Our hold calculation has historically differed meaningfully from management’s. Here is why.
Let’s face it, gambling is a volatile business. While the odds normalize over the long-term, even a big casino company can be significantly impacted by luck over a quarter. Gaming management teams have done a terrific job focusing on the hold impact on quarterly conference calls when hold is low but less so when hold is high. Go figure. And statistically speaking, as LVS and WYNN grow, hold volatility will be less important. Not in terms of absolute dollars as an analyst on the call seemed to believe, but in terms of percent impact.
So we certainly appreciate the company’s additional disclosure. The issue we have is not with the transparency, although management failed to comment on the positive EBITDA impact in the Q1 earnings release/conference call. Rather, it’s the methodology. As can be seen from the following chart, we calculate significantly smaller quarterly impacts from low hold than the company, especially in the most recent quarter.
Issues with LVS’s methodology:
- The company only calculates the hold impact on VIP business when the hold percentage at a property is outside the range of 2.70-3.00% and then calculates the differential to the midpoint (2.85%). So EBITDA from a property holding at 2.99% will not be adjusted at all but one that holds at 2.69% will be. A swing from 2.70% to 3.00% is material. For instance, that represents about $25-35 million in EBITDA for MBS alone. Our calculation just adjusts each property to its historical average.
- LVS does not adjust for abnormal Mass hold; but we do, using the 6 quarter trailing average.
- Since it’s impossible to calculate, neither LVS nor Hedgeye factors in that patrons play more when they win and less when they lose. Thus, volumes are not static across different hold percentages. This has the impact of dulling volatility so generally both methodologies will overstate the impact of hold variations.
- LVS doesn’t use the “range” methodology for LV operations like it does for Macau/Singapore. This doesn’t seem consistent.
- Mass hold in Macau/Singapore and table hold in Las Vegas is measured by the amount of chips bought. Thus, if the velocity of play changes, so will hold percentage, theoretically. For example, in down economic times, players may exchange dollars for the same amount of chips they always do but they may not gamble as much. The denominator (volumes) will be the same but the numerator (gaming win for the casino) will likely be less, resulting in a lower hold percentage than normal.
Q2 was certainly impacted by low hold percentage and that needs to be discussed. However, we calculate the impact was roughly half of what LVS estimated and almost all of the impact was concentrated at MBS as can be seen below.