scvsafewfdsvf
Let's assume his 17 kids (k) are locals. Let's say he lives in a city with 1 million people. That's 500000 potential other sex mates (assuming hetero) We're only interested in people more less our age, about 20% of the population (actually much smaller), so 100 000 = N
Chance of incest = 1 - p where p = N!/(N^k (N-k)!)
Using the sterling's approximation, and that half his kids are opposite gender, k -> k / 2 = 6.5
About 2 in 10 000 chance of incest. Which are really high for something many people consider worse than death.
Now, my assumption of 20% is too generous. What if it's 10%, and he has more, undiscovered kids (presumably who haven't gotten tested)? Let's say 30 kids (it's happened)
1 in 2 thousand for every mate. You wouldn't get into a car that had those odds of a fatal accident, never mind 6, 10, 30 times.