wecPPT0525f

Uploaded from authorPOINTLite
Views:
 
Category: Education
     
 

Presentation Description

No description available.

Comments

Presentation Transcript

Casinos’ “Gravity” Not Always According to Reilly: 

Casinos’ “Gravity” Not Always According to Reilly Presentation to the 13th International Conference on Gambling & Risk-Taking Will E. Cummings Cummings Associates May 25, 2006 [ With notes added May 30, 2006 ]

Casinos’ “Gravity” According to Reilly -- Amended: 

Casinos’ “Gravity” According to Reilly -- Amended Presentation to the 13th International Conference on Gambling & Risk-Taking Will E. Cummings Cummings Associates May 25, 2006 [ With notes added May 30, 2006 ]

Overview: 

Overview Gravity I: Reilly’s Law Gravity II: General decline w. distance The Problem: New casino(s) in Iowa Alternative Solutions An Empirical Test Conclusions / Future Directions

Measuring Markets: 

Measuring Markets Gaming Revenues Gaming Revenues / Market Size Per unit of population Per unit of economic activity Drawing boundaries: “Half way” to the next facility? Wide-open spaces? Draw rings . . . Segmented rings . . . “Gravity” models

Gravity I: Reilly’s Law: 

Gravity I: Reilly’s Law

Reilly’s Law:: 

Reilly’s Law: ms ~ S/d2 Where ms : market share S : “size” of each trade center d : distance

Newton’s Law:: 

Newton’s Law: F = m/d2 Where F : gravitational force m : mass (of each body) d : distance

Reilly’s Law:: 

Reilly’s Law: ms ~ S/d2 Where ms : market share S : casino size (capacity) d : distance

Reilly’s Law: weighting factor: 

Reilly’s Law: weighting factor weightij = sizejscoeff x (distij/7)grav1 weightij : weight of casino j in market segment i sizej : size of casino j scoeff : size coefficient = 1.0 distij : distance of casino j from market segment i grav1 : “competitive” gravity coeff. = -2.0 [ corresponds to Huff’s generalized “friction” (1964) ]

Gravity II: General decline w. distance: 

Gravity II: General decline w. distance

Las Vegas Visitation/Distance: 

Las Vegas Visitation/Distance

Las Vegas: log-log version: 

Las Vegas: log-log version

Las Vegas: slope of the curve: 

Las Vegas: slope of the curve

Laughlin: much steeper slope: 

Laughlin: much steeper slope

Mississippi Visitors/Day/000: 

Mississippi Visitors/Day/000

Mississippi: log-log version: 

Mississippi: log-log version

Mississippi: closer/cleaner: 

Mississippi: closer/cleaner

Mississippi: linear fit: 

Mississippi: linear fit

Mississippi: polynomial fit: 

Mississippi: polynomial fit

Mississippi: log-log version: 

Mississippi: log-log version

Smaller Snapshots: Casinos X and Y: 

Smaller Snapshots: Casinos X and Y

Casino X / players’ club data: 

Casino X / players’ club data

Casino X: log-log version: 

Casino X: log-log version

Casino X: rescaled $/distance: 

Casino X: rescaled $/distance

Casino X: rescaled log-log: 

Casino X: rescaled log-log

Casino Y: players’ club data: 

Casino Y: players’ club data

Casino Y: log-log version: 

Casino Y: log-log version

Casino Y: closer/cleaner: 

Casino Y: closer/cleaner

The Problem: New casino(s) in Iowa: 

The Problem: New casino(s) in Iowa

Market X in Iowa -- Today: 

Market X in Iowa -- Today 50,000 adults 100mi(+) from existing facilities (with 10,000 slots) “Aggregate” distance factor = .155 Current spending ~ $106/adult 100% to existing facilities Total spending at existing facilities ~$5,300,000

Market X -- With New Casino: 

Market X -- With New Casino Small: 300 slots, 12 table games New average distance ~ 10 miles “Aggregate” distance factor = .78 Total spending now ~$26,500,000 Old casinos’ market share = ((10,000x(1/10)^2) / ( “ + 300) = 100/(100+300) = 25% Total spending at existing facilities ~$6,625,000

Alternative Solutions: 

Alternative Solutions

Alternative Solutions: 

Alternative Solutions Absolute distance matters less? Size matters less? Relative distance, at extreme values, matters more Segment the market: at different distance scales, consumer needs / products that serve them differ

Absolute Distance Matters Less?: 

Absolute Distance Matters Less? If spending didn’t rise so much . . . Coefficient ~ -0.6 ? (vs. -0.7) “Old” casinos’ market share → 11% Not quite enough impact Belied by Δ spending when new casinos enter fringes of markets

Size Matters Less?: 

Size Matters Less? Why should size matter so much? Coefficient ~ 0.9 ? (vs. 1.0) “Old” casinos’ market share → 19% Not nearly enough impact Belied by fit of models elsewhere / in other markets

Relative Distance Matters More: 

Relative Distance Matters More “Competitive” gravity stronger? Different regimes at different distances? Coefficient ~ -2.2 ? (vs. -2.0) “Old” casinos’ market share → 17% Still too high, but -2.5 or -2.6 would do it Worth further investigation, but . . .

Distinct Market Segments: 

Distinct Market Segments Nearby: “Convenience” Gaming Distance matters a lot Middle distances: “Excursion” Gaming Distance matters some Large distances: “Destination” Gaming Distance matters less

Spending vs. Distance: 

Spending vs. Distance

Segmentation by Distance: 

Segmentation by Distance

Reilly’s Law: weighting factor: 

Reilly’s Law: weighting factor weightij = sizejscoeff x (distij/7)grav1 weightij : weight of casino j in market segment i sizej : size of casino j scoeff : size coefficient = 1.0 distij : distance of casino j from market segment i grav1 : “competitive” gravity coeff. = -2.0

Reilly’s Law Amended: 

Reilly’s Law Amended weightij = sizejscoeff x (distij/7)grav1 x (distij/7)grav2 weightij : weight of casino j in market segment i sizej : size of casino j scoeff : size coefficient = 1.0 (or 0.9) distij : distance of casino j from market segment i grav1 : “competitive” gravity coeff. = -2.0 grav2 : “aggregate” gravity coeff. = -0.7

Spending vs. Distance: 

Spending vs. Distance

Reilly’s Law Amended: 

Reilly’s Law Amended weightij = sizejscoeff x (distij/7)grav1 x (distij/7)grav2 weightij : weight of casino j in market segment i sizej : size of casino j scoeff : size coefficient = 1.0 (or 0.9) distij : distance of casino j from market segment i grav1 : “competitive” gravity coeff. = -2.0 grav2 : “aggregate” gravity coeff. = -0.7

Reilly’s Law Amended II: 

Reilly’s Law Amended II weightij = sizejscoeff x (distij/7)(grav1+grav2) weightij : weight of casino j in market segment i sizej : size of casino j scoeff : size coefficient = 1.0 (or 0.9) distij : distance of casino j from market segment i grav1 : “competitive” gravity coeff. = -2.0 grav2 : “aggregate” gravity coeff. = -0.7 [ Corresponds to Huff’s “friction” = -2.7 ]

An Empirical Test: 

An Empirical Test

New Casinos in Iowa: 

New Casinos in Iowa

New Casinos in Iowa: 

New Casinos in Iowa IOC Waterloo / Black Hawk County Riverside / Washington County Diamond Jo Worth / Worth County Wild Rose Emmetsburg / Palo Alto County

New Casinos in Iowa: 

New Casinos in Iowa

Projections Compared: 

Projections Compared Reilly Raw Reilly Amended Black Hawk $79.6m $83.7m Washington $63.3m $68.3m Worth $17.2m $23.9m Palo Alto $11.0m $14.8m ----------------- ------------------ Total $171.1m $190.8m Impacts ($34.8m) ($54.3m) ----------------- ------------------ Net New Win $136.2m $136.4m

Projections Compared II: 

Projections Compared II Reilly Raw Reilly Amended Marquette - 4.9% - 8.2% Dubuque - 3.6% - 4.8% Clinton - 3.7% - 3.3% Quad Cities - 3.4% - 4.9% Catfish Bend - 8.0% - 8.7% Tama -16.7% -32.8% Prairie Meadows - 2.7% - 2.0% Lakeside - 4.1% - 5.7% C. Bluffs/Omaha - 0.1% - 0.1% Onawa - 0.1% - 1.2% WinnaVegas - 0.1% - 1.0% Sioux City - 0.2% - 0.3%

Conclusions / future directions: 

Conclusions / future directions Tweaks to “gravity” parameters Size matters less? Absolute distance matters . . . ? Relative distance matters more Segmentation by distance intuitive, logical, and tractable More empirical data! More local studies / diverse markets

Casinos’ “Gravity” According to Reilly -- Amended: 

Casinos’ “Gravity” According to Reilly -- Amended Presentation to the 13th International Conference on Gambling & Risk-Taking Will E. Cummings Cummings Associates cummingsw@aol.com