Slots Strategy - Sweepering

[Pigsy]

Having now spent the last couple of months doing Ladbrokes promos and wagering through the bonuses ( my first ever Casino effort), I like many was caught out with a bonus balance and 20 x WR after the Moneybookers changes.

Having now dabbled with slots, has anyone used a sweepering card type strategy on slots. I am just a rank novice on this subject and I know they do carry their own risk but could a similar strategy help in reducing the variance?
As I understand it the return from a slot is between 90-95% but as I see it most slots are very dependent on the bonus rounds. If you fail to hit the bonus rounds then the damage can be much worse.

So can sweepering reduce this damage? For example if you played a slot at 50 x 20p spins and you fail to hit a bonus, then play the next 50 at 40p spins. If still no luck then play the next 50 at 60p and so on.
This depends on the slot and the bonus structure, but in theory the larger the spin value the larger (potentially) the bonus win. I have no idea what the expected return over the long term on this would be and there is still the chance of going 500+ spins without a win for a heavy loss. But if you have a large pot and can withstand a few hundred spins without bonus then is this a strategy that could help.

I tried it on demo with Cashville and grew my 200 balance to about 250 before going on a run of 540 spins without the bonus round. At 500 spins I had maxed out the bet to £2 a spin and was just playing through when I hit the bonus at about 520 spins. I won £200 bonus which took my balance to £300, but could have been unlucky and only returned £40 from the bonus.
If there was a slot that had statistically fewer average spins between bonuses and a more balanced bonus structure, would this strategy help.
Feel free to tell me if I am barking mad.

-snip-
I've moved your post to the Casino Strategy forum, thanks for posting on the questions thread but this is a topic that comes up time and time again but always it seems as an off topic post within an existing thread so it's always hard/impossible to find those threads and say 'here read this!'! Would be very useful to have a dedicated thread for discussion of slots sweepering 'systems'. If anyone knows of a dedicated thread that already exists, I'll merge it with that (I couldn't find one, hence the new thread!).
-snip-
--munk

[Gibber]

sweepering makes no difference whatsoever as the chances of hitting any feature, big win etc. Is exactly the same throughout. Plenty of us do increase stakes while playing slots bonuses though after not hitting anything in a while because it can feel like your owed a big win, it does seem to come off every so often and it is a good feeling hitting 5 scatters on £2.25 a spin . In reality though to reduce your variance you should always play min. coins.

[munk]

As I said in my PM to you Pigsy, there have been a lot of 'mini' debates about this in various threads, but I can't find any of them now! The only one I found is this:

There's no real strategy you can use to 'beat' the slots because they inherently have a 5-10% house advantage. On the face of it you might think that increasing your stakes if you go for a long period without a bonus round might increase the payout you receive if you hit a bonus round at the higher stakes. However that's really an illusion because every spin is independent of every other spin, so if you go for 500 spins without a bonus round, on the 501st spin your chances of hitting a bonus round are still around 1/150, statistically the fact you've gone for 500 spins without a bonus round shouldn't count for anything (because each spin is independent).

It's essentially the Martingale 'system' you're describing - again with that system any gain you make from employing it is only illusional, statistically every new hand you receive playing blackjack is independent of every other hand.

BUT having said that using a sweepering strategy can decrease the tedium of just playing flat stakes and make things a bit more interesting.

Also have a look at KK's site (casino forum moderator here):

SlotBeaters.com - Best online slots, hints & tips, guides, slot statistics, reel layouts - reelbands, slot percentages

also this thread might be of interest:

http://www.thegamblingtimes.com/boar...-strategy.html

which bizarrely was actually a reply to a question you asked (click the little read > above to see the thread). This is the kind of thing that people do ask about fairly regularly, but unfortunately there's no dedicated thread about it that I can find anyway... so is useful to have this.

So, I would stand by what I said above, any perceived gain from employing any 'system' when playing (fair/random) slots is just that, only a perceived gain. Each spin is independent of every other spin and the machine has no way of 'remembering' what came before (well, apart from perhaps some jackpot machines or pub type slots?), your chances of hitting a feature don't increase the longer you go without hitting it.

So for example if you have a 1 in 350 chance of hitting a feature and then you go for 700 spins without hitting that feature, on the 701st spin your chances of hitting the feature are still 350/1. The only thing that has changed is the level of variance you are experiencing.

On the other hand, some people might be inherently lucky and so using a system will appear to work when in fact all that's happening is that they have been lucky with the variance involved, so if they'd just played with flat bets then they would have won just the same over time.

All of this is bordering on the philosophical and is reminiscent of this thread which is definitely worth reading through if you can spare the time:

http://www.thegamblingtimes.com/boar...ndy-right.html

In that thread there's a discussion of whether variance ever actually completely evens itself out over a long enough period and the last page or two are quite interesting.

[Pigsy]

Interesting read, I do remember your response last time, and I have already got my head around +EV and tracking it over time. It is clear to me that every spin or every turn of a card is entirely independent from the last and has no bearing on the history of your play.
However where I am struggling is the sweepering strategy and whether it can add value in the long term, not over a short period of time. I know it clearly doesn’t as it would be the end of Casinos, I am just struggling to understand why.

Let’s assume you play a slot that has a house edge of 5%, i.e 95% return on investment. So everyone says that to limit variance play minimum spin value. First I have to understand how the house edge is broken down.
Let’s say The Pigsy Slot can be played at 20p spins minimum and has a bonus round on average every 150 spins. Assume the bonus round has 3 win scenarios, £2, £4 & £6 at 20p spins so an average win of £4 based on 20p.
So now say we play that slot every day 50,000 spins per day for 100 days on 20p spins.

If we use the ‘average’ stats:-

50,000 spins x 20p x 100 days = £1,000,000 Invested
Based on 95% return = £950,000 or £50,000 loss.
If the bonus is average £4 every 150 spins, therefore during the 100 days you would receive 33,333 bonuses at £4 a time i.e £133,333 or 14.03% of your return would be based on bonuses and 81.66% of investment based on general play. (950k-133.33k)/1,000,000

Now if we used sweepering techniques for this sample size how would it effect the return.

Every 50 spins we increase our spin value by 20p. If we use average bonus stat of 150 spins then this would equate to 50 spins @20p, 50 spins @40p & 50 spins@60p, or £60 invested. Assume that the bonus return is pro-rata to the spins value, i.e £6, £12 & £18 so average bonus payout of £12 for a 60p spin.

Assume that at the end of each bonus win we revert back to 20p spins and start again, the average spin value would be 40p (50 x 20p, 50 x 40p & 50 x 60p) so the overall investment would be much higher.
50,000 spins x 40p x 100 days = £2,000,000 Invested
Bonus average £12 every 150 spins = 12x33,333 = £400,000
General return of 81.66% = 2,000,000 x 0.8166 = £1,633,2000, total return £2,133,200 i.e a profit of £133,200

Clearly some bonuses would be won on 20p spins, some would be won on £2.00 spins, but the average over a large test should be 150 spins. Obviously my figures are far too simplistic, but what I wanted to discuss is whether increasing your stake affects the outcome in the return long run. Am I wrong in assuming that the bonus features have such an effect, as my brief experience shows that to come out of wagering up you need to hit a feature or two and so increasing the possibility of better feature rewards by higher spin value has some logic.

[munk]

Some great working out there.

Every 50 spins we increase our spin value by 20p. If we use average bonus stat of 150 spins then this would equate to 50 spins @20p, 50 spins @40p & 50 spins@60p, or £60 invested. Assume that the bonus return is pro-rata to the spins value, i.e £6, £12 & £18 so average bonus payout of £12 for a 60p spin.

If the sweepered spin values are 20p, 40p and 60p, there will be 3 lots of feature win scenarios:

2/4/6
4/8/12
6/12/18
(assuming the 3 feature pay levels are stake*10, stake*20 and stake*30 as in the first flat betting example)

so wouldn't the average payout per 150 sweepered spins would be £8, not £12? I think?! My maths could be out though. It is a Saturday night to be fair and I stared at that a bit too long!

Generally though there's no getting around the fact that if the ROI is always 95%, it doesn't matter if you stake 20p, 40p or 60p, you'll still be down 5% all things being equal.

Loving the pigsy slot though! You should submit it to Microgaming.

[Pigsy]

If the average bonus hit was on 150 spins then this would relate to a 3x20p spin value and a bonus of £6, £12 or £18 so average £12. Some bonus would be won on <50 spins (£4 bonus) some on <100 (£8 bonus) and some on £2.00 spins for example (£40 bonus).

But overall the average over a long period should be 150 spins, £12 bonus (I think)

As I said the maths shows that it does have an effect, but I am sure that just like EV, if you have a bad run then the line would run below the expected average.

[KasinoKing]

This "sweepering" system is how I've played slots for the last 5 years.
I have made profit playing slots for each of those 5 years, but probably only due to the number of bonuses I've used. I personally think the system helped me lose less, but I can't really prove it.

I just quickly added up my figures for last year:
2009 - total bonuses received: $15,397
2009 - total profit made: $8,464

$6,255 of those bonuses were from Ladbrokes;
Let's assume WR of Bx4 = $25,021 of wagering.
For the other $9,142 lets assume WR Bx20* = $182,843
(Used *20 even though most WRs are much higher, because with about half the bonuses I bust out before completing WR)
So in $207,864 of wagering I lost $6,933 = a return of 96.67%
These (very rough!) figures seem to show that the method is not a way to "beat slots" on it's own - no surprise there!
Did it help me lose less? No way to say for sure as I don't have $200,000 worth of wagering without using this method to compare it with!

At the end of the day it boils down to do you believe the gamblers fallacy that when an event has gone way past it's probability of occurring that it is more likely to occur, or not.
Mathematics says it's not - the brain says other things!

KK

[Grandthrax]

Sweepering will increase your variance and won't affect your ev.

If the average bonus hit was on 150 spins then this would relate to a 3x20p spin value and a bonus of £6, £12 or £18 so average £12. Some bonus would be won on <50 spins (£4 bonus) some on <100 (£8 bonus) and some on £2.00 spins for example (£40 bonus).

But overall the average over a long period should be 150 spins, £12 bonus (I think)

As I said the maths shows that it does have an effect, but I am sure that just like EV, if you have a bad run then the line would run below the expected average.

You're right in that £12 is the average bonus you'd get, but it would also be your average stake. Therefore massively more variance and no more ev.



But there are ways to decrease your house edge and variance when playing slots which a lot of people aren't aware of.

If the gamble feature counts towards wagering (which it should do unless explicitly stated in the t&c) then as this as an average return of 100% (ie no house edge) it will lower the overall house edge of the slot by 2-3%. So you can end up playing slots with a house edge of around that of roulette. Very manageable. If you gamble more than once then you can decrease it even further.

Even if the gamble feature doesn't count towards wagering. If you use it on small wins and not on large wins then you decrease the variance of the slot.

ps. loving the 'martingale' tag

[Cloverleaves]

Does this gamble feature also bring down the variance on JOB? I notice that a couple of ones I have played at have that option. Either to gamble all or half.

[Grandthrax]

Does this gamble feature also bring down the variance on JOB? I notice that a couple of ones I have played at have that option. Either to gamble all or half.

Umm not sure to be honest. I reckon on 100 hand it lowers variance but raises it on single hand but that's just a guess.

If you're two-tiering then betting the balance on multi-hand vp and using the gamble feature to a target can be a good way to do it. If your aiming for a high target it can also the most ev.