Sorry for digging this thread back up, and everything has pretty much been said already, but I'll stick my oar in anyway
Even though I'm sure everyone here at CM understands the concepts of variance/volatility perfectly well, my personal feeling is that sometimes human emotions get in the way of rational thought, and we start to see things that don't exist, build conspiracy theories in our minds, decide to focus on events that reinforce those beliefs, and so on.
The thing to remember is that when it comes to variance and expected results, slots are completely off the scale compared to other casino games/gambles.
At one end of the scale we have the infamous English Harbour incident, where a sample of just a few hundred gambles was enough to prove that a double-up game wasn't, ahem, 'working properly' in VP.
Slots are at the other end of this scale. Even low variance slots tend to have a couple of big wins on the paytable that you may very rarely, if ever, get to see. (Case in point is Boogie Monsters, a game I've been playing on and off for over four years, and yet it was only a couple of weeks ago that I got five Boogie Monster symbols on the reels, and also five scatters - and these two events happened pretty closely together too.)
When it comes to medium-high variance slots, the paytable is made up of hundreds of thousands of possible outcomes (do we have any actual numbers? I suspect it may be even higher than that) - so you can literally play these games for weeks and months, and still be seeing significant dead patches, hot runs, and everything inbetween.
As such, slots can often 'feel' cheaty even when they're nothing of the sort. (And yes I've been there and I've had those thought processes same as most players I'm sure.)
As a low-roller who isn't adverse to taking up big bonuses with significant WRs (I'd say my deposits are split 50/50 between bonus play and non-bonus play), I generally try to meet WRs by playing tens of thousands of spins at low stakes. (I'm not a bonus hunter/abuser as I'm not looking to find a mathematical advantage against the casino, I expect to lose long-term, and the WRs don't really exist these days where you'd expect to do any better than get your deposit back on 'average luck' anyway, I just enjoy playing the games and extending my play sessions where I can.)
Anyway, the point is that I've seen higher variance slots easily chew through £300+ playing at 30p spin with nothing in the way of a decent win or bonus round - (so over 1000x stake, if you're playing at say, £5/$5 per spin, get ready to say goodbye to £5000/$5000) - before dropping a big hit, and as is the nature of random numbers, sometimes a big hit or two close together.
My feeling when it comes to MG slots, and this is based on over four years of play and god knows how many hundreds of thousands (millions?) of spins, is that they work correctly.
Of course they're horrible when they're dead, the medium to higher variance slots in particular can be brutal, but this is the nature of random numbers. I've had really good results - (two £500 hits on Hitman at 90p stake within a couple of hundred spins of each other, and a five scatters during free spins on Summertime (Thunderstruck clone) for over £1000 from a 90p stake over the same weekend) - and they didn't make me think 'This game is broken! It's paying out too much!'
I've had really bad results too - (I remember getting absolutely leathered by one of the 5 payline basic reel slots one evening, over £1000 lost during an extended session....., and a couple of £500 down sessions as well, even with stakes of £1 or less) - and the thing is that at those times emotion kicks in and heart rules the head, 'The game must be cheating! It's not done this before!'
And what I do after those sessions is take a little break, and think back rationally and realise I just got beat by randomness and the house edge, no need for conspiracies, no need for accusations of cheating - the house just took their 5% like they always do, and I had a run of bad luck on top of that.
Ultimately if I really thought the MG casinos were cheating, I simply wouldn't play there - end of story.
Now that is not to say that they couldn't do things better, I honestly cannot understand why MG (and other software providers) don't rate the variance of their own machines, a simple scale from 1-10 or even 1-5 would do - along with an explanation of variance built into the help files, or the FAQ on the casino websites. Make the information available to players, give them a chance to make an informed choice about what style of game they wish to play before they start.
(As an aside, I think a lack of understanding of variance, and of how a long-term theoretical RTP actually works, is one of the main reasons new players tend to bust out from small bankrolls and stakes that are too high for their bankrolls, whilst not even having any clue as to the variance of the machines they're playing - and these players will most likely never come back.)
Also, provide the long-term theoretical RTP for each and every game, and display it on the pay table. I can see no valid reason whatsoever for them not to do this.
TL
R? Slots variance works both ways, it can be a blessing and it can be a curse. We don't call 'cheat' when it works in our favour, so we should think twice about calling 'cheat' when it goes the other way.