Ask the Slot Expert: Does the RNG keep spinning while a machine is not being played?

Question: When a slot machine is not being played, does the RNG still keep spinning or does it spin only when the button is pushed?

If it's spinning continuously, then the time one waits before each hit or pull could make a difference, right? If that's not correct, then how is it possible that the win/lose is picked the second one pushes the slot button?

Also if everything is random then why are there more wins when playing higher denominations? For example, if I play, say fifty cents, the lowest amount possible, the bonus comes up very frequently, but when I move up to $3.00 the bonus rarely appears.

While I would like to believe it's all random, there are just too many things happening which make me think otherwise, not that it makes a huge difference since I go to a casino for enjoyment and with a budget, being sure to leave all my debit or credit cards at home, which also causes me to play more cautiously since I want a long night of enjoyment.

By the way, am I the only one who watches on YouTube the high rollers who actually put $100,000+ into a slot machine at $10,000 a pull and wonder: Where do people get all that money? Is it illegal drug money? And even if one is legitimately wealthy, is it moral to spend that sort of money in a casino when just one single pull could feed thousands of starving people around the world? OK, enough moralizing.

Answer: The RNG constantly generates new numbers. It doesn't take a break when a machine is not being played.

That was not the case with early computer-controlled machines. They wouldn't generate a new number until someone pressed the Spin button. Because the RNG is just a mathematical function that takes an input and calculates a number using various mathematical operations, these machines were vulnerable to being cheated. Cheats with knowledge of the RNG function were able to figure out what the result of the next spin would be, so they were able to bet the minimum on losing spins and the maximum on winning spins.

The solution was to have the RNG constantly generate new numbers. Now the number generated when you hit the Spin button determine the results of your spin. You have to hit the Spin button at just the right time to have a winning spin. It's impossible to time hitting the Spin button so you have a winning spin.

Or is it?

Remember the Russian Slot Hack from a few years ago. A group of cheats -- again with knowledge of the RNG function -- figured out a way to know exactly when to hit the Spin button to have a winning spin. Their bizarre playing behavior led to their being caught. They would sit at a machine and wait. Then hit the Spin button and wait. Wait. Quickly hit the Spin button and wait. Wait. Wait. Spin.They didn't play with a rhythm. There was always a random length of time between spins.

Here in Nevada the RNG must generate an average of 100 numbers per second.

When you press the Spin button, the program running the slot machine gets numbers from the RNG in a process called Polling the RNG. These numbers, one for each reel, tell the program where the reel should stop on that spin.

It's true that when you press the Spin button has a huge influence on what you get for that spin. A losing spin might have been a jackpot winner if you had taken a sip of your drink between spins and had hit the Spin button a few seconds later. Conversely, a winning spin might have been a loser if you had played more quickly or more slowly.

Your results will be different, but not necessarily better or necessarily worse. Just different.

If you want to think of the RNG as a spinner in a board game, this is the most accurate analogy. The spinner is spinning and stopping, spinning and stopping, 100 times a second. Most of the time, no one is watching to see where it stops. Once the player hits the Spin button, we watch the next few spins of the spinner to get a number for each reel. Once we have our numbers, we stop watching the spinner. The spinner, though, keeps spinning and stopping, spinning and stopping. It doesn't care -- or know -- if someone is watching.

Why might there be more wins when you play higher denominations? As a general rule, the higher the denomination, the greater the long-term payback. A Double Diamond machine, for example, may pay back only 92% at quarters but 95% at dollars. Sometimes the slot designers lay out the symbols on the reels so that the hit frequency (the percentage of spins that pay something) is higher on the higher denomination machine than on the lower denomination machine.

What about multi-denomination machines? They can change reel layouts with denomination. In Nevada the help screens on the machine will say that different reel layouts are used for the denomination choices. The game screen itself will also say which reel set is in use.

Why don't you get the bonus as frequently with a higher bet? What I have heard from other players -- and experienced myself -- is being red hot with bonuses at a low bet. Once we've played a large number of spins and we're convinced that we're going to win so much that the casino will have a difficult time making payroll this week, we up are bets to really show the casino who's boss. After a brief cold spell at Max Bet, we drop down to protect our winnings.

Two things are happening here. One, sample size is radically different. We have many spins at the lower bet and far fewer at the higher bet. We can't really compare our results.

Two, randomness. We might have been lucky at the lower bet and unlucky at the higher bet.

And C, insufficient sample size. Hitting the bonus is a rare event. We need a large sample to guess how frequently it hits. A better comparison with small sample sizes is hit frequency. Did the spin pay something, yes or no? We need only a few hundred spins to compare the hit frequencies at the different bet levels.

The observation that makes you doubt the randomness of the machines is actually an example that shows randomness. Let's say you always got one bonus for every 100 spins you played on a machine. Something isn't random. If hitting the bonus were truly random at 1% of spins, you should sometimes go 100, 200, 300 or even more spins without hitting a bonus. And you should sometimes get two or more bonuses in 100 spins.

There was a gambling device at the turn of the century (19th not 20th) that paid out every 14 plays. That was built into its machinery -- 13 plays, jackpot, 13 plays, jackpot, etc. The machine didn't last long because it didn't take long for players to figure out the pattern.

I think you can be assured that the results on a machine are determined at random. Note though that my descriptions above are for Class III machines, like in Las Vegas and Atlantic City. The results on Class II machines at Native American casinos are also determined at random but using a different process.

Finally, I didn't know that there were videos of $10,000 pulls on YouTube. I watched a video of a few $100 pulls today. That's much more than I'm willing to risk on a spin. Much cheaper to do it vicariously.

I doubt those whales are betting illegal drug money. Nothing like throwing money around to draw attention to yourself.

Is it moral to bet that much? Well, is it moral to spend over $300,000 on a car? How about millions for a private jet or a yacht?

How do know that the $10g bettor doesn't give a great deal of money to charity?

I don't know whether it's moral. I do know that he's not sitting in front of the casino burning hundred-dollar bills. The money he loses goes to paying salaries and overhead at the casino.

Dr. Fauci was on with Wolf Blitzer this afternoon. The Blitz asked, "We're both baseball fans. What inning would you say we're in?"

After some dancing around a non-answer and some prodding, the Fowch finally said, "Something like the bottom of the sixth."

He should have said, "That's an inane comparison. A baseball game can only move forward to a higher-number inning. We can lose ground against the virus. Just look at India."

Here are the latest figures from https://www.cdc.gov/covid-data-tracker/#cases_totalcases.