Slot Machines (the 1-armed bandit type)

Retirement as a full-time RVer is great. You either have lots of spare time to waste, or you're involved in days of heavy, concentrated, mind-exhausting driving, or you're being tested by your rig's many idiosyncracies and hi-tech challenges for hours at a time. Right now I'm in the time-wasting mode, as we're comfortably set up at another RV park for at least a few weeks.

As such, today I read a post on vpFree which had a video of none other than my good buddy and co-host of the Bob Dancer radio show heard weekly by at least ten people--Mike Shackleford. In it he gave a very articulate description of how slot machines operate. I was interested in this because I may have played slot machines ten times in my life, and because they really don't interest me I thought I'd check in with an "expert" to see what I was actually doing when I played them.

Well, the first item made sense. He said the higher the denomination you play, the better the payback, and the crummier and further away from the STRIP the casino, also the better the payback. I did get this. Hip hip hooray! for The Wizard!

Then, in true advantage player form (which he has always said he is--even to me in person when he interviewed me for his site, which would be a highly educational read for any of you) he served up a helping of hypocrisy. He proclaimed that as soon as you hit the bet button or pull the handle, the machine's RNG would randomly select numbers for each individual reel that would correspond to a symbol, and those randomly selected symbols would them appear on the screen for the player's results.

However, he then went on to say that each machine's payback % is controlled by an EPROM chip, such as 95% for instance. So, how can it be that numbers are being randomly selected, as Michael says, AND there is a controlled payback %? Funny, this seems to touch on how I've always said the vp machines operate, only the "experts" like Michael swear everything is random there too--exactly in the same way he states that slot machine operation is random.

Did he stub his famous, math wizard toe here? Please comment.

My understanding is that each reel is not independent. Now this refers to the standard "three reel machine" and not to the new video machines. The way I was told reel machines operated was this:

The RNG would choose among a multitude of pre-set results. Let's say, for the purpose of an example, the machine has one payoff which is AAA on the reels. This particular machine has on its reels only the symbols ABC. The machine is set to deliver one win out of 5, so the following "results" are pre-loaded:

ABC
ABB
AAA (the winner)
BBB
BCC

When you push the button the RNG will choose one of those five pre set results. Because the RNG is fair, each and every push of the button gives you a 20% chance of winning.

Now, there may very well be some new technology that totally randomizes each reel. But if that were the case the the payback on the machine would be adjusted differently by controlling, for example, the numbers of winning symbols on each reel -- actual reels or virtual reels.

Again returning to the example of AAA being the only winning combination with each reel independent: the number of "A's" loaded onto the reels (actual or virtual) would have to match the expected result so that (in our example) AAA would show up 20% of the time.

I don't know if I am making myself clear on this, but you could have the RNG choose from a final set of what shows on the reels (AAA or ABC) or you could set up the machine so that each reel shows an independent result but in the end the machine is seeded with the appropriate number of A's on each reel so that AAA would come up the number of times necessary to keep to the machines expected payback.

Your right on the money, Alan. The difference between VP and slots is the symbols all have an equal chance of appearing in VP but that is not the case with slots. With VP you equally map the same number of random numbers into each card. With the slots some symbols have a more numbers mapped into them. This gives them a higher probability of occurring. While the selection is still random the distribution is unequal. That's why the VP regs. say random and fair and slots only say random.

Also, when I say symbol it could be either one individual symbol or an entire line as Alan indicated. Depends on the game.

A little confusing. Alan said the RNG is fair and that the machine could be set to win 1 out of 5. Arci agreed with him, but stated that the regs in slots say they must be random but not fair. Yet Shack said none of this, instead maybe leaving it open for interpretation--something not expected when we're being told "this is how it operates". Who can put it all together?

Is the audio of the radio program archived on a website? If so, give us a link so we can listen to what was said. Of course one of the problems with "radio" and even with "TV" is that you don't have the ability to review what has just been said you might miss some details. Radio is less effective than TV for communicating complex issues because radio lacks visual support.

With print and the Internet the reader has the ability to read, read again and even read a third time till the information is understood.

I saw it too and I tried to type in the youtube link correctly.

http://www.youtube.com/watch?v=GCf79SyXVHw

I clicked on it and it works.. Look for the Wizard guy in a video titled How Slot Machines work.

The video I got is about how to beat slot machines.

When you are watching anything on the website, simply copy the address in the address bar on the web page. No need to type it, just highlight and copy using "control C." Then paste using "Control V."

Rob: could you post the page to the discussion on vpfree that has this video?

Alan, I clicked on the link he provided and when I schrolled down I came to the video with Michael Shackleford that I mentioned.

Sorry, guys but I don't see anything related to how slots work or Michael or the Wizard. What is the "title" of the video on youtube? I will put a link direct to the video.

I did a search on YouTube and found this interview with Michael from 2010:

http://www.youtube.com/watch?v=J-Tu0K7A2NU

Is this what you guys are talking about?

Yes that's the one.

Slot machines are offered to casinos with a range of payback percentages available. Since Class III slots use the pseudo RNG to determine the outcome of a spin, the payback percentage (simply a function of coin in - coin out) is determined and realized long term simply by the number and type of symbols on the virtual reels or in older machines, the reel strips.

Long term payback percentage will be realized through billions of random results on the newer "money-grabbing, bad bet, guaranteed long-term loss to the player" machines.

There is no manipulation to the machines to reach their RTP %. Once again it is in the math.

The best paper and author on slot math, I believe is John Wilson. Another expert is Mark Nicely, former IGT executive.
Juhn Wilson is a former tech editor for the ACEME who publishes this free monthly online magazine with no fluff.

http://www.aceme.org/

John Wilson's "Slot Math Can Be Fun" is here: http://content.yudu.com/Library/A1ig...--Can-Be-Fun--

If you do nothing else ever, reading this document will explain the subject better than anyone else. It can (and should) be saved to your hard drive and there is a section on the newer videos. Arci, I'd like your opinion on this read if you have the time. And everyone's opinion as well.

John wilson, in writing for the ACEME (Association of Casino Enterprise Manager and Executives) has these published articles on volatility, what purists and defenders of randomness blame losing sessions on:

http://www.aceme.org/articles/januar...r-satisfaction
http://www.aceme.org/mp3-library/vic...-ii-volatility

And other ACEME articles on the subject are on this page:

http://www.aceme.org/find/results/volatility

Mark Nicely is described in a USA Tax Court document this way:

"Mr. Nicely has no published articles because in his industry
anything worthy of publication is a trade secret. There is a
code of silence with respect to sharing information--publishing
would amount to giving secrets away to the competition. For
example, Mr. Nicely has solved a very difficult math problem
associated with a process called “gambler’s ruin”. His
associates do not have this analytical technique at their
disposal, so they have to use simulators. Mr. Nicely’s
analytical solution is very powerful, and he would never publish
it because it would be “spilling the beans” to his competitors."

His website, Nicely Done Gaming is here: http://nicelydonegaming.com/index.htm and is described as "Of benefit to casinos, on-line gaming concerns, game developers, regulators, inventors and their counsel, tax attornies and accountants, Nicely Done Gaming provides a wide range of design, analysis and legal services.

Finally of interest is a recent, important Alan Krigman article published by Casino City Times: http://krigman.casinocitytimes.com/a...-greatly-60179

I'm only supplying these links because they will answer your question thoroughly as well as unasked questions.

I'm not sure if you misspoke or if I read this incorrectly: "the payback percentage (simply a function of coin in - coin out) is determined and realized long term simply by the number and type of symbols on the virtual reels or in older machines, the reel strips."

My understanding that on a virtual machine it doesn't matter how many symbols are displayed because the RNG will control the payouts.

When you refer to "older machines" do you mean mechanical slots that don't have computer chips? There the payouts are again determined by a finite number of pay combinations on the reels.

I'm not sure why you mentioned "coin in" because the number of plays has no bearing on when a payoff or jackpot is hit no matter what kind of slot game you play.

Please tell me if I misunderstood? Thanks.

Alan, I'll try to answer your questions, though I supplied the links because John Wilson literally wrote the book on slot math and he does a better job than I of answering through his articles.

I mention "coin in, coin out" becaust that is the definition of RTP %.

I agree "coin in" "has no bearing on when a payoff or jackpot is hit" or for that matter, any combination of symbols.

We should mention that at the moment, this thread is limited to Class III discussion, as Class II and WA state function differently as do racinos and VLT's.

The history of the evolution of slot math has much to do with the Telnaes patent whick was the basis of virtual reel technology and essentially took the luck and fun out of slots, imo.

"In 1984 Inge Telnaes received a patent for a device titled, "Electronic Gaming Device Utilizing a Random Number Generator for Selecting the Reel Stop Positions." International Gaming Technology IGT bought the Telnaes patent and now all slot manufacturers using what has come to be known as Talneas mapping or reel mapping must license the patent from IGT. This device allowed for the creation of what we call the virtual reel.

While a physical reel may have 22 symbols of stops on it a virtual reel can have 32 up to 256 or even virtual stops. The virtual stops are assigned to one of the symbols on the physical reel."

Quoted from: http://casinogambling.about.com/cs/s...slotreel_2.htm

Generally I don't rely on or endorse the above site or similar sites for technical info but this was a faster way to post Telnaes information.

Alan, let me add that this and other info has been provided to another site, slotmachinesforum.com, where for years I have engaged in debate over this topic. Lots of good videos, pictures, and other info there, too.

Heck, I think I even got Rob to briefly post there a few years back in the VP category though he'll deny it, lol.
I did like his description of the AZ Indian reservations and how tribal casino $$$$ have been spent there, lol.

Thanks Lucky your summary and clarification is good.