| AR said in his Presser that when the official shows him the coin before the toss he calls opposite of what is showing (I am sure that this is based on research). Official showed "heads" so AR called "tails", when the coin never turned over & he reflipped, the official showed "tails" but did not give AR a chance to call it "heads", which he would have done. I wonder if AR & MM could have put up more of a protest when it happened, or if there is protocol for a "reflip" scenario? |
| 1 |
Replies sorted oldest to newest
So a coin toss is not 50/50? Perhaps this is correct - at some point the Patriots called 19 out of 25 correctly.
Sounds like superstition, not research. A coin flip is always a 50-50 proposition.
If AR says they always call the side that is facing down I would assume they have a reason for doing so?
That assumes a fair coin.
Sure, superstition. The OT rules still strike me as inherently unfair and I'd like to see them move toward the college system where both teams get a possession.
michiganjoe posted:A coin flip is always a 50-50 proposition.
Provided, you know, the coin actually flips while in the air.
They showed on camera how the coin stayed flat the entire time it was in the air.
Just more craziness that is Packer playoff football
michiganjoe posted:Sounds like superstition, not research. A coin flip is always a 50-50 proposition.
Persi Diaconis, professor of statistics and mathematics at Stanford University claims it's 51-49 biased towards which ever face is shown before the flip. He does an absolutely horrendous job trying to explain why in that video above.
The Patriots are known to have a database with each ref's height and thumb strength from the combine. Couple that with the phase of the moon and current barometric pressure, they can get the odds over 53/47. As recently as 2013, it was suspected they were shaving the edges of coins to change the weight.
I remember back in high school a math teacher putting the 50/50 heads tails theory to the test. He had us break up into groups and toss a coin 100 times while keeping track of which side came up after every toss. There were ten groups flipping coins and none of them ended up 50/50. I forgot most of what I learned back then but that experiment has always stuck with me for some reason. LOL
michiganjoe posted:Sure, superstition. The OT rules still strike me as inherently unfair and I'd like to see them move toward the college system where both teams get a possession.
No, the college system sucks. I'm all for a system that if the first team scores a TD, then that team still has to line up, kick the ball off so the second team gets at least one possession. But the foolishness of erasing the kickoffs and just starting at the 25 yard line, no way should that make it into the NFL.
I like the old OT rules. First team to score wins.
Don't let the opponent drive down the field and score a touchdown. Again.
GEEMAN posted:I remember back in high school a math teacher putting the 50/50 heads tails theory to the test. He had us break up into groups and toss a coin 100 times while keeping track of which side came up after every toss. There were ten groups flipping coins and none of them ended up 50/50. I forgot most of what I learned back then but that experiment has always stuck with me for some reason. LOL
Binomial distribution would like a word with you.
More seriously, the margin or error for a random sample is 100/sqrt(x), where x is the number of items selected and the number is a percentage. So for x=100, the margin for error is 10%, which means that two standard deviations (about 95.5% of the time) of the outcome should be within 10% of the expected value. Since a fair coin should be a 50% proposition, with 100 flips you should wind up with somewhere between 40 and 60 heads 95.5% of the time. An exact 50-50 split isn't particularly likely.
If you go up to 400 coin flips you get down to a 5% margin or error, so 95.5% of the samples would be between 180 and 220 heads, with the exact 200-20 split being even less likely. And if you go all the way up to 10000 flips you'll be down to a 1% margin of error, with 95.5% of the samples being between 4900 and 5100 heads.
Back when I was a kid and we got our first computer, a TI-99/4A, I sort of knew about the binomial distribution, so I wrote a BASIC program to flip 10 virtual coins at the same time using the computer's randomize function. Not that they used the word virtual at the time, or more importantly, not that I knew a darn thing about pseudorandomness.
I know some of those words
Hilarious how the ref botching the coin flip in OT wasn't even their biggest fuk up of the game. This crew couldn't decide if they wanted to call a tight game or let them play, it seemed to vary quite a bit.
Predictable randomness ? LOL
There were missed calls on both teams. I like that the refs let the boys play in this one.
