Stuart S. Janney III: Dan Singer and Mike Salvaris have been observing and studying the Thoroughbred breeding and racing industry for the past seven years. They're back here at the Round Table Conference today with some very important observations and recommendations. So let me ask them to come to the stage. Thank you.

DAN SINGER: Thanks very much, Stuart, and good morning, everyone. I'm Dan Singer from McKinsey. This is my colleague Mike Salvaris. We're thrilled to be with you this morning.

As Ian and Jim mentioned earlier, in our work this year we identified four major disruptions in sports and gaming that created opportunities for racing but also raised the bar in competing for new fans and share of wallet and wagering.

We've organized our recommendations for racing according to these four growth themes. Let's start with the first theme, which is innovation in venues and the game-day experience. Notwithstanding the increase in TV coverage and the importance of digital media applications and social networks and developing fans, the track experience is still a critical step in developing a new racing fan.

64% of new fans say the single most important reason for becoming a fan was when a friend or relative took them to the track. On the flip side, 45% of lapsed racing fans said they stopped attending because” I moved and there are no racetracks nearby.” Another 27% said “they closed the racetrack where I used to go.”

Other major league sports have dramatically raised the bar on the game-day experience with amenities such as nightclubs, bars that allow social viewing of the game, and non-stop interstitial entertainment, not to mention basic elements such as interesting food selections from local chefs, microbreweries, clean and numerous bathrooms, and well-merchandised retail outlets.

Since 2011, 11 tracks have ceased Thoroughbred racing and many others have had limited capital for renovation. In our research, only 67% of all racing fans and only 61% of fans under 35 said they had an excellent or very good experience at their last track visit. That's versus 83% of racing fans we surveyed in 2011.

As a result, of the top 35 metro areas by population in the U.S., only five have a Thoroughbred track that we would describe as major league based on achieving a TripAdvisor rating of 4.5 stars.

Of course, there are other ways to evaluate tracks, but advantage of TripAdvisor ratings is they are done by the general public, often with hundreds of ratings per venue, and they're available across all U.S. tracks. They also enable comparisons with other sports venues. For example, in New York City alone, Madison Square Garden, Citi Field, Yankee Stadium, Barclays Center, MetLife Stadium are all rated 4.5 on TripAdvisor.

In racing, 14 cities have a track with a 4 rating, and four others are 3.5 or worse. And for those of you familiar with TripAdvisor, 3.5 is not a good rating. You're probably not apt to go visit any kind of venue or restaurant or hotel with a 3.5 rating. So 4.5 is gold standard, and anything less is okay.

There are 12 cities with no Thoroughbred racetrack at all. By comparison, out of 30 Major League Baseball teams, 26 have a stadium with a 4.5 or 5.0 TripAdvisor rating. There are two stadiums with a perfect 5 rating.

We believe it's imperative for the future of racing that the industry innovate and upgrade the track experience in the major population centers. As you heard from Simon earlier, the U.K. Jockey Club has made great strides in the four areas that are the biggest driver of track experience, and these are probably pretty intuitive to you: first, food and beverage; second, cleanliness; third, a unique atmosphere, like Simon described at the major events and festivals; and, fourth, a non-racing event, such as concerts and festivals, and he also mentioned the live concerts that the U.K. Jockey Club organizes.

MIKE SALVARIS: Good morning. Probably the biggest disruption since our 2011 report has been the rapid rise of mobile-based media. Consider since 2011 smart phone shipments have risen from about 500 million per year to just under 1.4 billion today. In 2011, the average American adult spent 24 minutes per day on mobile digital media. Today that figure has gone up over eight times. We now spend three hours and 20 minutes a day on mobile digital media, much to the annoyance of our dining companions.

In 2011, Facebook had about 580 million users; today it's over 2 billion. Instagram had 15 million users; today it has over 1 billion. Netflix now has more U.S. subscribers than all cable TV providers combined.

So that three-hour-per-day increase in the use of mobile media has taken away from TV, at least for millennials. As of 2017, young millennials, those aged 18 to 24, watched under two hours of live TV per day, which is down 44% since 2012. Older Millennials, those aged 25 to 34, viewing is down 32% over the last five years.

So what does all that mean for sports? Well, mobile media has dramatically changed the ways that fans engage with sports and the way that leagues and teams are developing new fans.

Consider the NBA. The NBA leads all sports in terms of developing social media fans. They've built a massive online audience. They have 36 million followers on Facebook, double that of the NFL; 27 million followers on Twitter; 30 million followers on Instagram.

And those are just the NBA league accounts. Players and teams have much more. For example, LeBron James, Steph Curry, and Dwyane Wade collectively have 52 million followers just on Instagram.

How did the NBA create such a massive digital audience? Well, they produce a large amount of content themselves. The NBA's in-house digital team is about 30 people in the league alone, not counting the teams, and they're having huge success with game highlights and, importantly, lifestyle content.

Quick celebrity interviews are popular, such as the one that Dave Chappelle did during the Warriors-Cavs Finals, as are other humorous content, such as one of the most popular videos of last year, which was a video of LeBron James and JR Smith dancing in the locker room before the game had even started. Presumably, that was before JR Smith cost him Game 2.

But the NBA realizes that 99% of fan engagement happens outside of the NBA's owned and operated platforms. People aren't going to NBA.com for their news. They're finding it on social platforms. So the NBA produces content specifically for those social platforms and optimizes them for it.

The NBA has created an open ecosystem. Unlike other leagues, they don't just allow fans to post highlights,. They actually encourage it.

The NBA believes having fans post game highlights on social media increases TV ratings rather than detract from it. As a result, the NBA fans watched over 10 billion NBA videos on social media and over 3 billion minutes of NBA footage during the last season. Game highlights are automatically identified by YouTube, and the NBA shares in the revenue no matter who posts the videos.

The NBA has cultivated a network of influencers, or people with large social followings, on Instagram, Snap, Twitter, and other platforms. The NBA gives these influencers behind-the-scenes access to games and events and, as a result, is enjoying over 500 million YouTube views per month from influencer content.

What do you think these influencers are posting? Is it shots of themselves in the crowd or massive dunks? Actually, the most popular live content for NBA influencers is lifestyle content. So that could be fashion, such as high-end sneakers, food and restaurants or music and other aspects of the NBA lifestyle.

The NBA is capturing emails and other identifiers for millions of fans through social media, even though social platforms do not give the NBA or any other publisher personally identifiable information. One of the most successful ways they do this is they ask fans to vote for All-Stars.

With one click of a banner during or after a video, the fan is taken to NBA.com to vote. The fan is asked for their name, their email, and their favorite team for voting. The league uses those emails to sell the fan merchandise from NBAShop.com, streaming packages such as League Pass, and passes those details on to their fan’s favorite team, so the team can market to that fan.

So what does all that mean for racing? Well, we think there are three implications here. First, racing needs to create more engaging mobile content, especially videos and photos. Now, many organizations in racing today currently produce and distribute videos on social media. For example, a leading racing association has posted over 275 videos to Facebook. But when we look across the landscape, we see few organizations who are currently at the scale required to be successful. Racing needs to increase its output dramatically and increase the engagement of the content that it's producing.

Leading sports sites produce over 200 pieces of content per day, including video clips, articles, podcasts, and tweets. Racing videos also currently get low engagement, typically under 1,000 likes, shares, or comments per post, which is less than half of that of even niche sports such as Supercross, poker stars, and bull riding, and a fraction of what the major leagues accomplish.

As with the NBA, lifestyle content is a fertile area for racing, from foals at the farm to fashion and food at the races. In contrast to mobile and social content, we believe racing is doing a very good job of distributing live streams. Casual fans already access all major races, such as the Triple Crown and Breeders' Cup, on live streams through broadcast networks.

Core fans can access virtually every Thoroughbred race through TVG and the ADWs. The biggest gap in developing new racing fans is in engaging occasional fans through quick, compelling digital content and inspiring them to visit a high-quality track on a good race day.

Racing needs to do a better job of getting race highlights and lifestyle photos and videos in front of casual fans so they'll develop a reason to seek out live streams of races, where they can already access them, if interested.

Influencers can play a much bigger role in racing content. As we mentioned, the NBA is generating over 500 million YouTube views per month through influencer content. In general, brands are spending about two and a half times as much on influencer marketing this year as they did just two years ago. This year some 24 million Instagram influencer posts will be paid for by brands.

Racing has made some attempts at cultivating influencers, but it doesn't have the reach of other sports. The top five influencer accounts in racing have a collective following of 424,000 users, compared with 590,000 for professional bull riding, 88.4 million for NASCAR, and 89.2 million for poker.

The second implication of this disruption is around promotion of content. Racing should promote its best content so that new fans can find it, and that means paying for it to be promoted.

Simply publishing content on social media isn't enough. We need to promote our content so casual fans can see it. Publishers of all types discover the vast majority of their consumption happens outside of their sites and apps, and therefore they need to invest to target fans and promote their content to them on social platforms.

Finally, racing needs to capture fan data from social media engagement and invest in targeted, personalized digital marketing to get new fans to the track. As marketing shifts rapidly from traditional media to digital, it's critical to be able to identify and communicate to fans with highly personalized offers, which requires fan databases and email addresses, cookies, and other identifiers.

Today, major sports leagues know about 40 to 50% of the fans who attend their events. Casinos and theme parks know 70 to 90% of their patrons. By contrast, most tracks can only identify 10 to 15% of the fans at the track. And unlike major leagues, there is no central database of fans and no sharing of fan data across the sport.

To recap, the rapid growth of mobile media has three implications for racing. Racing needs more targeted mobile content creation, racing needs to invest in promoting mobile content to new fans, and racing needs to collect fan data for personalized digital marketing.

While individual tracks and ADWs can and do execute these strategies on their own, there is enormous power in the industry working together. Digital media and digital marketing provide creative and technical skills that are scarce. Though some social videos may look easy to make, creating truly great content and getting that content to be widely seen requires a highly skilled team. There is a lot of science that goes into a cat video.

The industry has a much better chance of winning the war for digital talent with an at-scale digital capability. Data required to effectively market to fans is expensive to gather. As we saw with the NBA, when the league obtains an email and opt-in permission from a fan, it shares the information with the fan's favorite team so that both the league and the team can market directly. Racing could do the same.

The industry would benefit from shared knowledge around what digital marketing tactics would get fans to the track and how to get them to return. Leading digital marketers find the most effective promotions by testing hundreds of campaigns per month, including variations in content, pricing, offering and timing. While difficult and expensive to do by themselves, racing organizations could come together to collectively do something similar.

Finally, it is easier to engage fans with digital content with a common brand that they recognize and trust, who curates the content and alerts them to it.

[Click here for the interview with Chris Pollak.]

MIKE SALVARIS: So before we leave the topic of digital content, data, and marketing, we need to consider the role of television in fan development for racing.

Despite the growth and time spent on mobile media, TV remains a very important part of fan development and ought to continue to be supported by the industry. Although millennials and Gen X are watching less TV today than they did seven years ago, Americans over 60 actually watch more.

Also, even though ratings for sports on TV are declining five to six percent per year, the reach of sports has not changed. Just as many fans watch sports on TV; they just watch fewer events and fewer hours per event.

The biggest sports events are bucking the trends. For example, in racing, ratings for the Triple Crown and the Breeders' Cup have held steady since 2011. And TV remains a critical platform to reach and engage casual fans.

There is also evidence that televising a race increases the handle. Data from one track shows a 14% uplift in handle when a race is broadcast on a general interest sports network in addition to being televised on TVG.

Despite the four times growth in hours of racing on broadcast TV, 34% of racing fans we surveyed still believe there is not enough racing on television.

The legalization of sports betting has created a once-in-a-generation opportunity for racing. At least three leading sports networks are in the process of launching sports gambling-oriented shows and have indicated that they'd like to include racing as often as there are compelling stories to tell.

The video content has to be in HD in order for it to be aired on the shows, but only 24 out of 74 tracks broadcasting on TVG currently provide HD signals. Therefore, racing will need to invest in HD carriage to take full advantage of the TV opportunity from legal sports betting.

Then we're going to talk about our third initiative, advanced analytics. The rise of cloud computing and exponential growth in computer speeds, plus development of efficient tools in machine learning, have enabled businesses of all types to drive insights from consumer data and apply those insights to growing businesses.

This year we've identified four applications of big data and advanced analytics for racing. The first is in scheduling. Last year at the Round Table we shared a model we built based on 40,000 North American races. Our model identified races with purses greater than $50,000 that conflict with each other. We showed that many great races failed to dominate their slot, i.e., one or more races started within five minutes of another major race, resulting in low shares of concurrent purse. We estimated that defining the schedule of these races was a $400-million-per-year opportunity.

In April 2018 we held a pilot with eight tracks around the country. Our goal was to communicate the off-times from these participating tracks, both in advance of each race and in real time, to avoid starting major races within five minutes of each other.

From the four weeks of the pilot, we learned that using simple digital tools, such as group text and spreadsheets, the pilot improved communication among racetracks significantly. Our model predicted the handle for each race with and without conflicts, which helped identify the highest priority races to address.

The tactical, operational issues in coordinating schedules was fairly easy to resolve. That said, the pilot only reduced race conflicts from 53% to 50% for April races.

Optimizing schedules to remove all conflicts in high-handle races would require specific agreements with tracks regarding which races have precedence. That was beyond the scope of this pilot, but we believe there is no fundamental barrier in the industry eventually reaching alignment on these rules of the road.

DAN SINGER: Next we're going to talk about takeout. When we asked avid fans for their top three concerns with horse racing, 35% of them said high takeout rates, which is the fourth highest concern -- actually, it may be a little counterintuitive, but there are three higher concerns -- new fan development, medication, and race quality -- with takeout being fourth.

While takeout rates in absolute are high relative to other types of sports betting -- for example, 16 to 20% takeout rates in racing versus roughly 5% for pregame sports bets -- high-volume racing bettors receive rebates particularly from advance-deposit wagering sites that reduces the effective takeout rate that they face by half or more.

We decided to apply advanced analytics to test the theory that takeout rates are above the optimal level, which you'll hear from some race fans. To do so, we compiled takeout rates and handle figures from 63 U.S. Thoroughbred tracks from 2017, and we ran a series of non-linear regressions, the results of which you can see on the slide.

Now, we found the average Win/Place/Show takeout rate for these 63 tracks was 17.3%, and the average Win/Place/Show handle per race was $61,000. That's on the vertical axis. The average is that red dot. Sorry, that red dot down low and to the right.

Some tracks had Win/Place/Show takeout rates over 19%. So that blue dot on the right there is a very high WPS takeout rate. Our model found an optimal takeout rate of 15.8%. Now, that's the takeout rate that would maximize the Win/Place/Show handle per race and also the takeout for the track, but 89% of the tracks that we analyzed had Win/Place/Show takeout rates today that are higher than that level.

It's possible that the optimal rate is different for different tracks. So depending on the quality of the races, depending on the fan base, the types of fans, depending on the timing of the race and the racing calendar, that could affect the optimal takeout rate. But the 15.8% that we get from our analysis is the optimal rate for the average track in our sample given the data and analytics that we have today.

Next we analyzed exotic pools, so exacta, trifecta, multi-race bets. We saw little or no correlation between the takeout rates and handle per race, which suggests the takeout rates for exotics are not excessive, which may also seem somewhat counterintuitive.

Let's give you two potential reasons why that might be. The first is that other types of low-wager/high-payoff bets -- think of a lottery bet where you can wager a dollar to win many millions. Other types of low-wager/high-payoff bets also have high house edges.

For example, in sports betting, parlay events, which are multi-event bets, have a house edge over 30%, which makes exotic takeout rates around 20% seem low in comparison. Lotteries often have takeout rates over 35%.

The second reason this may be is the rebates for exotic bets are typically higher than they are for Win/Place/Show bets, which insulates high-volume bettors from the high list price for exotics.

So while we recognize the takeout rate on exotics may seem high to core fans, especially at tracks with the highest takeout rates, so far the analytics don't demonstrate that racing could increase revenue by lowering the takeout on exotics.

Let me give you one last caveat on this. We're looking at the takeout and handle across 63 different tracks. Therefore, this optimal rate we think is useful, but it's a rough indication of what the optimal might be for a particular track.

We don't have access to ADW data. ADWs have precise data on what the rebate is for each customer, exactly what they're betting, what the takeout rates they face are. And with that data, we could probably get a more precise picture of how fans react to takeout rates, the effective takeout rate after rebate.

But we still think this analysis makes the case for tracks to experiment, or at least the 89% of tracks with takeout rates higher than the optimal here, to experiment with lowering their Win/Place/Show takeout rate.

We have to recognize that it takes bettors a while, we think 6 to 12 months, to fully adjust to a change in the takeout rate. So in the short run, if you raise the takeout rate, you might see more takeout for the track, and if you lower it, in the short run you might see less. After bettors adjust over a period of 12 months, we think you're going to see results like the ones we showed in this model.

So the conclusion is we think it's worthwhile for tracks to test lower takeout rates for Win/Place/Show bets.

MIKE SALVARIS: Next we're going to talk about simplified betting. As we heard earlier, when we surveyed fans, they believe the complexity of handicapping is an issue for the sport. 33% of all fans rank it as one of the three most important issues in racing, and 20% of potential bettors said the main reason for not betting is complexity.

Anyone who has tried to read a past performance sheet for the first time knows the difficulty in understanding the jargon and translating that data into a betting decision. Equibase makes available for free a wealth of information; however, basic handicapping tools are not easily accessible.

For example, to access these tools on ADWs and Equibase requires a registration, STATS True Odds and Timeform US require a monthly subscription of $50 to $70 respectively, and DRF Ticket Maker is for more advanced bettors and only covers exotic bets.

Learning to handicap can be much more intuitive if we take a handicap strategy using data from past performances and radically simplify them for the new fan. The goal is to present these angles at the time when the casual fan is ready to place a bet without any effort required to search for the information or to register for the service or pay for the recommendations.

For example, we can use video boards near betting windows and free apps distributed by tracks and television networks to show fans which horse is highest rated based on the speed rating, wins per start, or winning percentage of the current jockey-trainer team.

Next we're going to talk about trainer metrics. Now, it's no secret that field size drives handle. For example, an increase from seven starters to eight starters grows handle by about 15% on average. And field size has declined marginally, about .7% per year since 2011.

When we asked industry experts why horses don't start often as they once did, most believe that trainers are seeking to maximize their win percentage, because the most popular metric used by owners in choosing a trainer.

The problem with win percentage is it has zero correlation with starts per horse per year, and it also doesn't correlate very well with the quality of the owner experience. When we spoke to them, most owners want to see their horses race more often and value in-the-money finishes, not just wins.

Racing would benefit from a trainer metric or metrics that encourages fuller fields and aligns better with owner goals.

Although no metric is perfect, we believe an example of a good metric is the stakes Win/Place/Show per starter per year. The metric counts the number of Win/Place/Show finishes in quality races per year for each horse in training for a particular trainer. It provides an incentive to start horses more often, provided they have a reasonable chance of finishing in the money.

This metric has a strong correlation of about .46 with starts per horse per year, and therefore should encourage fuller fields. Also, when we recut a list of leading trainers by earnings based on this metric, we found several well-regarded trainers still rank high. For example, Bob Baffert, Chad Brown, and Mike Maker are all within the top 15.

However, it also surfaced some lesser known trainers. Seven of the top 15 trained fewer than 60 horses each. This is just an example of a metric. No single metric can identify the best trainer. But field size should benefit if we can shift industry metrics that correlate with starts per horse per year.

DAN SINGER: We're on to our fourth growth theme, which is the legalization of sports betting. Now, many of you or most of you probably know when the U.S. Supreme Court overturned the federal prohibition on sports betting in May of this year, it was perhaps the most significant discontinuity of U.S. sports and gaming since the advent of internet video.

Legal sports books have begun operations in New Jersey and Delaware. I should mention they've been legal for 50 years in Nevada as well. But the new legal sports books in New Jersey and Delaware just started operating in the last couple months, and there is the potential for New York, West Virginia, Mississippi, and Pennsylvania to join them in the next 12 months.

We project that legal sports betting could exceed $120 billion in wagers by 2023, which is about 12 times the handle of racing today. While all forms of gaming ultimately compete for share of the consumers' wagering wallet, we believe that the cannibalization threat from legal sports betting is limited for racing.

Now, here's why. First, racing has continued to prosper in countries that have legal sports betting. For example, Hong Kong, Ireland, Australia, and France all have legal horse racing and all have legal sports betting. And in all four countries, the per capita of gross wins, if you think about the takeout per person in those countries, it's higher for racing than it is for sports betting in those countries, and it's higher for racing in those countries than it is for racing in the U.S. So it's clearly possible for racing to continue to generate handle in competition with legal sports betting.

The second reason that we think the threat from legal sports betting to racing is limited here in the U.S. is that in our research among U.S. horse racing fans, only 4% said they'd bet less on racing if sports betting were legalized in their state. 4%. So that may sound low, but one potential explanation that we think has a lot of merit is that many of those who intend to bet in legal sports books as sports betting becomes legalized, they're already betting informally with friends or illegally with a bookie or an offshore book, and therefore -- or when they happen to be in Nevada. So that limits the extra sports betting that will happen as it goes legal.

For example, last fall, 21% of all U.S. sports fans and 35%, a third, of avid sports fans said they already bet on pro sports.

Now, while we expect the cannibalization threat of legal sports betting to be somewhat limited, that outlook will depend on the racing industry's willingness to respond to that competitive threat in three areas.

First, cross-selling. Given there will be hundreds of millions of dollars per year invested by sports books in developing their websites, developing their mobile applications, investing in television and digital marketing, investing in promotions, and therefore they'll generate millions of accounts for online sports books, racing can benefit by cross-selling to sports bettors, many of whom will be betting online for the first time.

Because races are just two minutes long and they occur year-round, they're a natural way for sports bettors to fill downtime for another sports match or during a game break like halftime.

But cross-selling effectively depends on convenience in placing a bet. That means the bettor should be able to use the same account, set up with the same registration process, using the same payment method as he or she used for sports betting. Racing should just be another tab in the online sports book, just like other sports, and the user experience should be just as similar as possible.

Now, while that sounds fairly intuitive, if states give regulatory authority for sports betting to different state departments than those that regulate racing today, the new department may create different rules about registration processes or operating requirements or marketing requirements.

These are fairly tactical, mundane things, but in writing the new regulations, if those rules are different, it could make it infeasible or impossible to cross sell online. Therefore, we believe the industry should take action to ensure consistent regulations across both sports betting and race betting.

The second imperative we see for racing is innovation in the betting experience. For example, in response to the growth of online sports betting in the U.K., U.K. horse racing bookmakers have innovated by offering insurance or cash-outs or even in-race bets through exchanges, which we don't really have in the U.S. today.

There are many ways the U.S. could innovate in race betting. One we mentioned back in 2011 was single-pool wagering, which could enable faster development of new bet types and also reduce the volatility of odds changes.

But the third and perhaps most fundamental change that racing could make in the response to legalization of sports betting is to offer fixed-odds betting for Win/Place/Show bets.

The argument for fixed odds is that virtually all sports, at least in the U.S., virtually all sports bets are fixed odds, and therefore sports bettors are already accustomed to locking in their odds at the time they place the bet. Although pari-mutuel betting has significant advantages for the track or the race book, many horse bettors dislike the odds changes that can happen in pari-mutuel betting, especially big moves just before or even after the race has started due to the cycles of the tote.

One example. On July 7th, about a month ago, at the Dwyer at Belmont, 45% of all the wagers were received with less than one minute to post. As a result, the odds for the eventual winner, a horse named Firenze Fire, were cut in half just at post time, which generated some frustrating tweets and posts from racing fans that didn't like it. Obviously, that would not happen with fixed-odds betting.

The experience of Australia and the U.K., which offer pari-mutuel betting in parallel with fixed-odds betting, suggests that bettors prefer fixed odds when they have a choice. For example, in Australia, fixed odds now represents 39% of Tabcorp's racing revenues versus only 14% in 2013, and fixed-odds revenue in Australia has grown at 25% per year since 2013, while pari-mutuel revenue has declined at 5% per year since then.

Now, for fixed odds to be successful in the U.S. for U.S. racing, we'd need, first of all, a commission on all Win/Place/Show bets. That should be roughly equal to the net from off-track pari-mutuel bets today. So when a bet is placed off track or in ADW, about 8 to 10% of the wager goes to the track, taxes, and the purse. There'd need to be a commission that kept all of those about the same in fixed-odds betting.

There would have to be a mechanism for setting odds. Right now, the pari-mutuel pool sets the odds. But in fixed-odds betting, the odds maker would have to set the odds, or a third-party odds maker could do it. That's already done in Europe even for U.S. races, and it's done routinely in the U.K. by U.K. bookmakers for U.K. and European races. So it's clearly feasible.

The third thing and importantly is we'd need regulatory changes both in the state hosting the track and in the states in which the bets are placed. But we would expect a fairly warm reception from regulators on fixed-odds betting, assuming the industry was aligned on the terms that were required.

Now, that concludes our review of the findings from our 2018 strategy for racing. Let's do a quick recap of the opportunities that we discussed.

First of all, theme one, innovations in venues. We think racing should invest more in improving the racetrack experience to meet fan expectations that are set by venues in other sports.

Theme two, digital content, data, and marketing, which Mike discussed and Chris as well. We think racing has to invest to create more content for digital distribution, especially social networks. Racing can use the interaction around that content to generate data on fans and then create highly targeted, personalized digital marketing to get casual fans or new fans out to the track, which is still a critical step in fan development.

At the same time, we think racing still has upside on television, building on the successes of the last seven years. And as sports networks start to program studio shows towards new sports bettors, it's a golden opportunity for racing because those shows need racing video every single day to round out  the coverage.

Theme three was advanced analytics. We talked about scheduling to avoid running major races within five minutes of each other, takeout optimization, new trainer metrics, and simple handicapping tools for novice bettors.

Finally the last theme, legalization of sports betting. We think racing needs to work on cross-selling racing on digital sports books, innovating bet types and experimenting with fixed odds for Win/Place/Show bets.

Notwithstanding the rapid pace of disruption in sports and technology, we think racing still has tremendous upside potential.

Just on a personal note, I've had the good fortune to be working alongside The Jockey Club and racing for the last seven years. And often I've been -- as I travel around the country looking at tracks and interviewing race fans and non-fans, I've had the good fortune to be in the company of very wise, experienced, and very outspoken racing experts, including trainers, vets, owners, track managers, online betting operators, breeders, stewards, regulators, buglers and broadcasters. I've interviewed lifelong and veteran handicappers and interviewed and videoed people making their first bet ever.

I, myself, have bet on races at an off-track betting location in downtown Toronto. I've been at the window at Churchill Downs on Oaks Day. I've bet on a PC in a windowless conference room at McKinsey in New York City, which is not a great venue for racing, but we tried it, and I've been at Aqueduct on cold, wet weekdays in the dead of winter, which is for the truly dedicated.

What I can tell you, just in my experience and I think in our collective experience, racing is truly unique. Notwithstanding the competition from other sports, from digital video, video games, from legal sports betting, and from new sports and gaming that we're barely even aware of today, we think the best Thoroughbred racing is going to be exciting forever, at least for the foreseeable future.

Largely that's because of the people who love racing, the people that are passionate about racing, the experience we have together at a great day at the track, as I fully expect today, this afternoon at Saratoga, will be. That's unique, that's special, and that's permanent.

So Mike and I and our colleagues have appreciated your collaboration and the generosity of many others in racing in giving their time and experience to us through this process, and we look forward to returning again in the future to celebrate continued innovation in growth and racing with you.

Thanks very much, and we'll hand it over to Stuart.