This is the second of three pieces examining the structural condition of American horse racing.In Part 1, I showed that the racing habitat is collapsing not because horses can't run, but because race days are disappearing, 182 per year since 2022 and accelerating. More than 30 independent jurisdictions, with no centralized calendar, explain where the collapse is worst: isolated states are dying while connected corridors survive. And where the subsidy is chained to the calendar, as in Pennsylvania, the schedule is written around the slot floor, not the horse.But structure is only part of the puzzle. To understand why race days are disappearing we need to follow the money. And when you do, you arrive at a conclusion the industry has been reluctant to say out loud: horse racing, as a wagering business, is no longer self-sustaining.Horse racing has two revenue inputs. People buy horses, and people bet on horses. Everything else, purses, salaries, track maintenance, is derived from those two. Start with the betting. Total U.S. handle in 2025 was approximately $11.8 billion.About 20% to 30% of that, roughly $3 to $4 billion, comes from computer-assisted wagering operations. Love them or hate them, CAW players are the reason the handle is still at $11.8 billion. Without them we are looking at roughly $8 billion, which in inflation-adjusted terms is less than a third of the 2003 peak. The human bettor, the person at the track or on their phone wagering because they have a hunch about a horse, has largely left the building.Now follow the money through the system. A track's revenue comes from taking a percentage of that handle, the takeout. Assume a 15-25% takeout, adjust for the rebates tracks give to CAW and other high-volume players, and you end up with net wagering revenue of $1.2 to $1.8 billion.Total purses in 2025 were $1.38 billion.Purses alone consume most or all of the net wagering revenue. What is left to run the racetracks, to pay stewards, starters, ambulance crews, maintain the surface, market the product, pay HISA? Under realistic assumptions, near zero or negative.The wagering business cannot keep the lights on.So what does? In 2001, purses were about 7.9% of the handle, a ratio stable for decades. If that ratio held in 2025, the $11.8 billion handle would support about $938 million in purses. Actual purses were $1.38 billion. That $438 million gap, roughly 32% of all purse money, comes from the same machines we met in Part 1, casino slots and historical horse racing terminals at racinos.This is not a supplement. Take it away and purse levels across the country collapse overnight. And it is revenue that has nothing to do with the quality of racing. It comes from slot machines in buildings that happen to be attached to racetracks, distributed through state-level agreements that can be renegotiated or eliminated at any time. In Part 1, the slot money dictated the racing calendar. Here it dictates the purses. Record purses while the handle declines is not a sign of a healthy industry. It is the signature of a subsidized one.Meanwhile the broader betting market is booming. Since the Supreme Court struck down PASPA in 2018, legal sports betting has exploded. Americans legally bet $167 billion on sports in 2025, an 11% increase from 2024. That does not even include prediction markets like Kalshi and Polymarket, where people wager millions on elections, on weather, on the date of the next Fed rate cut, on whether Jesus would return in 2027. There is no shortage of appetite for betting in this country.Our entire industry handle is 6.6% of the legal sports betting market. We are not at the table.So why not?Credit: Coady MediaConsider this year's Kentucky Derby. The broadcast drew a record 24.4 million viewers on NBC, up 11% from last year and the most-watched Derby in the network's history. And handle on the race fell 3.8%, from $234 million to $225 million. A record audience watched and did not bet. The problem is converting that attention into action.Our bets tell a layered story, but they are not easy to understand. They require a significant amount of knowledge just to grasp what the bets are and how to place them. Any new player must overcome a high degree of friction, and most will abandon it for something else.Now look at what we compete with. On platforms like FanDuel's sportsbook, binary bets, whether the Knicks score on the next possession, facilitate in-game wagering. These menus create continuous action, like scrolling through a feed. A bet resolves in seconds, then another, then another. They have figured out that this rhythm creates habits.This is where we need to evolve. The goal is to turn a curious newcomer into a regular with as little friction as possible, and build the habit that fuels long-term adoption. The science of operant conditioning tells us two things turn a reward into a habit: how unpredictable it is, and how quickly it arrives. We have the first covered. Horse racing sits at the intersection of chance and skill. You can study a past performance, read the track bias, watch the workouts, and still get beat by a 23-1 shot nobody saw coming. You can get good at this game, but you can never master it. It is the second we have never built. The shorter the gap between an action and its result, the more powerfully the behavior is reinforced. That is the loop the sportsbooks engineered, and it is the one thing our structure works against. A race lasts about two minutes, but the wait for the next one at a single track can run half an hour.Imagine instead a centralized system that synchronizes post times across the country, so a race goes off every couple of minutes and a first-time player can bet the next one from a single screen, with a single tap. One race resolves, then another, then another, the same rhythm that hooks a sportsbook customer, except every result is live horses coming down the lane. That is a stream no sportsbook can replicate. The unpredictable part has always been ours. All we have to build is the rhythm.And the long-term health of our pools depends on the human bettor. CAW has fueled our growth, but even the algorithms benefit from a deeper human pool, because of how they bet. They run profitability decay models: they calculate expected value, size their bets, and stop at the point of diminishing returns. That point is not fixed. Every human who leaves for a sportsbook or a prediction market thins the pool. The odds compress. The expected value shrinks. The machines bet less. Human and machine leave together, each accelerating the other's exit. The decline is not a cliff. It is a drain that gets wider the longer it runs. CAW will follow the humans wherever they go. Our job is to give them a reason to stay.But while racetrack economics generate the headlines, there is a larger economic force holding this industry together, one that gets almost no attention. The largest single source of capital in American horse racing is not the bettor and not the casino.It is the owner.The post State of the Industry: The Handle appeared first on TDN | Thoroughbred Daily News | Horse Racing News, Results and Video | Thoroughbred Breeding and Auctions.