While watching a recent television story about the permanent closing of retail giant Lord & Taylor due to COVID-19, one commentator pointed out that the closure was not really caused by the pandemic. Rather, he said, the pandemic simply accelerated the trend of big box stores being in trouble. Likewise, Foreign Affairs magazine recently featured an article headlined “The Pandemic Will Accelerate History Rather Than Reshape It.” The accompanying article pointed out that many global cultural shifts occurring now were already underway, but, again, COVID prompted their immediate arrival.
Could the same shifts be happening in gaming? It may be months, even a year or more, before we know those answers. But early reports seem to prove the hypothesis.
Ive theorized for months that the tipping point for ETGs (Electronic Table Games) was coming soon, and a recent CDC Gaming article by Howard Stutz seemed to credit COVID with accelerating that trend. “Ironically, the pandemic is the primary reason the casino industry has focused its attention this year on electronic table games,” Stutz writes. “The games offer social distancing and are a product that lends itself to the health, safety, and cleaning protocols that have been implemented in response to the coronavirus.” Scientific Games VP Rob Bone added, “I’m confident this segment is going to grow by 10% to 25% in the next 12 to 24 months. The demand is out there.”
Again, time will tell on ETGs. But even more dramatically, some data is starting to arrive that confirms and validates some other more dramatic trends that had been simmering for months. And, again, COVID seems to have accelerated the curve.
The question led me to a recent discussion with Nick Hogan, CEO of the data science firm ReelMetrics. The company has been collecting slot and player data here in North America for six years and has comprehensive meter sets on literally billions of sessions of slot play from participating operators, both before and during the pandemic.
Hogan confirmed my suspicions, explaining that the pandemic’s operational disruptions have created a cornucopia of new data points that have kept its analytical team in overdrive since March. “It’s as though the tide’s rolled out, exposing facts and trends that have been submerged for decades,” explained Hogan.
He went on to say that this period of COVID restrictions has proven particularly valuable in cementing two conclusions:
- The overwhelming majority of slot floors are grossly over-diversified
- They are also horribly unbalanced
There’s a lot of science behind Hogan’s two simple statements, but it comes from the ReelMetrics team’s work on the fundamental concepts of Slippery slots versus Sticky slots and their interplay with time- and budget-constrained players. As you might imagine, a Slippery machine is an inherently unproductive asset that loses players almost as fast as it can attract them. The latter is the exact opposite, in that players tend to stay and play a Sticky game much longer.
So along comes the pandemic, and suddenly many casinos had to make drastic cutbacks of slots by 50% or more. While no one wanted or needed this horrific virus, it did present some opportunities to test concepts and try to extract some advantage from this adversity.
“The pre-pandemic data clearly supported the over-diversification and imbalance hypotheses,” explains Hogan. “However, although we had great resolution on both symptom and cause, we’ve struggled to find statistically meaningful paths to evaluating treatments. Done properly, the controls would require the temporary disablement of thousands of Slippery slots. Few would ever have done that on their own. But with the pandemic, it became an instantaneous reality.”
Sifting through the resultant data sets is leading to some startling discoveries.
For example, ReelMetrics has seen many examples where, despite dramatic unit cuts, lower visitation and changing demography, some operators are experiencing gains in unit level win that have allowed them to meet or exceed 2019 results.
“We’ll see a 1,500-unit floor with a $200 average daily win per unit that was forced to eliminate half their machines. Fewer people are coming through the doors, scores of the most valuable customers are absent, but, despite this, unit-level performance has tripled to around $600 per day,” explains Hogan. So, that’s a net revenue gain of 50% and a WPUPD increase of 200%.
Of course, basic math alone would say that if you reduce the denominator in half, you should have seen some gain from that alone if supply drops and demand remains the same. Not surprisingly, the disproportionate nature of the gains piqued the interest of the ReelMetrics analytical team.
“The temptation is to credit lower unit counts and pent-up demand and call it a day,” Hogan says. “These are contributing factors to be sure, but there’s a lot more to this story and, generally speaking, it’s good news for operators.”
In explaining their findings, Hogan talked about both the company’s Slippery / Sticky analytical model and a resulting tendency they call ‘Ricochet.’ This is the common practice of players jumping from machine to machine until they find one they like.
“What we see is players pinballing between Slippery units until they get snagged by something Sticky,” says Hogan. “While ricocheting, a player’s wallet remains closed, cutting spend, while his or her frustration mounts, which harms value perceptions and loyalty.”
“Ricochet is an enormously destructive force, and we’ve yet to see a floor that’s not heavily affected by it,” he explains.
“Pre-pandemic, we postulated that simply reducing the volume of Slippery product would drop ricochet levels, yielding disproportionate gains in revenue and satisfaction.”
With the inventory changes forced by COVID-19 restrictions, ReelMetrics could test their theory in live gaming environments. “It’s effectively natural selection,” says Hogan. “Whereas, in the past, Slippery product vastly outnumbered their Sticky equivalents, the unit cuts have recalibrated those ratios profoundly. In turn, ricochet levels have plummeted, and the disproportionate productivity and performance gains abound.”
One can’t help but ask if this isn’t obvious information. “Add more good games and get rid of bad games” has been the first lesson in Slots 101 for years. Doesn’t everybody know that?
There’s a surprise in the data, though. “The empirical evidence reveals a gigantic gap between principle and practice,” explains Hogan. The highest percentage of productive to non-productive games that Hogan and ReelMetrics have seen from new clients is in the 25% range, with the mean at roughly 20% and the lowest recorded south of 10%. In other words, our Slots 101 lessons were not being applied.
These numbers validate Hogan’s #2 theory that most floors are horribly unbalanced. They just have too many weak games and not enough good ones (and, yes, there really are plenty of good ones available.)
One encouraging note from Hogan was that these floor imbalances are not due to malfeasance or poor management by your slot teams. Rather, this is just the way the industry has been structured and functioned for decades. New games arrive on the floor, maybe perform well, but often slowly diminish. Given the multitude of new products introduced every year by the manufacturers, the volume of unproductive games can get, and has gotten, out of control.
“Trial & error has always been inherently inefficient. Had viable alternatives existed, of course we would have used them,” Hogan says. “Now they do. And thanks to the data gathered coming out of the shutdowns, we have a better understanding of the problem in terms of its scale and severity, and we’re learning what to do about it.”
Here’s some more good news. Hogan and his team of data scientists have joined the emerging high-tech sector phenomenon of wanting to use their technology to benefit others during these extraordinary times and sharing what they’ve learned with the public at no charge.
ReelMetrics now has a version of their subscription service called Basic that is free to the industry. Most importantly, it includes a module called Recovery which can provide important insights and help with the issues discussed in this article.
“We’ve taken cues from our data science counterparts in life sciences and biotech,” explains Hogan. “Your ability to survive, let alone master, a new reality is a direct function of the speed and accuracy with which you come to understand it. And nothing clarifies and accelerates that understanding more reliably than deep, meticulously standardized pools of communal data.”
As you might imagine, the paid version of this software is far more productive and will generate even greater insights to your floor performance. But the features in the free version are still impressive, with floorplan analysis, game optimizations and pricing analytics.
With ReelMetrics, and others, like Eilers & Krejcik, now gathering and sharing productivity data from hundreds and thousands of games throughout North America, many of the common mistakes we’ve made in the past can be avoided completely, or rectified quickly, with the improved analytics now available.
That, and the fact that I haven’t had a single cold since I’ve been wearing a mask, are the only two good things I’ve noticed during these dreadful times.
# # #
