[PPT] - What Secondary Market Data Do and Do Not Tell You Rob Stavins PowerPoint Presentation

SLIDE 1

BOSTON CHICAGO DALLAS DENVER LOS ANGELES MENLO PARK MONTREAL NEW YORK SAN FRANCISCO WASHINGTON

Winning Strategies for Ticket Pricing: What Secondary Market Data Do – and Do Not – Tell You

Rob Stavins

Albert Pratt Professor of Business and Government John F. Kennedy School of Government, Harvard University

MIT Sloan Sports Analytics Conference

March 3, 2012

SLIDE 2

Agenda

Background, Purpose, and Objectives Analytical Methods Outputs and Illustrative Results

PAGE 1

SLIDE 3

Background and Purpose

Every day, in ordinary markets for all sorts of goods and services, the

interaction of supply and demand yields market-clearing prices,

which just barely sell inventory – no severe shortages or surpluses
In professional sports, the true market-clearing price (market value) of

tickets is in some cases greater than face value, and in some cases less

And the true market value almost always varies more than the face value –

according to location, amenities, timing, and other factors

We know this from the secondary market, but very careful analysis is

required to develop reliable estimates of true market-clearing (or other revenue-enhancing) prices

In fact, secondary market prices on their own are often highly misleading!
The analysis employs advanced, but well-established statistical methods

which we have used successfully for MLB and NFL teams, and can be adapted for other sports

PAGE 2

SLIDE 4

The Miami Dolphins turned to us to help

develop a pricing plan for 2010 season tickets and individual game tickets

Using an econometric analysis of

secondary market transactions, we evaluated consumer demand, estimated market-clearing prices, and recommended a new pricing approach

As a result, the Dolphins increased ticket

prices for 55% of seats, left prices unchanged for 30%, and lowered prices for 15% of seats

The team also created new seating

increase both attendance and revenue

Miami Dolphins raise some ticket prices, freeze and lower others For $5 more, Miami Dolphins fans can have it made in the shade

PAGE 3

Example of Real World Application: Miami Dolphins

Miami Dolphins raise 2010 ticket prices in many lower-deck sections

March 4, 2010

SLIDE 5

Possible Objectives of Ticket Value Analysis

I. Increase attendance and/or revenue

What are the right relative prices among existing ticket categories?
Can revenue be increased by adjusting or creating new ticket categories?
How much could be gained by temporally differentiated pricing (variable

pricing)?

II. Improve the quality of ticket services for fans

Fans dislike sitting next to someone who paid less for the same-quality

seat, or paying the same amount as someone who has a better seat

Better pricing reduces price differentiation by source, and increases price

differentiation by quality

III. Address a wide variety of investment questions

For example, adding new seats, new seating areas, or amenities

IV. Improve understanding of the real market value of assets

The present discounted value of net revenue that could be earned from

ticket sales is a key component of a team’s real market value

PAGE 4

SLIDE 6

Understanding the True Demand for Tickets

For all of these objectives, one must know the true market demand for

tickets

How much people value specific seats to specific games
How much people value individual attributes of seats and games
But true market demand is unknown in advance,
because face values do not reflect true interaction of supply and demand

PAGE 5

SLIDE 7

Section A sells out, but leaves $ on the table due to insufficient differentiation

Capacity

Price

Sec A Price

Revenue Impacts of Inefficient Pricing of Some Tickets

Sec B Price

Section B does not sell out, and leaves $

n the table due to:

Limited differentiation Price too high for some seats within section

Section A Section B

PAGE 6

Demand

SLIDE 8

X X Price

Sec A Price Sec B Price Sec C Price

Revenue Improvement Due to Not Overpricing Tickets

Section B sells out by creating sub-section C at lower price (but still leaves considerable $ on table due to limited price differentiation in Section A)

Section A Section B C

Capacity

PAGE 7

Depending on demand, it may be possible to increase revenue in Section A by raising price and sacrificing attendance, but here we focus on selling out

SLIDE 9

Price

Sec C Price Sec B Price Sec A Prices

Revenue Improvement Due to Increased Price Differentiation in Section A

Section A leaves much less $ on the table, due to price differentiation of new sub-sections

Section B C A1 A2 A3 A4

Capacity

PAGE 8

SLIDE 10

X Price

Sec C Price Sec B Price Sec A Prices

Revenue Potential Due to Higher Demand Due to Temporal Effect (Key Game)

Sections A, B, & C sell out, but A & B leave more $ on table due to higher demand for key game

A1 A2 A3 A4 Section B C

Capacity

PAGE 9

Revenue Potential Due to Higher Demand for Specific Games (e.g., Key Opponent)

SLIDE 11

X Price

Sec C Price Sec B Price Sec A Prices

Revenue Potential Due to Higher Demand Due to Temporal Effect (Key Game)

Variable Pricing with higher prices for “A sections” for key game leave less $ on the table

A1 A2 A3 A4 Section B C

Capacity

PAGE 10

Revenue Improvement with Higher Demand and Temporal Price Differentiation for Key Game (Variable Pricing)

Not to be confused with dynamic pricing

SLIDE 12

Agenda

Background, Purpose, and Objectives Analytical Methods Outputs and Illustrative Results

PAGE 11

SLIDE 13

Hedonic Analysis of Secondary Market Data

Hedonic analysis is a well-accepted statistical method that reveals

the relationship between a product’s attributes and its market price

Has been extensively applied to value many products, including

automobiles, computers, and airline tickets

PAGE 12

SLIDE 14

Hedonic Model Inputs

Secondary Market Prices Seating Attributes Temporal Attributes Estimated Ticket Values (in Secondary Market) Value of Individual Attributes Hedonic Price Model Estimated Market-Clearing Prices

PAGE 13

Analysis Outputs

Price-value ratios by ticket category
Recommended ticket categories
Estimates of incremental revenues

Additional Team & Stadium Information

SLIDE 15

= secondary market ticket sale Temporal Attributes

Opponent
Day of week/time of game
Expected weather (climate)
Playoff contention

Location (Seating) Attributes

Seating section
Distance to field, home plate
Angle of view
Elevation and viewing pitch
Other amenities – club seats,

parking privileges, etc. Secondary Market Sale Price

Basic Data for Analysis

PAGE 14

Ticket Sale Attributes

Number of seats together
Time from sale to game day

SLIDE 16

Collection of Market Price Data

Data on secondary market transactions (sales) can be obtained directly

through relationships with secondary market ticket sellers

Alternatively, ticket price data can be obtained through automatically

harvesting, systematically organizing, and interpreting observations from numerous secondary-market websites that resell tickets

PAGE 15

SLIDE 17

Execution of Hedonic Model

Estimated Ticket Values (in Secondary Market) Value of Individual Attributes Hedonic Price Model Estimated Market-Clearing Prices

PAGE 16

Analysis Outputs

Price-value ratios by ticket category
Recommended ticket categories
Estimates of incremental revenues

Additional Team & Stadium Information

Secondary Market Prices Seating Attributes Temporal Attributes

SLIDE 18

20 15

5

20 $0 $10 $20 $30 $40 $50 $60 Field box Behind home plate Tuesday night Opponent is Red Sox

Execution of Hedonic Model

Econometric (statistical) analysis
Here’s a simple example of the contribution of each attribute…

PAGE 17

Note: Contribution of each attribute affects contributions of others

SLIDE 19

Execution of Hedonic Model

Econometric analysis is based on secondary market activity
While the sample is large (statistically precise), it is not representative (biased)
Tickets on the secondary market are not necessarily representative of all tickets
We use the statistical analysis to adjust for this
Then, we simulate the value of tickets for specific games and/or combinations of

attributes

PAGE 18

Unadjusted Ticket Prices Estimated Ticket Values Ticket Category Face Value Secondary Market Sample Average Regression- Adjusted Value Estimated Market Clearing Price Section A $40 $185 $120

SLIDE 20

Moving from Secondary Market to Market-Clearing Prices

Estimated Ticket Values (in Secondary Market) Value of Individual Attributes Hedonic Price Model Estimated Market-Clearing Prices

PAGE 19

Analysis Outputs

Price-value ratios by ticket category
Recommended ticket categories
Estimates of incremental revenues

Additional Team & Stadium Information

Secondary Market Prices Seating Attributes Temporal Attributes

SLIDE 21

Section B $250 $120 $160 $175

Moving from Secondary Market to Market-Clearing Prices

Buyers on the secondary market may not be representative of the larger fan base
We incorporate additional information from the team, and use a proprietary

methodology to correct for this bias, and thus estimate the true market clearing price in the primary market …

PAGE 20

Unadjusted Ticket Prices Estimated Ticket Values Ticket Category Face Value Secondary Market Sample Average Regression- Adjusted Value Estimated Market Clearing Price Section A $40 $185 $120 $70

SLIDE 22

Agenda

Background, Purpose, and Objectives Analytical Methods Outputs and Illustrative Results

PAGE 21

SLIDE 23

Execution of Hedonic Model and Preliminary Outputs

Estimated Ticket Values (in Secondary Market) Value of Individual Attributes Hedonic Price Model Estimated Market-Clearing Prices

PAGE 22

Analysis Outputs

Price-value ratios by ticket category
Recommended ticket categories
Estimates of incremental revenues

Additional Team & Stadium Information

Secondary Market Prices Seating Attributes Temporal Attributes

SLIDE 24

7,500 1,000 5,500 15,000 20,000 3,000 5,000 1,000 900 800 1,200

Illustrative Analysis Output

PAGE 23

120 105 105 85 55 75 95 280 350 350 500

Lower Central Lower End Section Lower Sideline Lower Corner Upper Corner Upper Sideline Upper Central Club End Section Club Sideline Club Corner Club Central

161 130 120 90 55 65 75 195 209 200 230 41 25 15 5

10
20
85
141
150
270

34% 24% 14% 6% 0%

13%
21%
30%
40%
43%
54%

Face Value ($) Market Clearing Price ($) Difference ($) Difference (%) Capacity

Face value and market-clearing prices are correlated, suggesting the team understands the relative value of seating categories; however, there is still evidence of “mispricing” in several sections Modest percentage increases in large capacity sections can have significant revenue implications

SLIDE 25

Additional Output from Analysis: Refined Pricing Strategies

PAGE 24

Spatial Refinements in Ticket Pricing
Improve boundaries of existing sections
Create new sections for better price differentiation
Consolidate sections
Intertemporal Refinements: Variable Ticket Pricing
Peak-season
Time of day/day of week
Opponent

SLIDE 26

Conclusion

PAGE 25

Secondary market data can be very helpful in improving ticket pricing,

but careful analysis is crucial

Proper analysis of secondary market data can lead to better pricing

strategies, ….. improving both fan experience and team revenues

SLIDE 27

Contact

Andy Parece

Managing Principal Analysis Group, Inc. aparece@analysisgroup.com

Noam Kirson

Manager Analysis Group, Inc. nkirson@analysisgroup.com

Matt Notowidigdo

Neubauer Family Assistant Professor of Economics Booth School of Business, University of Chicago noto@chicagobooth.edu

Rob Stavins

Albert Pratt Professor of Business and Government John F. Kennedy School of Government, Harvard University robert_stavins@harvard.edu

PAGE 26

SLIDE 28

Appendix

PAGE 27

SLIDE 29

A Word on Dynamic Pricing

Dynamic pricing – changing ticket prices during the season in response

to changes in demand

Prices can be adapted in real time if and when demand changes due to

changing factors, such as team record, player status, or weather

But, best dynamic pricing policy will build on smart basic ticket pricing
Dynamic pricing should only be used to address real-time changes in ticket

demand, rather than as an ongoing attempt to correct for poor primary pricing

Dynamic pricing can introduce perverse incentives for fans
Getting basic ticket prices as close to optimal (highly differentiated) as possible

reduces need to rely on dynamic pricing and the uncertainty it introduces

PAGE 28

SLIDE 30

Going beyond Market Clearing Prices: Further Opportunities for Revenue Enhancement

PAGE 29

Market clearing prices are the prices at which sections will sell out
But increased attendance is only one of several possible team objectives
Pricing above market clearing prices may further enhance revenue, even if

attendance is reduced (depends upon shape of demand functions)

So, the hedonic analysis can be used to examine a wide variety of