(Submitted by reader Dave R)
On Sunday, 9/11/2011, the first three horse races at Belmont Park in New York City ended with the horses numbered 9, 1, and 1 winning the races, respectively. A spokesperson said the odds must be a million to one against that happening. I’m not sure how many horses were in each race so I can’t figure the exact odds, but it certainly isn’t million to one against. If there were 10 horses in each race the odds of that particular combination would be 1 in 1000.
However so many people bet on that exact 3-pick due to it being the date, a $2 bet only paid off $18, or 9:1.
[EDITOR: While the odds of that combination are 1 in 1000, I imagine the odds of it occurring on that particular day drive it up quite a bit. Anyone wish to do some math for us?]
Ha! I finished reading this at 9:11 a.m. Four stories down on my Facebook news was a post from Gawker with 9 likes and 11 comments—so glad I’m not superstitious.
We need to know which horses were favored to fully determine the odds. If the NY Yankees were playing a minor league team, your statistics are saying that each would have a 50% chance of winning when of course we know this isn’t the case.
If a $2 bet only turned into $18, I’m guessing they put the “favored/best” horses in the 9, 1, and 1 positions.
Good thinking. This one may, indeed, have some outside help from the people coordinating the race, driving down the odds.
This is exactly the thought process most people DON’T bother to undertake when they hear stories like this. Thanks for the feedback.
“While the odds of that combination are 1 in 1000, I imagine the odds
of it occurring on that particular day drive it up quite a bit. Anyone
wish to do some math for us?”
Assuming (for simplicity) that each race had 10 horses and that every horse had exactly the same chance of winning (completely random) then the odds are still 1 in 1000.
The maths:
We turn up to the races on 9/11/2011.
Probability of the date being 9/11/2011: 100% (1)
Probability of horse 9 winning the first race: 10% (0.1)
Probability of horse 1 winning the second race: 10% (0.1)
Probability of horse 1 winning the third race: 10% (0.1)
1 x 0.1 x 0.1 x 0.1 = 0.001 (1/1000)
The pay off was actually $18.60 for a $2 bet so the ratio was 9.3:1, not 9:1.
From what I can find it looks like there are 13 races per racing day. So there are 11 chance for 3 sequential numbers to come up.
9/11 happened on Sep. 11, not Sep. 1. So to combine separate races into one number means you double the odds of a race or races matching a date. You increase the odds even more if you truncate the tens digit to try to match a number (e.g. 18 becomes 8).