Home The Bridge
Options

What is the odd in Gauntlet of having 9/10 rounds with Med traits

VioletBlueVioletBlue ✭✭✭
edited October 2017 in The Bridge
Certainly since Med is the least used trait in gauntlet, typically I have 1, esp with my current gauntlet has the Sci as the main trait.

Right after Phlox was out, I got 9 Med in a row (with 9 merit refresh).
Then the 10th round where I have hard to lose match (with my Valeris 45% vs Tuvok 45%), Valeris crits 0 time and Tuvok 2 times.

Certainly the crits are questionable and a lot of ppl disagree with this and DB always think this is part of their design.

But this is not the first time I got these ridiculous Med traits like this.
Anyone else have had this?

Comments

  • Options
    Was medicine the chosen skill?
    “What's a knockout like you doing in a computer-generated gin joint like this?”

    Proud member of Patterns of Force
    Captain Level 99
    Played since January 2017

    TP: Do better!!!
  • Options
    Paladin 27Paladin 27 ✭✭✭✭✭
    If this is the one that started last night it was science. That one had Valeris as a 45% crit, but I don't see how tuvok was 5% since Vulcan was a trait. In fact disguised tuvok is also 45%.
  • Options
    Well, assuming they're using the same probability number that the success rates in shuttles are using... given that there are six different skills that could come up as either slot, and that numbers actually don't matter at all in this game...

    The odds of a gauntlet having 9 out of 10 rounds pull med are 145%, using DB math.
  • Options
    I gave up on keeping track of gauntlet stats long ago (and decided to just quit stressing over it, and doing just as many rolls with me as an underdog as with me as the favorite, because it's so stupidly random), but anecdotally, I sure as hell notice times when the featured skill has seemingly changed mid-gauntlet. A sudden run of 7 or 8 MED (or SCI or ENG) rolls, when the featured skill was CMD, can really ruin one's day, and certainly gets noticed.

    An alternate theory to the return of the "gauntlet skill switching bug" would be something others have mentioned before: DB's random number generation may be very lacking.

    There's a lot of reading you can do on the subject (which I did after the theory was first mentioned; I recommend wikipedia's Random Numbers page, as well as random.org), and it comes down to the fact that true RNG (or even pretty good RNG) can be a bit difficult to do, and the only times it is legally obligated is in casino/lottery/regulated-gaming situations. DB (and pretty much all mobile game developers) are so far able to skirt this requirement, and we have no idea exactly what methods they are using for their RNG.

    It's very possible that DB is just using some kind of built-in RNG feature included in whatever OS or programming language they started with, in which case, it's almost certainly a very weak pseudo-RNG. The only way to get PRNG systems to provide reasonably fair degree of randomness is to be very careful about your selection of the "seed."

    A weak PRNG mechanic combined with less-than-robust seed generation could go a long way to describing the "clumpy" lack of randomness that DB's "R"NG seems to serve up. It's not just in the gauntlet where weird runs of non-featured skills happen several times per week, either. I just had a run of 4 days where I got 4* crew on 3 Voyages out of 7.

    All 3 were Agent Harris.

    That, my friends, is an almost comically outrageous statistical improbability. It almost convinces me that the "fin foil hat brigade" were right all along when they started talking about DB's RNG being fundamentally flawed. I mean, I more or less concede that DB probably has a less-than-optimal PRNG system (because good RNG is expensive and/or complicated, and not required by law -- yet), but more recently, I'm starting to think there could be something to DB's seed choices being rather suspect, too (such as using our DBID numbers as an element of the seed, which would explain why individual players might experience said "clumping" instead of true randomness).

    End conspiratorial rant.
  • Options
    Banjo1012Banjo1012 ✭✭✭✭✭
    edited October 2017
    I noticed it too. It seems med comes up more than any other regardless. I have countered this by using Phlox and Waitress Ezri every gauntlet and it works for me.
  • Options
    It would be nice if they showed the rolls we actually got versus what was needed to win after the fact. Then we could gather the precise results and see if the histograms follow the correct distribution, which if the probabilities are all the same, should look completely flat.

    I was stunned to get the Gorn captain twice in a row on voyages yesterday, the odds to get any two of the same purples in a row is probably in the 1/60k range. To hear the above anecdote about 3 agent Harrisses in a row (maybe 1 in 15 million or so) makes me curious if the phenomenon is widespread.
  • Options
    Paladin 27 wrote: »
    If this is the one that started last night it was science. That one had Valeris as a 45% crit, but I don't see how tuvok was 5% since Vulcan was a trait. In fact disguised tuvok is also 45%.

    Sorry, I missed the 45%. It was 45% vs 45%, but valeris crits 0
  • Options
    [RO]BitterDevil[RO]BitterDevil ✭✭✭
    edited October 2017
    Certainly since Med is the least used trait in gauntlet, typically I have 1, esp with my current gauntlet has the Sci as the main trait.

    Right after Phlox was out, I got 9 Med in a row (with 9 merit refresh).
    Then the 10th round where I have hard to lose match (with my Valeris 45% vs Tuvok 45%), Valeris crits 0 time and Tuvok 2 times.

    Certainly the crits are questionable and a lot of ppl disagree with this and DB always think this is part of their design.

    But this is not the first time I got these ridiculous Med traits like this.
    Anyone else have had this?

    well...considering i just had 5 MED in a row(the featured skill being SCI ofc) i'd say the chances are pretty high...

    I would say something about DB's "it works perfectly" but i'm under probation for saying trolling things like that or pointing out flaws in the game.

    LE. make that 6 times in a row

    LLE. can you say 7?
  • Options
    STT_Junkie wrote: »
    It would be nice if they showed the rolls we actually got versus what was needed to win after the fact. Then we could gather the precise results and see if the histograms follow the correct distribution, which if the probabilities are all the same, should look completely flat.

    I was stunned to get the Gorn captain twice in a row on voyages yesterday, the odds to get any two of the same purples in a row is probably in the 1/60k range. To hear the above anecdote about 3 agent Harrisses in a row (maybe 1 in 15 million or so) makes me curious if the phenomenon is widespread.

    In initial voyages when they started I used to get 3-4 copies of same card and just got 4 doniks in the week plus several 3 star tuvoks which is why I am working on a Tuvix
  • Options
    Well, MED shows up in about 1/4 of the rounds (if we assume it's not featured and the featured trait doubles the chances). The chances of MED in 9/10 or more is about 0.0029%, or 1 in 34500.

    Do you think there were more than 34500 runs of 10 gauntlet matches played/refreshed today? I suspect there were more like 100000 of them if not many more--remember that if you run 20 gauntlet matches, you've made 11 groups of 10 runs--meaning around 3 people saw a run with 9/10 or 10/10 MED.

    You were one of those lucky 3. Congrats!
  • Options
    IrialIrial ✭✭✭✭
    I mean, I more or less concede that DB probably has a less-than-optimal PRNG system (because good RNG is expensive and/or complicated, and not required by law -- yet), but more recently, I'm starting to think there could be something to DB's seed choices being rather suspect, too (such as using our DBID numbers as an element of the seed, which would explain why individual players might experience said "clumping" instead of true randomness).

    End conspiratorial rant.

    I have noticed some "clumped" luck (i.e., both good and bad) when farming certain items from missions, and had the following thought cross my mind: could the 'seed' possibly include the current time? That is, when my luck was good, had I just picked the perfect time of day to run that particular mission, and vice versa when I couldn't get the item I was wanting to drop ...

    ... please resume your conspiratorial investigation :p
  • Options
    Kirian wrote: »
    You were one of those lucky 3. Congrats!

    yay. i am buying loterry tomorrow

  • Options
    Of course it works across 100,000 people. That's why when I fail shuttles at 99/99/97 % it is not MY shuttles that count it's EVERYONE ELSE'S shuttles. Somehow. Now I understand. Will everyone atop sending shuttles please so I can start getting successful missions? Ta!
  • Options
    You want more ppl to send more shuttles and then hoping others take your spot & fail.
Sign In or Register to comment.