As a slot manager at a large Midwest casino I enjoy reading your column every chance I get. There are many "slot experts" out there but you are by far the most accurate I have seen on the Internet. I especially like your analogies using coins and dice to explain the randomness of random number generators and the predictability of overall game outcome.
I thought would write to weigh in on one of your columns: Common Experiences with Random Events. Your response to Gamblin Flatlander was right on the money. If a coin toss was used as the RNG for a game, there are two possible outcomes: heads or tails; yes or no; win or lose. The randomness comes from not being able to accurately predict the outcome of the coin toss. The predictability comes from knowing in advance what the results will be once the coin toss is complete. The coin doesn't care if 'heads' is a winning or losing outcome; it's just a coin.
This is an oversimplification of a true RNG, but the end result is the same; a random event is determined with no regard to the ramifications of what that event means. The RNG doesn't care; it's just a random number generator.
Tom's question about leaving a machine after winning a royal flush reminded me of when I was a young slot tech in Laughlin, NV during the mid-eighties. Back then, fully electronic games were still a fairly new technology. Slot managers in the early to mid-eighties that "came up" during the sixties and seventies when all slot machines were mechanical or electro-mechanical in nature didn't fully trust these "newfangled contraptions".
Management at the casino that I worked for at the time had a policy that required us to thoroughly check any video poker machine that had hit a royal flush jackpot more than twice in a month. It was explained to me at the time that the odds of hitting more than one or two of these large jackpots in a 30-day period was so remote that surely "either something is wrong with the machine or someone has gaffed it." Of course, we never found anything wrong. Even on machines that had hit two or three royal flush jackpots in a 24-hour period we found no problems.
After several months this practice was discontinued. Management finally came to accept that these electronic machines worked like they were supposed to. The frequency of royal flush jackpots is not determined by an arbitrary number of days, but rather by the volume of play they get. A game that is played more often will, on average, hit more jackpots.
As in the example you gave in your response, players have an equal chance of hitting a royal flush on any given hand. I once saw a player, after being paid $4,000 for a royal flush on a dollar poker game, get a straight flush on the very next hand! It can and does happen. There's nothing wrong with the game when this happens. It's just one of the many possible outcomes due to the randomness of the RNG.
In response to Gene's comments about "resets" of the RNG causing a hesitation in the game: if the RNG seed value is changed from time to time (and this would be dependent upon the way the RNG programming code was written), it would have no visual impact on the operation of the game. What I mean by this is, if or when re-seeding of the RNG is happening, it is going on behind the scenes with respect to the operation of the game — you wouldn't notice it.
There are other factors that can cause an apparent hesitation in the game play, one of which is a built in pause. Sometimes a pause is programmed in to slow the game down a bit. Why? If the game takes your money too quickly, your perceived value (enjoyment) of that game is reduced. The win/lose outcome won't be affected, just the speed at which the game takes your money.
This isn't done by the casino, it's done by the manufacturer. Here's why: you take your $$$ to your favorite casino and get a better overall playing experience (more playing time) when you play brand X slot machine than when you play brand Y slot machine — you will be more likely to play brand X slot machines when you visit the casino. Later, when I am making my slot machine purchase decisions, I look at which games are the most popular, i.e., profitable. Brand X is likely to get more space on my gaming floor!
Lastly, I like your analogy about Sammy and Sally Slotdirector. Some slot managers are still making their payout percentage decisions based on denominations. However, the more enlightened among us make those decisions based on the average wager of a game. If the average wager on a two-cent slot machine is around 50 cents, then it makes sense to use a payback percentage in the same range as a 3-coin quarter game.
To me it make sense to give today's two-cent player the same payback as yesterday's quarter player so long as the average two-cent player is wagering the same amount as a quarter player on a 3-coin game. If my games take your money too fast, why would you continue to play my games? As a slot manager, it's part of my responsibility to give my players a reasonable value for their gambling dollar. If I don't, they'll take their money elsewhere.
Sorry for being so long winded.
Rick the Slot Guy
Thanks for the kind words about my column and thanks for sharing your experience and knowledge. I wasn't aware that some manufacturers program in a pause during play to stretch a player's playing time.