SOOPOO
Joined: Aug 8, 2010
• Posts: 8108
February 7th, 2012 at 7:51:29 AM permalink
Quote: Wizard

From 2000 to 2010 the probability of a safety was 5.8%. That makes it fair laying 1633 against a safety. The fact that there were safeties in Super Bowls 2008 and 2012 will not deter me from making the same bet next year. It isn't whether you win or lose, it's whether or not you had a good bet.

Wiz- you did not include in your analysis which two teams were playing. Teams with strong pass rushes (the Giants) against a team coached by a man not afriad to be aggressive and call pass plays deep in his own zone (the Patriots) would have a far higher likelihood of a safety than, say, a chicken shit team like the Bills who would run 3 times rather than 'risk' something bad happening. Your assumption that all matchups have an equally likely chance of a safety may be incorrect. If the average number of TDs in a game is 7, but the Ravens are playing the Bills, I would guess that 5 1/2 is the fair line. If its the Saints and Packers, maybe 9. Safety chances may vary, too... but its not as easy to track.
DorothyGale
Joined: Nov 23, 2009
• Posts: 639
February 7th, 2012 at 7:52:37 AM permalink
Did anyone post this link yet? Good stuff, that ...

Fan who bet on first score safety and won \$50k is donating it to charity

--Ms. D.
"Who would have thought a good little girl like you could destroy my beautiful wickedness!"
hhhccc
Joined: Jun 10, 2010
• Posts: 59
February 7th, 2012 at 7:57:25 AM permalink
Quote: Nareed

Heretic ;)

There have been very few safeties in the Super Bowl, becasue there have been few safeties overall. It's a rare occurence, after all. So it's a good bet to wager on no safeties. And hindsight is 20/20.

As for a TD to open the score, that's reasonable. Given two highly defensive teams, I'd tend to favor a field goal. Otherwise the first score is often a TD.

Question: what's more unusual, ie what happens less often, a game with a safety or a game without any TDs?

Good bet on no safety? YES!
Good bet on first score is TD? YES!

My question is simply betsizing.

I have no doubts that they were very good, solid bets. I also trust and agree with Wizard's numbers on the probability of a safety.
All I am questioning is relative bet-sizing versus other prop bets out there. Surely there were many other good prop bets, and thus to lose, even after losing the safety bet, seems like a poor spreading of bankroll.
Wizard
Joined: Oct 14, 2009
• Posts: 23333
February 7th, 2012 at 9:02:01 AM permalink
Quote: SOOPOO

Wiz- you did not include in your analysis which two teams were playing. Teams with strong pass rushes (the Giants) against a team coached by a man not afriad to be aggressive and call pass plays deep in his own zone (the Patriots) would have a far higher likelihood of a safety than, say, a chicken shit team like the Bills who would run 3 times rather than 'risk' something bad happening. Your assumption that all matchups have an equally likely chance of a safety may be incorrect. If the average number of TDs in a game is 7, but the Ravens are playing the Bills, I would guess that 5 1/2 is the fair line. If its the Saints and Packers, maybe 9. Safety chances may vary, too... but its not as easy to track.

I factor in the spread and the total in the probability for the specific game. In the case of the safety bet it doesn't make much difference, so I didn't want to muddy the waters with discussion about that. However, since you bring it up, I put the probability of a safety in Sunday's game at 6.1%, a little higher than the 5.8% average.

I'm sure my numbers would be a little sharper if I considered individual team strategy, but I don't deem it a good use of my time.
It's not whether you win or lose; it's whether or not you had a good bet.
Nareed
Joined: Nov 11, 2009
• Posts: 11413
February 7th, 2012 at 9:27:57 AM permalink
Quote: Wizard

I'm sure my numbers would be a little sharper if I considered individual team strategy, but I don't deem it a good use of my time.

Do you try individual team stats? I mean, if the Springfield Isotopes, for example, comit more safeties than average, does that influence your calcualtions? Of course, safeties being so rare it probably doesn't.

Was there a bet on whether a team would make a 4th down conversion? That would be a good bet to make on the Pats on any close game.
Donald Trump is a fucking criminal
dwheatley
Joined: Nov 16, 2009
• Posts: 1246
February 7th, 2012 at 9:48:21 AM permalink
Quote: Wizard

I factor in the spread and the total in the probability for the specific game. In the case of the safety bet it doesn't make much difference, so I didn't want to muddy the waters with discussion about that. However, since you bring it up, I put the probability of a safety in Sunday's game at 6.1%, a little higher than the 5.8% average.

I'm sure my numbers would be a little sharper if I considered individual team strategy, but I don't deem it a good use of my time.

Philosophical aside about probabilities: I have a strong background in statistics and probabilities, even if I mess up the calculations occasionally. I can explain some really hard stuff to colleagues when I need to. That being said, I am always bothered by these probabilities of events occurring, especially during human contests like sports. It's not a die being tossed, it's people! You can look at the historical data and grab a sample mean, and hope it's the 'best estimator' for the true probability... but what was the REAL probability of it happening? Impossible to determine, I'm sure. Coin Flip + opening snap + first coaching calls of the game all came together to cause the safety. The probability even evolved over time, like a die spinning on the table.

And then there's always the chance, maybe not this time, but in any sport, that the players were in on it somehow. Blows my mind.

tl:dr? Probabilities that can't be expressly measured make my head hurt.
Wisdom is the quality that keeps you out of situations where you would otherwise need it
Wizard
Joined: Oct 14, 2009
• Posts: 23333
February 7th, 2012 at 9:55:44 AM permalink
Quote: dwheatley

That being said, I am always bothered by these probabilities of events occurring, especially during human contests like sports. It's not a die being tossed, it's people!

I'd rather bet on something perfectly quantifiable too, but it isn't easy find a card game with a player advantage any more. Maybe my number crunching in sports betting isn't perfect, but it has served me well through the years.
It's not whether you win or lose; it's whether or not you had a good bet.
hhhccc
Joined: Jun 10, 2010
• Posts: 59
February 7th, 2012 at 1:57:08 PM permalink
dwheatley, take two random teams and put them in the Super Bowl. what do you think the Safety line would be? Of course, it would be very close to the same as NE/NYG! How do I know? Well, because it is that way every time there is a Safety line regardless of the teams in the Super Bowl, regardless of the defenses, offenses, line or total. That tells us the line is just wrong, the books or the Yes bettors didn't know anything special about this game that the pro bettors didn't know.
charliepatrick
Joined: Jun 17, 2011
• Posts: 2425
February 7th, 2012 at 5:03:42 PM permalink
Quote: DorothyGale

Did anyone post this link yet?...Fan...won \$50k is donating it to charity...

Yes ...here... see later in the same thread when tmz reported it with a picture. This was my original post before that appeared and it seems the UK bookies were initially a bit more generous in their odds.
Quote: charliepatrick

(1) fwiw The odds for a safety at the first play for each team on one of the uk bookmakers a few hours before kick off was 125/1.
(2) Seems someone managed \$200 to win \$15,000; and the BBC on their live showing of the Superbowl were saying bookies might have lost £20k (though can't find anything about it on the internet).

And this win seems to be on several websites ..
\$50,000 won - ...the punter placed \$1000 and only won \$50,000. So I think the odds were 50-1. However same story here.

btw when an outsider wins (such as a UK horse race) it is usually the best result for the bookmaker. However I remember one race (many years ago) where it was the worst for one bookie as they had taken a bet of £150 to win £5,000. Also at Newton Abbott where a 66/1 won the last race it was easy to see that two people on the track had found the winner, everyone else had left!

98Clubs
Joined: Jun 3, 2010