It all begins Macau and Las Vegas was built on return to player or RTP, the wily and often fickle hidden hand shaping the odds in the house’s favor.  There has long been a debate between casino executives about the role of RTP and whether initiatives to lower RTP (i.e. increase win sometimes termed hold) actually drive more revenues or simple diminish play-time.  We have explored this subject at length; most recently we spoke about the ‘elasticity of RTP’ at the UNLV conference on risk.  This work found that changes in RTP on the margin are impactful – at least up to an optimal inflection point - but increases past this inflection point actually reduce value.  This finding coincides with our experience from the field.  For instance, one Differential advisors oversaw the shift from a lower-edge baccarat variant to higher edge variant.  This subtle change resulted in a multi-million dollar EBITDA improvement on an annual basis.an idea. Maybe you want to launch a business. Maybe you want to turn a hobby into something more. Or maybe you have a creative project to share with the world. Whatever it is, the way you tell your story online can make all the difference.

We leveraged a curated data-set from our strategic partners at GMA consulting which details monthly slot turnover and slot actual win data for ~200 different properties which we have termed the slot win compendium. In total, the slot win compendium includes ~6800 month x property data points dating from the early 2000’s stretching to COVID.  Among the studied properties the avg win/hold rate is 7.8% (90% upper range is 9.8% and 10% lower range is 5.9%).  Perhaps more interestingly, hold / win rate was sub 7% in 2005, grew consistently to 2016 when hold / win rate was 7.7%, and then accelerated at a dizzying pace to 2020 when hold / win rate topped 8.3%.  See chart below

We were recently asked about the typical flow through from increasing slot hold win-rate.  More specifically, let’s take a specific example: say a property currently has a floor-wide hold / win-rate of 7% and earns $1m per day.  Mgmt, worried about achieving the budget, begins an initiative to increase hold / win-rate by seven-tenths of a percentage point which is a +10% improvement (.7% divided by 7%).  After these changes, and over the long-run, will the casino earn an extra $100K on top of the current $1m: i.e. a ten percent improvement? To answer this question, we first turned to our network of slot managers across the country.  These managers range from senior executives at major resorts to hands-on operators at smaller operations and all of whom are opinionated and well versed.  When posed with the scenario above none of the slot managers thought there would a linear impact from hold on revenues (linear meaning a 10% increase in hold delivers a ten percent increase in win).  Even so, the managers equivocated on the actual ratio that flowed to the top-line.

Item descriptionTo answer this question, we designed a natural experiment using slot hold compendium data-set, described above: the logic goes that variation in RTP, whether by a machine setting or by volatility, has the same impact on the player.  To study the relationship on RTP and actual win

Data in hand, we developed an econometric model designed to tease out what economist would term RTP elasticity (i.e. the change in actual win from a corresponding change in RTP).  This macro-analysis differs fundamentally from previous micro-work which analyzed behavior at a player level.  A macro view captures concepts like dropped trips more succinctly than the micro-model.

Item descriptionWhile our hold/win rate model is still in it is infancy we can see several clear trends.  Over the 15 year period of the model and across 200+ properties, the elasticity is .3.  Let me explain what the means:

• Most importantly, the elasticity of .3 is positive meaning that increases in hold correspond to increases in actual slot win

• We can then convert this elasticity into a ‘flow-through’ rate which can be applied.  Continuing with the example of a property which wants to increase the hold/win rate by 10%.  We can calculate the flow-through as 1.1^.3 = 3% (i.e. 1.1 captures that we want to increase hold/win rate by 10%).

The conclusion from the analysis above, a ten percent increase in hold / win rate will increase win by 3% or a 30% flow-through rate.  While we love the simplicity and concision of an over-arching macro model, surely the .3 elasticity while vary given the property’s starting hold / win percent.  In economics this concept is know as diminishing marginal returns.  While on average, a +10% increase in hold / win rate delivers a 3% increase in GGR, our model indicates that impact range from as low as 2% and as high as 4%.  We found that properties that increase hold / win rate from a higher starting point (i.e above 7%) have an increase in GGR from an increase in hold / win rate at the lower end of this range.  Likewise large increases in hold / win rate deliver disappointing increases in GGR.  While we can see a diminishing marginal return pattern in the data, our model does not yet capture this dynamic.  We are updating and please stay tuned.

This is not to say that increases in hold / win rate initiatives are not worthwhile.  Although, we advocate for a more considered approach than a broad increase which is rooted in database analysis.  For instance, we recently supported a regional operator where a 6% increase in hold/win rate delivered a ~3.5% increase in slot win which is a 58% flow-through.  To achieve this lift, we first segmented the database by preferences like risk vs entertainment and wallet vs time constrained.  Each resulting segment has different trip characteristics and hold / win rate elasticity.  We then helped district the floor into regions targeting these patrons, aligned the hold/win rates and supported with marketing.

Item descriptionThis is not to say that increases in hold / win rate initiatives are not worthwhile.  We advocate a more considered approach which is rooted in database analysis rather than a broad increase to optimize flow-through.  We recently supported a regional operator where a 6% increase in hold/win rate delivered a ~3.5% increase in slot win which is a 58% flow-through.  To achieve this lift, we first segmented the database by preferences like risk vs entertainment and wallet vs time constrained.  Each resulting segment has different trip characteristics and hold / win rate elasticity.  We then helped district the floor into regions targeting these patrons, aligned the hold/win rates and supported with marketing.