Paying tribute to Godfather of video poker
March 13, 2019 3:00 AM
by Elliot Frome
The number 21 is heavily on my mind as I write this week’s column. But it’s not blackjack or any of its variants on my mind.
It’s the number of years since my dad passed away. It seems surreal to me.
Not a day goes by that I don’t think of him. But as the years go by, I sometimes forget how many years it has truly been.
Linkedin reminded me when it posted up that I had been the President/CEO of Gambatria for 21 years. When my dad died, the family decided to keep running the business, which after a relatively short time turned into me, with my mom’s help, continuing to fulfill book orders. It would be four years later that I would decide to turn to gaming analysis as a full-time job.
For my dad, he made his initial mark in the industry in video poker. As the story goes, he was in one casino and saw a video poker machine with a sign above it announcing its payback. Shortly thereafter, he was in a different casino, saw a similar video poker machine with the same pay table declaring a different payback.
This didn’t make any sense to him. He set out to find out which one was correct. if either. My father was not a trained programmer, to say the least. He was a brilliant electrical engineer who could do complex math in his head, but he was a beginner at programming.
My dad knew that there was an absolute correct payback for video poker. The regulations required that the video version play identically to a real-life version of cards. So, all he had to do was figure out what was the best way to play each of the 2,598,960 possible initial deals. Unlike traditional poker, video poker had no bluffing. The value of a final hand was whatever the pay table said it was. You didn’t have to worry about beating out another hand. For each of these nearly 2.6 million initial deals, there are 32 ways the hand can be played. These range from throwing all five cards to keeping all five cards and everything in between. Depending on how many cards are discarded, there can be as many as 1.5+ million possible draws.
Using today’s computers, you couldn’t run every possible hand. My father was doing this in the late 1980’s using an Amiga computer that probably had 1/1000th of the processing power of my computer.
It’s been a lot of years since I saw my dad’s program. My recollection is that he did a lot of math longhand — or maybe using a spreadsheet program — and then plugged it into his code to speed things along. Thus, if you held a pair, and discarded three cards, rather than run all 16,215 possible draws, his program essentially ‘knew’ that you would want up with a certain number of pairs, trips, full houses and quads. The program had to take into account all the discards as well. For a pair, this is simple, but it gets more complex for other hands.
That was step one of the process. This step determined how many of each paying hand would happen if you held/discarded certain cards. Step two was to multiply the payouts of each winning hand by these occurrences and sum up the values.
Step three was to divide this number by the total number of draws so that each of the 32 ways could be compared on an even basis. This was called the expected value. Step four was to compare the expected value of each of the 32 ways. Whichever way had the highest expected value was the best way to play the hand. The expected value of each of the nearly 2.6 million hands were summed to get the payback of the overall game.
When I got into the business in 2002, I wrote a similar program as a sort of test for myself. My program runs in about an hour or so, maybe less. I can’t even imagine how long my father’s program took to run a single version/pay table of video poker. When he released his book “Winning Strategies for Video Poker,” there were 60 different pay tables in it. It probably took weeks just to let the program run to spit out the payback and strategy.
I don’t know if either of those two original machines had the right payback for the game. I wouldn’t be surprised if neither of them did.
My father long suspected that the original “strategy” that was derived was more of a common-sense strategy, where the developers assumed that players would play hand a particular way. I’m dumbfounded that they didn’t see that there was a mathematically perfect solution that could be calculated relatively easily. This might explain why some of those games also boasted paybacks so close to 100 percent or in some cases over 100 percent. Casinos don’t tend to put games in with over 100 percent paybacks.
Thirty years after he created those programs and 21 years after he is gone and video poker is a staple in the casino. Is it any wonder that many in the industry gave him the nickname the Godfather of video poker.
Lenny Frome didn’t invent video poker. But, can anyone really argue that without him, it never would have succeeded?
At Gaming Today we are dedicated to providing valuable up-to-date information on the casino industry and pari-mutuel race wagering. With news and features, plus expanded coverage in key areas – race and sports analysis, picks, tips, and handicapping.