Stay informed with the
Recent Articles
Best of John Robison

# Ask the Slot Expert: American slots and English fruit machines: Divided by more than a common language

17 January 2018

Question: I am an English college student currently studying in London. As part of a research task, I decided to look into the way mathematically true randomness was integrated in slot machines.

As I have gathered, you are an expert in slot machines both in how they work and how to maximise the player's profit from it.

I had a few questions concerning this research. If you would have the time to answer them, it would be greatly appreciated:

To your knowledge, how do slot machines operate: is there a mechanism behind the spin that predetermines the outcome, or is it totally random so the odds between the player and casino are even?

If the outcome isn't random, have you been able to establish a method to determine in advance the outcome of a spin?

In your opinion, would it be better if the machines were random or if they weren't to maximise the player's gain?

And lastly, to do with the addictive side of gambling would you say this happens with slot machines and what is your view on this?

Thank you for taking the time to read my email and to answer it if you come to have the time to do so.

Your e-mail gave me the opportunity to take a memory off the shelf and relive it using my Pensieve. (I figured that a Harry Potter reference would be more appropriate than the Wayback Machine from Sherman and Peabody.)

About 20 years ago (20 years!! Sheesh), one of my co-workers told me about his friend who programmed slots in England. I was just beginning my career as a slot journalist and the company we worked for was a market research company, so probability, statistics and gambling were frequent topics of conversation.

My co-worker's friend said that the slots in England worked differently from the ones in the United States. In the United States, the output from the random number generator (RNG) is used to determine the result of the spin, and the result determined by the RNG must be displayed without any alteration. Back in the 1980s, the early days of computer-controlled slot machines, machines from a company called Universal had an additional step after polling the RNG. If the spin was a losing spin, this extra step discarded the result from the RNG and chose an exciting losing combination from its pool of losers. So, instead of landing blank-blank-bar on the payline, the player might get jackpot-jackpot-blank. Much more exciting. Both results are losing results, so Universal didn't see any harm in the alteration. Regulators determined that the altered combinations gave an incorrect impression of how likely symbols were to land on the payline. All jurisdictions in the United States prohibit so-called "secondary decisions" and require that the result from the RNG be displayed.

The past few weeks we've talked about how machines that don't get much play can show widely varying paybacks from month to month. Machines go through hot, warm, cold and tepid streaks and there's nothing in the programming of a U.S. slot machine to try to correct a streak.

Remember the anecdote I told about the introduction of the Blazing 7s machine? In the first casino in which they were installed, the machine's performed close to their long-term payback percentage. But in the second casino, the machine's went on a hot streak (or a cold streak, from the casino's point of view) and players were hitting combinations of 7s right and left. The machines eventually regressed to the mean and their performance neared the long-term percentage. The casino was happier, the machines were still very popular with the players, and the rest, as they say, is history. Blazing 7s became one of the most popular slot titles and it is the example usually given of a low hit frequency machine.

Machines in the United States exhibit these phenomena because their results are as random as they can be even though they were generated by a deterministic function (the RNG) and the output of that function must be displayed without change.

Machines in England, on the other hand, worked differently according to my co-worker's friend. Those machines, he said, have a supervisor function that monitors the performance of the machine and can force certain outcomes to keep the machine's actual payback from straying too far from its long-term payback. My co-worker's friend worked on the supervisor function. Unfortunately, I never had a chance to meet the friend and get more details about the algorithm.

The Computerization subtopic on the Wikipedia page for Slot Machine has a description for how early computer-controlled slots in the U.K. operated. When a coin was put into a machine, it went to either a cashbox for the owner or the payout channel to be eventually paid to a player. The computer monitored the number of coins in the payout channel. When it was full, the machine displayed more winning combinations. When it was nearly empty, it displayed fewer.

The Wizard of Odds gives a current description of the "adaptive logic" used in the Fruit Machines typically found in pubs (the slots in casinos work like U.S. slots but with limits on bet size and jackpots) in this blog post about gambling in the United Kingdom.

Apparently, the goal of these games is to ensure the proprietor of the game will make money consistently on a weekly basis. The game has a memory and keeps track of how much money it has been taking in and giving out. If it has been overpaying beyond its target return percentage, then it tightens up. Likewise, if it has been underpaying, then it will loosen up.

How do the slots adapt? Despite reviewing a (rather sparse on details) patent titled Control System for Gaming Machines about the process and visiting numerous websites, the Wizard was not able to find the algorithm anywhere and ended this section of the blog post with a plea for anyone who has worked on the process to contact him.

So much for background. Let's move on to your questions.

How do slot machines operate: Is there a mechanism behind the spin that predetermines the outcome, or is it totally random so the odds between the player and casino are even?

Slot machines have a function in their programming called the random number generator (RNG). The RNG function is just a series of mathematical operations like addition, multiplication, etc. The function is more correctly called a pseudo-random number generator because there is nothing random about the mathematical operations in the function. The RNG generates a stream of numbers that appear to be randomly generated and satisfy many tests for randomness, but are not truly generated at random. The RNG runs continuously, generating numbers, even when the machine is not being played.

After the player initiates the spin by pressing the Spin button, the Bet Max button, or pulling the handle, the program running the slot machine polls the RNG (that is, gets a number from it) to determine where the reels will stop. The program typically polls the RNG once for each reel on the machine.

So, there's no mechanism that predetermines the outcome; the outcome is determined solely by the numbers from the RNG. Is it totally random? Sorta. The number stream from the RNG isn't truly generated at random, but it's close enough. The odds between the casino and player are not even. The machines have house edges that range from a fraction of a percent to maybe as much as 18%.

If the outcome isn't random, have you been able to establish a method to determine in advance the outcome of a spin?

Nope. There's no way to know the outcome of a spin in advance — at least, not legally.

Would it be better if the machines were random or if they weren't to maximize the player's gain?

In the long run, I don't think it matters. If the symbol probabilities and winning combination values yield a pool of outcomes that pays back 95% in the long run, I'll get back 95% of the money I play through the machine with or without the adaptive logic. The adaptive logic, though, will act like bumpers in the gutter of a lane in a bowling alley, limiting how bad — or how good — an individual session can be.

All in all, however, I prefer random. I don't want to penalized for a previous player's good luck. But I guess it's also possible that I can be the beneficiary of prior players' poor luck. Nevertheless I think it's cleaner and better for the odds to be the same on every spin for every player. Sounds like a very democratic way of determining the outcome on a slot machine.

It's ironic that the adaptive logic can create the "take" and "pay" cycles that some players used to claim were programmed into slot machines. A slot has to take before it could pay, and then it had to go back to taking to be able to pay future players. Some of the websites the Wizard visited claimed that a player could tell when a machine was going to pay. If that's true, then players might be able to win more by concentrating their play on machines that are in a "pay" cycle.

And lastly, to do with the addictive side of gambling would you say this happens with slot machines and what is your view on this?

All casino games can be addictive. You have four common addictions right in the casino: gambling, booze, tobacco and food. Everything besides tobacco can be and is used or engaged in safely by the majority of people. And only food is essential for survival.

All casino games give intermittent rewards. Slots have the additional addictive factor of the near miss — "The machine is trying to pay you."

A bartender can tell when a patron is drunk and should be cut off. Can a casino identify a problem gambler or a gambling addict? I checked a number of sites with the signs to look for and none of them are behaviors that would be exhibited in the casino. The American Gaming Association has a Code of Conduct for Responsible Gaming, but there's nothing in it about a casino proactively identifying a problem gambler. I found a document online called Problem Gambler Identification Policy from Christchurch Casino, but the general indicators for problem gambling apply to every video poker pro I know.

Some people claim that the casinos have all of the data they need to identify problem gamblers in the player tracking systems. The systems know how frequently we visit, how much we play on each visit and how much we win or lose. But how do you filter out the video poker pros in my circle, who collectively play many millions of dollars through the machines and gamble nearly every day? Whether we win or lose? Then what about the pro who just happens to have an off month or year?

I think it's ultimately up to the players and their families and friends to take action when gambling becomes a problem for someone.

John Robison

John Robison is an expert on slot machines and how to play them. John is a slot and video poker columnist and has written for many of gaming’s leading publications. He holds a master's degree in computer science from the prestigious Stevens Institute of Technology.

You may hear John give his slot and video poker tips live on The Good Times Show, hosted by Rudi Schiffer and Mike Schiffer, which is broadcast from Memphis on KXIQ 1180AM Friday afternoon from from 2PM to 5PM Central Time. John is on the show from 4:30 to 5. You can listen to archives of the show on the web anytime.

#### Books by John Robison:

The Slot Expert's Guide to Playing Slots
John Robison
John Robison is an expert on slot machines and how to play them. John is a slot and video poker columnist and has written for many of gaming’s leading publications. He holds a master's degree in computer science from the prestigious Stevens Institute of Technology.

You may hear John give his slot and video poker tips live on The Good Times Show, hosted by Rudi Schiffer and Mike Schiffer, which is broadcast from Memphis on KXIQ 1180AM Friday afternoon from from 2PM to 5PM Central Time. John is on the show from 4:30 to 5. You can listen to archives of the show on the web anytime.

#### Books by John Robison:

The Slot Expert's Guide to Playing Slots