i've had similar situations occur even on public GAs too, but adding in the fact that it was their very first two GAs makes it a pretty unusual (and pretty funny) case. xD

all RNGs are nonsense imo...

there is no "random", but that would be too much math and physic...

i had the luck to win 2 games on a single train too but told the creator he can reroll one if he wants... but he was ok with winning both...another time i won around 6/7 games in around 30 hrs and sometimes you don't win something for month...

Yes, I know there is no real randomness with computers, but take a train with 10 giveaways.
If 100 users ride this train, every single giveaway in that train will be entered by a subset of these 100 users.
Now if you take 10 completely unrelated giveaways, there might not be a lot of users who entered more than 1 of these giveaways, there could be thousands of different users who entered these 10 giveaways (let's assume for a same total number of entries than the 10 giveaways in the train). So of course it will be more likely to have a same winner on 2 giveaways in a train than in random giveaways.

I'm leaving this here: https://www.random.org/randomness/

What Sighery suggests here is not only true but also the worst nightmare of any coder once tried to produce a bunch of real random numbers in a tiny amount of time. One hack/solution is to recreate the seed before each number generation, which is very expensive in terms of server and cpu loads, considering a system like SG needs a new random number every __ ms.