provider that that is what you get. When you pay with real dollars in order to increase this chance, and you are not getting this chance, then, from my understanding, this amounts to fraud. If it were an honest mistake, that would be one thing, but DB has been made aware of this problems a thousand times, through the forums and through tickets. They know about the problem, they CHOSE to remain inactive about it. I am not a lawyer, but if there are lawyers out there, I am dying to hear them elaborate on this issue.
4
Comments
If you have data to share about shuttle events, please do in this post. Please provide only event results, or at least give separate calculations for events and regular shuttle missions. Also, please indicate how often approximately a booster was used (or if you have it and like to publish it, do show the exact boosters used in the raw data). As I indicated, in my case a skill boost was used virtually the whole time (about 95%). Please do not post here if you simply want to state your opinion or give the recommendations to people just to let it go. This is a post for hard facts, if you don’t have them, please consult different threads. This thread is to provide data in order to help/force DB to deal with a serious issue they ignored for far too long. If we can provide hard evidence they will have no choice but to address the issue. Of course, if you have hard facts that would support that the displayed percentages are correct, please provide those too.
As for me I am not really sure if I will collect further data, simply because I am not sure if I stay with the game. The experiences described under 2. are so frustrating that I am not sure it makes a lot of sense to stay with the game if you get burned all the time, just because you do not throw incredible amounts of money at DB.
I made an extensive spreadsheet so I could track EVERY little detail.
My Event Crew are:
Niners Kira 4/4* 100
Rura Penthe Reed 4/4* 100
Nurse Paris 2/2* FF 100
Major Kira 2/2* FF 100
Captain Proton Paris 4/4* 90
Combat Medic Paris 1/4* 90
Tactical Michael Burnam 1/4* 90
Cardassian Kira: 2/4* 70 (Not used yet - not sure why)
After first 48 hours of this event "The Akritirian Job", here are some stats:
Total Shuttles so far: 72 (first 14 were not counted in time, which were the ones that returned with 30 minutes or less, so below are the 58 that I started counting).
Federation Success/Fail: 13/6
Bajoran Success/Fail: 13/6
Klingon Success/Fail: 14/5
4 currently out on mission
# Shuttles started with 2250 VP, completed with 2750 VP: 4
# Shuttles started with 2750 VP, completed with 3500 VP: 8
# Shuttles started with 3500 VP, completed with 4000 VP: 4
# Shuttles started with 3500 VP, completed with 2750 VP: 8
**No rhyme or reason for the VP changes. Doesn't jive with the Wiki at all**
Shuttle Missions (Successes unless otherwise noted):
Bon Apetit: 2
Crusher Et Al: 3 (1 Fail)
Delay Tactics: 1
Departed Souls: 4 (1 Fail)
House Calls: 1 (Fail)
Insidious Methods: 4 (3 Fail)
Interquadrant Call: 4
Long Distance Call: 1
Looks Like Rain: 2
Penance: 5 (1 Fail)
Pointing Fingers: 1
Put 'Em Up: 5 (3 Fail)
Reporting for Duty: 1 (Fail)
Rosetta Stone: 1
Security Override: 2
See You in Court: 1
Sick Day: 2
Teach A Man to Fish: 1
The Grass is Greener: 2
The Invisible Plan: 3
The Pot and the Kettle: 1
The Red Wire: 2
The Resistance Rises: 4 (1 Fail)
The Right to an Attorney: 2
The Unrepentant Okana: 1
Thicker Than Water: 2 (1 Fail)
Tunnel Kings: 4 (2 Fail)
I will report stats upon completion of the event, and supply my spreadsheet if possible.
The reason is that refresh glitches cause the wrong VP value to often be displayed. To view the ACTUAL VP value of a mission, tap on the VP icon in the lower left corner of the mission screen; the pop-up window will display the true VP value; this should jive with the wiki.
No, I checked & these were ACTUAL values, not the visual bug!
It belogs in Engineering, since it's either a bug or fraud. It is to make DB finally take action, since they ignoring the issue since the beginning. My hope is that if we can prove that the numbers are incorrect, they will take action, and if not...well, the US is the country of the class action law suit...
As for your data, it would help if you provide percentages...what is the average displayed percentage and what is the percentage you actually achieve?
If I calculated correctly, your average displayed percentage is 73.36%, your actual success rate is 66.13 percent, so more than 7% total off.
You should provide this data in your sheet, such that DB can view it quickly and take action.
@DB: TAKE ACTION!!!!
I was waiting until the end of the event to post percentages, since they will change throughout.
If i have only around 80% chance of success, forget it, they will fail.
Spread sheets with all the numbers would be best, but at least post the following.
- How many missions did you do
- What was the DISPLAYED average success chance
- What was the ACTUAL average success chance
Thanks.
This is the summary, with a portion of the spreadsheet attached for the breakdown.
Final Score 259700
Final Rank 1608
Shuttle Missions: Sent Successes Fails: % Success
Federation 39 30 7 76.92
Bajoran 39 24 9 61.54
Klingon 29 22 5 75.86
Shuttle Missions Successes Fails Total
92 29 121
Overall Average Overall Success
Shown Actual
Federation 72.69% 79.06%
Bajoran 75.27% 70.82%
Klingon 74.95% 72.00%
I could have made the stats down to the most minute detail, but I didn't feel the need to dive that deep for this exercise. (Such as which crew were on the most failed shuttles, etc.)
I continue to reason that the stronger the crew, combined with strong boosts in the right places give you a far higher % success. I used 29 different crew: 11 were 4/4* FE, 4 were 4/4* lvl 100, 3 were 2* FE (Bonus Crew), and the rest were lvl 90 4* & 5* crew, and a couple times, there were 5* lvl 80 crew used.
Most of my fleet that complain about shuttle fails, also don't have nearly the level of crew I do (and my top 12 are in limbo, or I would have done even better during this event.
Hope this helps!
[Note: this is NOT the visual bug that has been heavily talked about, but the actual VP points listed on EACH INDIVIDUAL MISSION. I clicked on every ** one of them!]
Shuttles sent out at:
2250 – came back 2750: 4
2750 – came back 3500: 12
3500 – came back 2750: 7
3500 – came back 4000: 5
overall less or equal 90% between 40% and 60%
missions 414 272 60
successes 298 173 29
failures 116 99 31
theoretical success 79.90% 72.30% 53.62%
actual success 71.98% 63.60% 48.33%
difference total success rate -7.92% -8.70% -5.28%
relative failure increase +39.39% +31.40% +11.39%
It got a little better, but of course not because I performed above displayed average, but because the negative bias was not so bad this time. It's still 40% more fails than expected.
There is a minor error in the subtotals, but the totals are right.
Edit: spelling
If you want me to post all 600+ screencaps...well, no I'm not going to do that, but rest assured it is a visual anomaly just like when you see one number on the faction for VP points, and it is a different value when you send from what you saw, and sometimes a different value when the mission returns from when it left.
I have all the proof you need. PM me if you want a selection of screencaps to show you what I mean.
Here are some caps to validate this problem. First 2 are what you were talking about, the order changes from sending to return, but otherwise no big problem I see there. The last 4 are the progression of a shuttle mission: shows points going out at 3500, added crew, then boost, then returns with 4000. This happens the other way more often, i.e. send out at 3500, and comes back 2750.
The last screenshot is a different mission. The completion text is for Fool Me Once, not How Are You Feeling. You can also see that the last one is a 2 slot mission, not 3.
https://stt.wiki/wiki/The_Daystrom_Award/Faction_Missions