here's my current breakdown Tb = time boosted, 3* = stat boost, 9H = 9 hour loot boost
edit: *p/*f is pass or fail
Thanks! This sort of validates what I have seen, your shuttle align nicely with hitting the first trait on AND nodes. As soon as I did that, it worked great. However, when I started, I used this crew with the displayed success rates:
However, if you assume that the AND shuttles are First-Second, not Highest-Lowest, they re-calculate like this:
My results match the rate from the second chart....
I ran the same 4 shuttles every round, every day. Same crew, same boost, every time. I ran the three seater with the DIP and SEC, MED and DIP, SEC and MED ("By Force, If Necessary"). I had Katra McCoy in the middle seat and Rakal Troi in the far right seat. The left seat varied between Romulan Kirk or Tuvix. Depending on boost, the success was 87% to 92%. This mission succeeded all day Thursday and Friday. Then it failed all day Saturday and Sunday. It failed 9 times in a row. It was so reliable that late Sunday I swapped it out for something else and got my 4/4 results back. Personally, if I had access to the codebase and the database, I'd be searching through those logs and debugging. If that really is legit RNG, I'm going to be trying my luck at something I normally fail / have bad odds at.
Thanks for tracking that. Confusing highest/lowest with first/second is such an easy typo to make, I've suspected there may be an issue on the server side with it, but I never had the data to prove it.
I ran the same 4 shuttles every round, every day. Same crew, same boost, every time. I ran the three seater with the DIP and SEC, MED and DIP, SEC and MED ("By Force, If Necessary"). I had Katra McCoy in the middle seat and Rakal Troi in the far right seat. The left seat varied between Romulan Kirk or Tuvix. Depending on boost, the success was 87% to 92%. This mission succeeded all day Thursday and Friday. Then it failed all day Saturday and Sunday. It failed 9 times in a row. It was so reliable that late Sunday I swapped it out for something else and got my 4/4 results back. Personally, if I had access to the codebase and the database, I'd be searching through those logs and debugging. If that really is legit RNG, I'm going to be trying my luck at something I normally fail / have bad odds at.
Want to know what’s really bizarre? I didn’t see this mission once from the start of the event until this morning on my final round of shuttles. I had no idea there was an event mission with MED in it for almost four days.
Want to know what’s really bizarre? I didn’t see this mission once from the start of the event until this morning on my final round of shuttles. I had no idea there was an event mission with MED in it for almost four days.
I recommend always opening all the shuttles to scope out the best options
I think I'll chime in as someone else who hasn't noticed this phenomenon. I did have two 95%+ shuttles fail while ramping up, but once I hit 4000 shuttles my displayed percentages varied from 75-90% and I usually got 3/4 or 4/4 (occasionally 2/4) success.
One thing I don't do is cherry-pick the best missions and run only those over and over (I can't spare the merits or time). After the jump-start I usually just open 4 and run them, regardless of percentages or number of seats (although I'll assign 5 seaters last). If specific missions are bugged, that may explain why some are seeing this suspecious pattern.
Want to know what’s really bizarre? I didn’t see this mission once from the start of the event until this morning on my final round of shuttles. I had no idea there was an event mission with MED in it for almost four days.
I recommend always opening all the shuttles to scope out the best options
With all of the bonus crew I have (between Vulcans, Kirks, and Romulans), there was no need to waste the merits.
I ran the same 4 shuttles every round, every day. Same crew, same boost, every time. I ran the three seater with the DIP and SEC, MED and DIP, SEC and MED ("By Force, If Necessary"). I had Katra McCoy in the middle seat and Rakal Troi in the far right seat. The left seat varied between Romulan Kirk or Tuvix. Depending on boost, the success was 87% to 92%. This mission succeeded all day Thursday and Friday. Then it failed all day Saturday and Sunday. It failed 9 times in a row. It was so reliable that late Sunday I swapped it out for something else and got my 4/4 results back. Personally, if I had access to the codebase and the database, I'd be searching through those logs and debugging. If that really is legit RNG, I'm going to be trying my luck at something I normally fail / have bad odds at.
There might be something screwy with the way shuttles have been functioning, but I don't think it was really at play with that specific mission with those specific crew members. I also ran that mission with the exact same crew allocation on repeat [EDIT: ...except that I always used RomKirk...], and it worked well for me. Can't say that it never failed, but then again I only had 88% success rate projected. Failing one out of ten times seems to fit my recollection.
I suspect you just got really, really unlucky on that one...
I wonder. Was this issue ever reported in Engineering section?
The problem is perception. I look at the mission success and the percentages and expect them to match, because I'm am playing my instance of the game, which in most other games is isolated from other players. But DB bases the percentages upon the missions of the entire player base.
So while it is valid from their point of view to be the 1% that fails more frequently, it does not fit most individuals perception.
I wonder. Was this issue ever reported in Engineering section?
The problem is perception. I look at the mission success and the percentages and expect them to match, because I'm am playing my instance of the game, which in most other games is isolated from other players. But DB bases the percentages upon the missions of the entire player base.
So while it is valid from their point of view to be the 1% that fails more frequently, it does not fit most individuals perception.
The whole "based off the entire player base" thing is, likely, a bad explanation from a CS rep who has a poor understanding of probability. I'm pretty sure they are trying to convey the concept of "a small sample size is not necessarily indicative of the population as a whole", which, in general statistics terms, is 100% correct.
The idea that there is some sort of "global" pull on the direction of shuttle failures and successes (especially one that isn't quantified in the "Expected Success" output) is really kind of outlandish. If they DID implement something like that, I would expect that it would be accounted for in the shuttle output, otherwise, the number truly is intentionally lying.
This issue seems a lot simpler than that: a discrepancy between the client-side algorithm for the displayed success rate vs the server-side algorithm for determining if a shuttle actually succeeded.
Comments
Thanks! This sort of validates what I have seen, your shuttle align nicely with hitting the first trait on AND nodes. As soon as I did that, it worked great. However, when I started, I used this crew with the displayed success rates:
However, if you assume that the AND shuttles are First-Second, not Highest-Lowest, they re-calculate like this:
My results match the rate from the second chart....
edit: grabbed the wrong tabs the first time....
Want to know what’s really bizarre? I didn’t see this mission once from the start of the event until this morning on my final round of shuttles. I had no idea there was an event mission with MED in it for almost four days.
I recommend always opening all the shuttles to scope out the best options
One thing I don't do is cherry-pick the best missions and run only those over and over (I can't spare the merits or time). After the jump-start I usually just open 4 and run them, regardless of percentages or number of seats (although I'll assign 5 seaters last). If specific missions are bugged, that may explain why some are seeing this suspecious pattern.
With all of the bonus crew I have (between Vulcans, Kirks, and Romulans), there was no need to waste the merits.
There might be something screwy with the way shuttles have been functioning, but I don't think it was really at play with that specific mission with those specific crew members. I also ran that mission with the exact same crew allocation on repeat [EDIT: ...except that I always used RomKirk...], and it worked well for me. Can't say that it never failed, but then again I only had 88% success rate projected. Failing one out of ten times seems to fit my recollection.
I suspect you just got really, really unlucky on that one...
The problem is perception. I look at the mission success and the percentages and expect them to match, because I'm am playing my instance of the game, which in most other games is isolated from other players. But DB bases the percentages upon the missions of the entire player base.
So while it is valid from their point of view to be the 1% that fails more frequently, it does not fit most individuals perception.
The whole "based off the entire player base" thing is, likely, a bad explanation from a CS rep who has a poor understanding of probability. I'm pretty sure they are trying to convey the concept of "a small sample size is not necessarily indicative of the population as a whole", which, in general statistics terms, is 100% correct.
The idea that there is some sort of "global" pull on the direction of shuttle failures and successes (especially one that isn't quantified in the "Expected Success" output) is really kind of outlandish. If they DID implement something like that, I would expect that it would be accounted for in the shuttle output, otherwise, the number truly is intentionally lying.
This issue seems a lot simpler than that: a discrepancy between the client-side algorithm for the displayed success rate vs the server-side algorithm for determining if a shuttle actually succeeded.