Sunday, April 9, 2017

Crew Rolling Experiment Results April 9th

Crew Rolling Experiment

People have been complaining about the Random Number Generate of the Crew Roller has been 'fixed', with no BB or DoS crews popping up. So, to check the claims of this I set out doing 4 different roll experiments. With the following rules 
  • No collecting crews until all uranium is spent 
  • Must all be done in one session
  • If the game or computer crashes, restart 
  • Only roll crews using the Advanced Crew option 
One with 170 rolls, one with 226, and one with 391, and the last with 403. I then complied all in a spreadsheet, and made a total one and an average one. 

Here are the links to the 4 trials
Now that we have the basic data, here is the total between them all. 

The average? 

(Although these are the averages from 1447 rolls, some crews were never rolled. I will be doing another 403 rolls every week, so in a month new data will be available to compare.)

Here are the averages that matter 
  • 37% for rolling a Common Crew 
  • 61% for rolling an Uncommon Crew
  • 1.5% for rolling a Rare Crew 
  • .41% for rolling a Legendary Crew 
And for individual crews 
  • .21% for rolling a Bullseye Brigade
  • .138% for rolling Grease Monkeys
  • 3.8% for Midnight Marauders 
  • 4.98% for Gearheads 
But the biggest shock from it all?
  • The average chance of rolling a common crew is higher than an uncommon crew 
What do you mean? 

Well, if you take the number of times that you can roll a uncommon crew and divide it by the number of total crew rolls, and then compare to the number of common crews over total crews; you'll find that some common crews are easier to get than uncommon crews. Which crews? 

You have a higher chance of rolling a Gearheads crew (4.98%) than you do 
  • Dragonslayers (4.35%)
  • Fearless Blitzers (4.28%)
  • Junkyard Dogs (3.39%)
  • Lucky Bastards (3.94%)
  • Midnight Marauders (3.8%)
  • Molotov Maidens (4.7%)
  • Motorheads (4.15%)
  • Salty Dogs (4.42%) 
  • Sea Serpents (4.7%)
  • Silent Hunters (4.35%)
  • Steelheads (4.01%)
  • Wolfpack (4.63%) 
Holy shit that's a lot of crews. Gearheads isn't the most common crew either, its Lucky Bastards at 5.6%, more than any uncommon crew. Which, isn't right at all. 

The chance of any given crew rolled being uncommon is 61%, and common is 37%; but you are more likely to roll common Lucky Bastards than you are any given uncommon crew! 

What can be done to fix this? 

Well, two things

  1.  Introduce a cap to where once a set number of the same crew is rolled, its taken out of the system for that session. Example? If you roll a Lucky Bastards crew 20 times in a single session, the chance of rolling another Lucky Bastard is 0%. This seems way too much in the player favor, but when you look back at the data; for every 361.75 rolls I only rolled a single crew 20 times, with 8 other crews being 5 rolls or closer. What would happen to the lost 5.6% of the LB crew? Give it to another crew, at random. What does it accomplish? It allows for rarer crews to come out of their shell, and have a higher player satisfaction at rolling crews. How often do you go on the forums are see people complaining about BB crews being gone?
  2. Move more crews to common, but keep the tier percentage the same. Meaning with an average 37% to roll a common crew for the different 8 crews, giving 4.63% per crew is higher than the 61% for the 14 uncommon crews at 4.36%. By adding more common crews, but keeping the tier percentage the same at 37%, you would increase the number of uncommon crews rolled overall. What crews could you add to common? Dead-eye Destroyers, Radar Runners, and Dragonslayers. What would this do? Well, 37% over 11 crews is 3.36% per crew, while 61% for 13 crews is 4.69%. A difference of 1.33%! 
Either way, both changes would need a redo in the crew system. Maybe we need another change in the system. Keep the rarities as they are, just fix the roll chances or move some crews around.

I'm going to continue this for another month or two. Just depends when I get a chance to sit down and roll for an hour or two.

As always Film at 11



  1. This is one of the biggest bunches of crap I've ever read in my life. Ten minutes of my life wasted.

    1. Explain why? People have been complaining in forum for a long ass time; and I wanted to find out the averages per crew, the title should be clue enough that it would be a bore to read

  2. As long as you keep putting out garbage like this to stir up the tin hats and lower levels I will be at you base at least 1-2 weeks a month every month and friends will be visiting the other two weeks. Your info is so bad,that your now considered the new specs and he was one of the most hated mods to date on forums. Oh for someone trying to teach people how to build bases,PS your rhino is a joke at stopping subs ;) Sincerely 9nines,your new battle log buddy

    1. Thanks for the tip! Will try to improve my skills on the next post. P.S. I didn't build that rhino for stopping subs. Can't wait until I see you again, when it takes you 3+ tries and rockets

  3. You know NOTHING about basic math

    1. I double checked everything, what did I possibly miss?

  4. Quite honestly I found the post interesting even though I might not agree with it, the algorithm the computer uses to generate the crews is way to complex for any of us to figure out or to be consistant but for those who insist it is random let me point out that by its very nature a computer is incapable of generating something that is totally random. Don't let the trolls get to you man do your posts on the thing you feel are fitting

    1. RNG's use a basic percent rolling method, picking 1 number out of 10 is the same as 10%, the format doesn't really matter. If you have another possible solution to forum's outrage, please tell me. I'm doing another post like this, but with more data, in a month. I'll add it in. Thank you

    2. Honestly I only half pay attention to the forums. Its largely populated by people who would bitch if you gave them a million dollars because it wasnt 2 million or by people who would be giving Doomrooster a colon exam headfirst if he stopped too fast lol

  5. I usually roll 300k of U to try for skull / GM etc and see the "set" roll thru many times; in the past 6 months I have not rolled a Skull and only 4 GM

  6. What would be interesting to know, is whether it's a double roll or not. I'm guessing it is, i.e. if the roll "hits" on legendary, it rolls again to figure out which legendary. That might explain how DOS is rare AF (assuming it's still possible to get), having a 1% chance to hit legendary, then a 1% (or lower) chance to hit DOS. Actually come to think of it, I'd put money on double roll, as I seem to get a legendary crew every X uranium or so, but it's often one of the less worthwhile ones. I highly doubt blizzard coded a .138% chance to drop unless it's just a by product of the rand units/grouping they used, but I don't see a obvious pattern, everything that is tidy and would make sense is just a bit off (uncommon @4% each, common @5% each, leg's sharing a 1% chance would leave rares sharing a 3% chance, which seems too high, for example) maybe it's a 2 roll system and they just ran with the awkward numbers, or maybe the rand is just really high. It would be cool to know tho.

  7. Appreciating the time and energy you put into your website and in depth information you present. It's good to come across a blog every once in a while that isn't the same old rehashed information. Fantastic read! I've saved your site and I'm including your RSS feeds to my Google account. rainbow six siege boost