John Robison Gaming Guru

• John Robison Homepage
• John Robison Archives
• Books by John Robison

Ask the Slot Expert: Triple Double Bonus video poker and volatility

Question: Here is the info I should have provided [about my unusual video poker experience].

At 25 cents played per hand, this is the first time I have had a session $1000 or over. I have had a number in the $600 to $700 range. I do not keep records, so do not know the actual number. (I also do not keep "net results" per session.) I play twice a week sessions with one seven hours and one five.

Thee $1000 session was also strange in that four of the quads I hit were aces with the kicker at $200 each.

When playing $1.25 a hand for a long time, I on occasion would have gross wins around $5000 per session, with a lot of them in the $3000 to $4000 area. Of course, I also had the sessions with much, much less than that.

Triple Double Bonus is a very volatile game.

Yesterday, during a 25 cent/hand five-hour session I did not have a quad for the first three hours. During the next two hours, I had a number, including one with aces and kicker for $200. I ended up with a gross win of $400.

I have played this game for a long time. My experience has been that unless you are ready to have a lot of patience, and not switch games back and forth, and not give up early, it is best that one not play it.

I am curious what your reaction is to that strategy.

Answer: This is a follow-up to my column two weeks ago about my reader's unusual video poker experience. He won over $1000 playing Triple Double Bonus at 25 cents per hand. I correctly speculated that he must have gotten many quads and that many of them must have been premium quads.

I also said that I was not as qualified as my reader to comment on how unusual his experience was. I have never analyzed the paytable and I have played only a tiny number of hands of the game.

I have to admit that I'm a bit confused with terminology. First, "session". To me, a session is a period of playing similar games with only short breaks. If I play NSU for a few hours with only a bathroom break, I would consider that one session. If I play for a few hours in the morning, then go to the buffet for lunch (Oops! I forgot. No buffets now in the casinos I go to.), and then play some more, I would record two separate sessions.

After playing NSU for a couple hours this morning, I played four different slot machines. I could have recorded that as just one session playing Electronic Gaming Devices, but I recorded my play as two sessions -- one NSU and one slots.

The IRS doesn't define what a session is, so we are free to use our own definition, as long as it's logical and we use it consistently.

My sessions are shorter than yours. We just have to be careful in assuming that we have the same definition of a session. Ideally, we would specify the number of hands played. Time on a video poker machine is measured in hands, not ticks of the clock. The next best metric is the length of time played, from which we could estimate the number of hands played, but our estimate may or may not be accurate. All in all, "session" is meaningless.

I'm confused about "net results" and "gross win". When you said you had a session of $1000, doesn't that mean that the amount you cashed out minus the money you put in was $1000?

In any case, I was planning to discuss volatility this week, so your last comment is quite timely.

I saw something surprising about volatility looking at one manufacturer's data sheets for its slot machines many years ago. In addition to a volatility index, this manufacturer provided a handy chart for each game. The chart showed the number of ways to make each payout listed in the paytable. Each payout had a bar above it. The taller the bar, the more ways there were to make the payout. The value of the payouts increased going to the right.

Following are two sample charts. They are not based on real payouts and probabilities. Don't read too much into the height of one bar compared with another. A payout with a bar that is twice as high as another isn't necessarily twice as likely to hit, just more likely. I want you to just look at the general shapes of the charts.

Which machine do you think is more volatile? The machine on the left or the one on the right?

The chart on the left reminds me of the profile of a ski-jump ramp. The chart of the right looks much tamer.

The chart on the left is from the machine with the lower volatility. There are many ways to hit its low payouts, so the machine pays its low payouts frequently and it keeps you in tray money to keep you in the game. The higher payouts hits less frequently.

The low payouts on the machine on the right, on the other hand, don't hit that much more frequently than the higher payouts. The slot designers took a bunch of the low payout hits and moved them to higher payouts. But the swap isn't one for one. They might have swapped 1000 1-coin hits for one 1000-coin hit. Large decreases in the heights of the low-payout bars yield only small increases in the heights of the high-payout bars.

The machine with the chart that looks like what you see at the top of the first hill on a roller coaster actually gives you less agita about your bankroll than the machine with the chart that looks like smooth sailing.

On video poker, the designers can't alter how frequently certain hands are hit. They can only alter what is a paying hand and how much it pays. (A player's strategy does have a small effect on how frequently some hands are hit.)

I checked the probabilities of getting a high pair and two pair for 9/6 Jacks and Triple Double Bonus on the Wizard of Odds site. The probabilities are about the same for each, about 0.21 for a high pair and about 0.12 for two pair. Triple Double Bonus (TDB) pays for the premiums on the quads by decreasing what you get for two pair, not by decreasing how frequently you hit two pair.

You still get two pair on TDB about as often as on 9/6 Jacks, but they pay half as much so they don't increase your bankroll. If you don't get your share of the premium quads that the decreased payout on two pair are funding, you'll probably have to feed the machine often.

I've compared video poker and slots to a Bizarro world roller coaster. In the normal world, you climb very slowly up the first hill on a roller coaster and then you plummet down.

On the Bizarro world roller coaster, you go down slowly and then shoot up quickly.

You can only lose as much as you bet on each play, so your bankroll decreases slowly. But then you could hit a big jackpot and your bankroll could skyrocket in one play.

In Jean Scott's The Frugal Video Poker Scouting Guide, she and Viktor Nacht give TDB their highest volatility rating, $$$$. In contrast, NSU is only $$ and 9/6 Jacks is $.

Jean once told me that one of the reasons she likes Deuces Wild over 9/6 Jacks is that hitting four deuces can lessen the sting of an extremely bad session -- or even put you into the black -- while 9/6 Jacks has nothing similar. It's just not likely that you're going to hit enough quads and straight flushes to save a really bad session.

If you're going to play TDB, you have to have both patience and bankroll. You need the bankroll to see you through the cold spells when you don't get your share of quads. (Similarly, I've had plenty of times when I've put $100 into NSU, played 20-25 hands, and then had to feed the machine again!)

You also need patience because the high volatility means it will tend to take more hands to home in on the paytable's long-term payback.

Some advantage will use the high volatility to their advantage. Say a casino has two paytables with roughly the same long-term payback. One machine has higher volatility than the other.

I would probably choose the lower volatility paytable, but they might choose the higher one. Because some casinos suspend your mailer after a winning month, they would rather have many losing months and a few big months in which they win big than fewer losing months and more winning months in which they win less. They're trying to concentrate their wins in fewer months.

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

I think there is a problem with the CDC's number for total cases in Nevada. New York Times has the same number. NVHealthResponse.nv.gov reports 428,934 cases as of 10/11/21. I doubt there were 10,000 new cases from 10/11 to 10/12.

Send your slot and video poker questions to John Robison, Slot Expert™, at slotexpert@slotexpert.com. Because of the volume of mail I receive, I regret that I can't reply to every question.

Copyright © John Robison. Slot Expert and Ask the Slot Expert are trademarks of John Robison.

Ask the Slot Expert: A casino at Belmont Park?

Ask the Slot Expert: An unusual video poker experience?

Ask the Slot Expert: Where were you when the lights went out in the casino?

Next 10 Articles >

• Featured Articles

Hit Frequency, Reno, Atlantic City, Double Up

Slot Machine Pay Off

Declare Your Independence from the Slots

Guide to Winning at Slots: The Best Number of Coins to Play