Ask the Slot Expert: Rebutting an article about why slots are not random - Part 1

I found an article detailing why someone believes slot machines are not random while researching last week's column. The writer makes the common mistake of assuming that everything about random events is completely unpredictable.

The article looks on solid ground. The writer recaps what the casinos tell you:

• When the play button is pressed, the RNG randomly selects a winning or losing combination from among millions of available combinations.
• Slots don't get hot or cold - Slot machines have something in common with goldfish: they have no memory. Every spin is independent of the previous and following spins.
• Every spin is considered a brand-new spin, with an outcome of any possible symbol combination.
• Slot players are said to either win or lose; they never "maybe win," "maybe lose," or never "almost win."

So far, so good. There aren't necessarily millions of combinations possible on a given machine, but that doesn't affect the truth of the rest of the statement.

The heading of the next section is: I call BS! - Why I Believe Slot Machine Games are Not Random

Slots are not random, and I will never be convinced otherwise; I have been playing slots for many years and seen so many examples of them being NOT random, making it incredibly difficult to believe that slot machines are random.

Let's look at some of the examples he cites. Two say the same thing, so let's deal with them together.

If slot machine games are designed to fit into a math model, which they are said to be, then they cannot be TRULY unpredictable.

How is it possible for a slot machine to be a random device and for a machine also to have to pay back a certain percentage of the money played through it?

At the Super Bowl (sorry, Big Game -- calm down NFL lawyers) they just used a coin toss to determine which team gets the kick/receive choice. Isn't a coin toss used because a fair coin fits into the math model of 50% chance heads and 50% chance tails? Doesn't the coin toss fit a math model and isn't it also unpredictable?

Let's repeat this fair coin toss 1000 times. Don't we expect to see that about half the tosses were heads and about half tails? Isn't this expected outcome predictable even though the coin toss is random?

There's a lot we can know about random events.

Say you have a snack bag of 25 M&Ms (I'm sure it's just a coincidence that I'm using a candy theme for this example and I'm writing it on Valentine's Day.) and the bag contains 5 red, 8 green and 12 blue. You pick an M&M at random from the bag and eat it.

Don't we know that in the end you will have eaten 5 red, 8 green and 12 blue M&Ms? We don't know the color of the M&M you will draw next, but the mix of M&M colors you will enjoy once you've finished the bag is totally predictable.

Now let's say that instead of eating the M&M you draw, you put it back in the bag after noting its color. I'm going to switch from M&Ms back to the ping pong balls I usually use for this example because I can't imaging drawing an M&M and not eating it.

Don't we expect to see the ratio of red ping balls drawn to the number of draws to be close to 5/25, and the ratio for green to be 8/25, and the ratio for blue to be 12/25?

Let's upsize the example one more time to a slot machine. If the reels are laid out such that the three symbols needed to trigger a bonus round appear on 0.7% of the possible combinations on the machine, don't we expect to get a bonus on about 0.7% of our spins?

The concept here is that the characteristics of our draw results get closer and closer to the characteristics of the population from which we are drawing as we make more drawings.

If we've done the math, looked at all the combinations on the reels and added up what the winning combinations pay, and calculated that this machine pays back 95%, the machine will return 95% of the money played through it because the characteristics of the combinations that land on the payline get closer and closer to the characteristics of how the symbols are laid out on the reels as the machine gets more and more play.

What's unpredictable is what will happen next. The big picture, on the other hand, is predictable.

Let's turn this argument around. If the paybacks on machines were unpredictable, why would casinos have them? Casinos don't gamble with their money. How would the pay for volcanoes, guitar-shaped hotels, fountains and pyramids?

In the beginning of the article there is a disclaimer that "the article contains opinions only from has been experienced from playing slots...."

That's a shame. A little bit of research on randomness and random events and the article could have contained facts and been instructive.

Next time we'll look at some of the writer's experience that disprove randomness.

Click here for the latest Covid data.