I suspect the health benefits of Exile-style martial arts are benefits that come from ki control in and of itself, rather than physical fitness in and of itself. Though I
also suspect that having control of one's ki means one has a lot more control over the physical state of one's body.
Also, little trick of betting odds: they're not the actual odds. But rather, the odds that are expected to make a profit.
Yeah. The thing is, if the betting odds are poorly correlated with the real odds, in the long run you'll
lose.
Suppose that you offer bets on matches between comparative unknowns with a reputation for being at best dilettantes in martial arts, versus well-trained, experienced competitors in the top of their game, at 7000:1 odds. If the true odds of the nobody winning are in fact
worse than 7000:1, you will gradually make a profit off of this. But to be confident enough of making a profit that it's worth the huge volatility that results if you lose, you'd have to offer hundreds of thousands of such bets, over hundreds of thousands of separate matches, and never take enough such bets that you could bankrupt yourself paying them.
There's actually a whole industry that works this way: the insurance industry. My car insurance represents me taking on a bet at X:1 odds (for some value of X) that my car won't get wrecked this year. The insurance company is taking the high-probability, low-payoff side of that bet ("yeah, your car probably won't get wrecked") while I take the low-probability, high-payoff side. My incentive for doing so is that while the consequences of me paying my insurance premium are small and predictable, the consequences of a car accident are large and
not predictable. The "bookie" (that is, the insurance company) makes a profit in large part because I'm willing to accept an unfavorable bet if it means security from disaster.
But here's the problems.
One, the insurance industry gets away with this because they accept
thousands or millions of such bets (insurance policies) every year. They can average out even very low risks, like the chance of a house getting struck by lightning.
Two, the insurance industry employs an entire specialist profession of statistician-financiers whose
one job is to figure out the true odds of bad things happening and rationally assess the risks associated with them. Without that, the whole thing would fall apart in short order, because if the true odds of a house getting struck by lightning are one in a hundred thousand, and you estimate them as one in a million, you can lose a lot of money pretty fast.
...
Bookies don't get to issue bets on thousands and thousands of events like this so that they can accept "small payoff 4999 times, huge loss 1 time" situations as something that averages out smoothly over time. For them, it's more like "Maybe I make a tiny profit, and once in a very long while someone who does something like this goes out of business." Most small and medium businesses don't like to take odds like that deliberately.
And, well. As mentioned, you have to be
really freaking confident to offer a bet at 7000:1 odds. An event has to be so rare that you could go to similar events every day for twenty years and plausibly expect to never see it happen. I'm pretty sure a good actuary would advise the bookie that just lost three quarters of a million zeni to Kakara something like:
"While the odds of Cynthia, with her known resume, defeating Mitsuba, with
her known resume, may seem vanishingly small... That isn't what you're offering odds on here. You're offering odds that a little-known challenger who
thinks they should be at a tournament against the world's strongest fighters will defeat one of the elite of the tournament circuit. The odds of
that happening, in a world where secretive training is a thing that can happen, may be very low, but I'm pretty sure they're not 7000:1."