I’m playing in a game and invested heavily in experimentation early on. So far there have been 7 galactic cycles and my scientists have unlocked experimentation 5 of those times- the last 3 in a row!! The other two were in terraforming. This seems way out of balance. If it is really random maybe I should go to vegas.
As far as I know and have read here, yes I believe it’s supposed to be completely random.
I would love to know what algorithm is used to decide it though!
I just use pythons radom.choice method to pick a tech from a list of all techs.
You should definitely buy a lottery ticket.
first game, i just got experimentation 3 out of 4 cycles. exp, bank, exp, exp… algorithm definitely needs a change.
a co-worker has gotten exp, bank, exp … so far.
I posted this on the G+ page in response to a suggestion on the issue:
An alternative solution would be to decide how many hits in the same thing you would accept in a row before you got a hit in everything else.
So if you say experimentation must give you a hit in everything once before it can cycle again, every 7 cycles you’d take the list [1,2,3,4,5,6,7], randomize that, and pull the next tech out of the list each turn.
If, for example, you wanted to allow for the possibility of hitting 2 in a row you’d do the same thing, except every 14 cycles and using a list like [1,1,2,2 … 7,7].
I think that would be kinda nice, and would lend a little strategic predictability to experimentation.
1 Like
To echo jon/Dilbert’s comment, random is random.
If you get HyperSpace 5 times in a row, the odds are still 1 outa 7 you’ll get it again. I’ve played enough that I’ve seen (mostly) different tech each time … and below is a screenshot (click to see big) where I got Weapons five outa six times … but a teammate (it was a team-based game) was our Weapons dude … 
I think Experimentation is one of the better aspects of the game (and don’t think the random nature should be changed) … although its importance diminishes later one.
BTW, I’ve heard rumors that Experimentation might be biased for the AFK’ers … if so, I actually think that is a GREAT move.
AFKers are not cheating on Experimentation, but thats a good idea because it would mostly hidden and could be chalked up to randomness
1 Like
Without tipping the scales too much, I’m all for giving the AFK’ers some advantages.
I.e. being able to trade even if not in Scan range, making moves/trading/etc. intra-turn, and yea, biasing experimentation. That is until you get the full sentient AI working and they start attacking players to take over the galaxy! 
But you still want noobs to eventually become veterans, right ?