Superbowl teams

[NinerSickness]

I'm not sure where you got that idea, either. AFC North should be a difficult division compared to the South and West.

Whoops! You wrote Broncos & for some reason I read Bengals.

 

[Southieinnc]

No, the first fraction 1/28 (should have been 1/27 because it took 27 years for a repeat Super Bowl to happen; I counted SB 1 by mistake). That takes into account both teams returning.

And saying there's a 1/47 chance because it's happened once is another example of how you're not understanding statistics and odds. Odds are based on how many times something happens ONE TIME out of how many times it takes for it to happen. You can't factor the Super Bowls from 29-47 in the denominator until there's a second occurrence of a repeat Super Bowl. Otherwise, the 1 in the 1/27 is skewed. Once you factor any Super Bowl after that, the 1 becomes a 1.05 or something because the odds don't all of a sudden double when there's a second repeat Super Bowl. So 1/27 VS 1/36 are accurate odds.

Also, by ignoring the odds of one team making it back to the Super Bowl, you're implying that one team making it back somehow lessens the odds of the other team making it back, and that violates a fundamental law of adds and statistics as well. One has nothing to do with the other (either statistically or logically).

You're going to have to concede this point because it's not an opinion; you're just wrong about this one.

Not going to concede and you are not going to reconsider. You may get someone to agree that the odds of the 2 same teams repeating as finalists are (whichever) odds you are qouting above. I'm not buying it. Of course I'm not saying that either team is not good enough to return. I am saying the likelyhood of both teams returning is very, very poor. It is a wager I'd bet against, any year, any 2 teams.
Unfortunately, when they don't make it to the Superbowl (together) this year, you'll quite correctly say it was just not good odds.

 

[cdumler7]

I'm not ignoring the departures; I'm saying the arrivals out-weight the departures from a personnel standpoint. Losing Pitta will hurt a lot, but I think they're gaining enough on defense (Webb, Dumervil, etc) that they're still stronger than last season by a little bit.

The Pats, on the other hand, have had the offseason from hell. I think they & the Broncos are WORSE than last season. The Texans & Benagls? Let's just say I have no faith in Schaub & Dalton. I don't think either one of them will ever win a Super Bowl.

And I don't know what gave you the idea that the Bengals' division is easy. And the AFC South has gotten much stronger, so it's not like Houston's a lock for the 1 seed. The Texans might actually be the 2nd best team in the AFC from a personnel standpoint now that they drafted DeAndre Hopkins; he & Johnson will give defenses fits. But when it comes playoff time, I think Schaub will blow it.

Couple of things...You seem to be missing a couple of parts to their defense at this point. The first being the 2 ILB positions are either going to be manned by a player that had a spinal cord injury and has yet to be cleared to play and a Rookie who just had hernia surgery or even worse the back ups to those two players. Throw in they have a rookie and an average Safety replacing Ed Reed and Pollard. Not sure the middle of that defense is as strong as it was the year before.

Offensively they are now missing 2 of Flacco's favorite targets from the year before and Jacoby Jones now might be the number 2 receiver on the team (they could have the worse receiving core in the NFL at this time). Ed Dickson is injured as well so they are now on their 3rd string TE at least for the time being. Their Left Tackle that is known for being overweight and out of shape can't even make it through practice right now because he didn't condition during the off season.

Also as a Bronco fan I loved Dumervil but there is a reason not many teams were getting on board to pay him as much as the Ravens and Broncos were offering. He is a very one-dimensional player and good OL players usually eliminate him completely from games. He has big games against inferior OL players and then disappears in many other games. His Run Defense also leaves something to be desired. He was a nice addition to the Ravens but whether he is actually worth what was paid for him is still very much up in the air.

 

[NinerSickness]

I'm not saying that either team is not good enough to return. I am saying the likelyhood of both teams returning is very, very poor.

This is my entire point: picking any 2 teams to go to the Super Bowl is very, very poor odds (getting 'em both right). But the odds of a repeat Super Bowl are slightly higher than the odds of picking 2 playoff teams at random using history as the measure. That point is just numerators & denominators.

 

[NinerSickness]

Couple of things...You seem to be missing a couple of parts to their defense at this point. The first being the 2 ILB positions are either going to be manned by a player that had a spinal cord injury and has yet to be cleared to play and a Rookie who just had hernia surgery or even worse the back ups to those two players. Throw in they have a rookie and an average Safety replacing Ed Reed and Pollard. Not sure the middle of that defense is as strong as it was the year before.

Offensively they are now missing 2 of Flacco's favorite targets from the year before and Jacoby Jones now might be the number 2 receiver on the team (they could have the worse receiving core in the NFL at this time). Ed Dickson is injured as well so they are now on their 3rd string TE at least for the time being. Their Left Tackle that is known for being overweight and out of shape can't even make it through practice right now because he didn't condition during the off season.

Also as a Bronco fan I loved Dumervil but there is a reason not many teams were getting on board to pay him as much as the Ravens and Broncos were offering. He is a very one-dimensional player and good OL players usually eliminate him completely from games. He has big games against inferior OL players and then disappears in many other games. His Run Defense also leaves something to be desired. He was a nice addition to the Ravens but whether he is actually worth what was paid for him is still very much up in the air.

These are all fair points, and I've taken them into consideration. However, I think the ILBs are going to be fine at the very least by the time the playoffs roll around, and I think Dumervil's overall value to a defense is much higher than Paul Kruger's. I don't think safety will be a problem either. And Arthur Jones (one of my favourite players in the '10 draft class who had no business lasting 'til the 5th round) is really starting to come into his own.

The defense is why I'm taking the Ravens over the Broncos. I don't think the Broncos are going to be able to control the clock & wear other teams down like the Ravens will. I also think Joe Flacco is becoming a really good QB, and he's not going to be easy to beat.

In fact, I think Houston might have a better overall roster than the Broncos do.

 

[cdumler7]

These are all fair points, and I've taken them into consideration. However, I think the ILBs are going to be fine at the very least by the time the playoffs roll around, and I think Dumervil's overall value to a defense is much higher than Paul Kruger's. I don't think safety will be a problem either. And Arthur Jones (one of my favourite players in the '10 draft class who had no business lasting 'til the 5th round) is really starting to come into his own.

The defense is why I'm taking the Ravens over the Broncos. I don't think the Broncos are going to be able to control the clock & wear other teams down like the Ravens will. I also think Joe Flacco is becoming a really good QB, and he's not going to be easy to beat.

In fact, I think Houston might have a better overall roster than the Broncos do.

I can get on board with Dumervil being better than Paul Kruger although I think long term Kruger would have been a better investment as Dumervil has hit the dreaded 30 and older mark. The MLB position I have a hard time agreeing with. They lost both starters then went and got a guy in FA to be a starter who all of a sudden retired so they already are now looking at 3rd string type guys to step in and start. The fact that McClain has yet to be cleared from his neck injury should be a very worrisome sign of how productive he can be and how healthy he can actually stay for the season. Brown was a good college player but it always takes rookies some time so he will have some growing pains same at the safety position.

Also the fact that you would take the Ravens defense over the Broncos is interesting. Considering the Broncos were a top 5 defense last year and the only real player they lost was Dumervil and they brought in about 3 different guys to replace his production plus a bunch of secondary help that should make the back end a lot better. I guess for me I just would rather take the proven talent over the potential at this point.

Flacco is a decent talent but who is he going to be throwing to? Right now other than Torrey Smith he is throwing to a bunch of no ones. Yes I understand players can develop but they lost some major proven talent to where I think teams can now load up the box against the run and force Flacco to have to try and beat them with only one proven receiver. You role double coverage towards Smith's side and load the box and this offense I think will struggle.

Houston I can agree has just about as talented of a roster as the Broncos. Biggest difference would be the quarterback spot where nobody would think Shaub is anywhere near Peyton's talent. I think the Broncos offense looks a lot more dangerous than what Houston has but I think Houston can have the more talented defense so that could be a wash at this point. Right now in my opinion it is Broncos have the best roster in the AFC followed by Houston then Cincinnati.

 

[France_Steve]

No, the first fraction 1/28 (should have been 1/27 because it took 27 years for a repeat Super Bowl to happen; I counted SB 1 by mistake). That takes into account both teams returning.

And saying there's a 1/47 chance because it's happened once is another example of how you're not understanding statistics and odds. Odds are based on how many times something happens ONE TIME out of how many times it takes for it to happen. You can't factor the Super Bowls from 29-47 in the denominator until there's a second occurrence of a repeat Super Bowl. Otherwise, the 1 in the 1/27 is skewed. Once you factor any Super Bowl after that, the 1 becomes a 1.05 or something because the odds don't all of a sudden double when there's a second repeat Super Bowl. So 1/27 VS 1/36 are accurate odds.

Also, by ignoring the odds of one team making it back to the Super Bowl, you're implying that one team making it back somehow lessens the odds of the other team making it back, and that violates a fundamental law of adds and statistics as well. One has nothing to do with the other (either statistically or logically).

You're going to have to concede this point because it's not an opinion; you're just wrong about this one.

The bolded part is bad math and bad logic. Taking more information into account will not skew the statistic it will make it more accurate.

 

[France_Steve]

Here is a better way to look at it: What is the probability that a team from either division returns to the super bowl? Since 1970(because that's when Wikipedia starts the list), the NFC champion has returned 6 times, so 6/43= 14%. The AFC champion has returned to the superbowl 10 times, so 10/43= 23%. What is the probability that both of these independent happen at the same time? Multiply them together. .14x.23= .0322. So there is approximately a 3% chance that the same teams play in the super bowl. 1 in 27 gives you 4% chance, so the probability remains the same, but you got that probability using incomplete data.

 

[NinerSickness]

The bolded part is bad math and bad logic. Taking more information into account will not skew the statistic it will make it more accurate.

That is 100% wrong. Any statistics professor will tell you when you're factoring odds using real-life occurrences, you record your denominator every ONE TIME the occurrence you're testing happens.

For example, if you're rolling a 500 sided die and you're doing a real-life test to see what the odds are of rolling a 499 are, you can't arbitrarily roll it 20 times and get 499 zero times and conclude the test shows there's a 0% chance of rolling a 499. On the same token, you can't roll it a bunch of times and then finally roll the a 499 (say after 521 tries hypothetically) and then arbitrarily decide that you're going to roll the die 75 more times (whether you roll a 499 or not) to get a "more accurate" reading. One would only do that in order to purposely skew the data because the odds are that you're not going to roll a 499 in 75 tries. The only way that would work is if you happened to roll a 499 on the 75th attempt. If you were to continue rolling the die for better accuracy, you would have to keep rolling it until you rolled a 499 a SECOND TIME and then record your denominator.

The Super Bowls following Super Bowl 28 are like rolling the die more times. You can't use those Super Bowls as part of the odds unless you were to argue that there was a 0% chance of a repeat Super Bowl in those years (and we know that's not the case). If SB 28 is any indication, it makes sense that there hasn't been another repeast SB since then because using the first 28 SBs as the first set of numbers (1/27 chance) one would predict that another repeat SB would occur some time around SB 55 (28 + 27).

This is how odds and statistics work. Put quotation marks around the word math all you want, but this is math.

 

[NinerSickness]

By the way France Steve, you counted wrong. There have been 13 repeat SB teams, not 16.

If you want to go that way, you'd take .1413 X .1413, which is 2.00%

Picking a playoff team from the previous year at random would give you .1072 X .1072, which is 1.15%.

2.00% > 1.15%.

 

[Southieinnc]

I really, really, wish I could take that bet. (bold print to imply I have junk math figured out)

 

[NinerSickness]

I really, really, wish I could take that bet. (bold print to imply I have junk math figured out)

You can take that bet. Go ahead. The problem is that there's a 96.87% chance we're both going to be wrong (.9800% X .9885) going by history.

But what the heck; let's do it. I'll take the Niners & Ravens. You take 2 randumb playoff teams from last year. If you're right, I'll use some pre-approved avatar of your selction for an entire year. If I'm right, you use this picture as yours for an entire year. Deal?

And by the way, calling math "junk math" doesn't change the fact that it's still math.

 

[rmilia1]

IMO the 3 best teams in the NFL all play in the NFC ( SF, Seattle, Atlanta). Id put Denver 4th ( lets be honest it took a MIRACLE for Baltimore to win that game last year and Denver didnt even really play well). So that pretty much leaves me with Denver in the AFC. As to the NFC participant Id give Atlanta a slight edge right now because of the injuries to Crabtree and Culliver for SF. I do think SF has more talent than Atlanta overrall but the Niners weakest point of D last year was their secondary and losing Culliver certainly isnt going to help that. IF Seattle can get homefield Id go with them to represent the NFC but I think the more difficult schedule probably results in a 11ish win season and thats not going to be enough for homefield.

 

[geneh_33]

I'll go with Atlanta and Denver.

Atlanta: I see Seattle and San Frans QB's falling back to earth this coming season. Felt like Atlanta's playoff loss last year was more of their own doing than anything else. Think they are a better team this year.

Denver: At some point Peyton starts showing his age. Don't see it being this season though. Do they have a hangover from last seasons devastating playoff loss? I say no. Picking up Welker helps Denver more than it hurts New England, (Amendola). Losing two LB's might hurt their defense, but Peyton can play with the the clock as good as any QB in the league, and that benefits any defense. Had Ravens beat last year. Won't make the same mistake again. I see them going to the SB.

Good man!

Yeah, we Atlantans want our revenge on Denver and what they did to us in the Superbowl last time.

AND WE MEAN TO HAVE IT!!!!!!

BAHAHAHAHAHAHAHAHAHAHAHA!!!!!!!!!!!!!!!!!!

 

[France_Steve]

That is 100% wrong. Any statistics professor will tell you when you're factoring odds using real-life occurrences, you record your denominator every ONE TIME the occurrence you're testing happens.

For example, if you're rolling a 500 sided die and you're doing a real-life test to see what the odds are of rolling a 499 are, you can't arbitrarily roll it 20 times and get 499 zero times and conclude the test shows there's a 0% chance of rolling a 499. On the same token, you can't roll it a bunch of times and then finally roll the a 499 (say after 521 tries hypothetically) and then arbitrarily decide that you're going to roll the die 75 more times (whether you roll a 499 or not) to get a "more accurate" reading. One would only do that in order to purposely skew the data because the odds are that you're not going to roll a 499 in 75 tries. The only way that would work is if you happened to roll a 499 on the 75th attempt. If you were to continue rolling the die for better accuracy, you would have to keep rolling it until you rolled a 499 a SECOND TIME and then record your denominator.

The Super Bowls following Super Bowl 28 are like rolling the die more times. You can't use those Super Bowls as part of the odds unless you were to argue that there was a 0% chance of a repeat Super Bowl in those years (and we know that's not the case). If SB 28 is any indication, it makes sense that there hasn't been another repeast SB since then because using the first 28 SBs as the first set of numbers (1/27 chance) one would predict that another repeat SB would occur some time around SB 55 (28 + 27).

This is how odds and statistics work. Put quotation marks around the word math all you want, but this is math.

If you are going to factor odds using real life occurrences then you need that occurrence more than one time. Otherwise you risk basing your theory on a statistical anomaly. This sample size is too small. Since it has only happened one time that statistic is incomplete and worthless. you can't roll the 500 sided dice twenty times, get 499, and say that the probability is 1 in 20 without rolling any further.

The second bolded sentence is also incorrect. One would predict that on average a repeat super bowl would happen every 28 years, not that it would happen precisely on a 28 year cycle. Every year there is a 3% chance it happens. Just because it does not happen for twenty years does not mean it has a greater chance of happening. A better way to say it would be "I predict that after 58 super bowls there will have been two repeat super bowls." It is not happening on a 28 year cycle, it is happening on average 28 years.

I never put quotation marks around the word math. My only point was that using 1 in 27 is inaccurate. The method I used is more accurate for predicting the probability of the same teams playing each other in the SB. Also, I rechecked and 17 is the correct number, unless Wikipedia is wrong.

List of AFC champions - Wikipedia, the free encyclopedia - 72, 73, 75, 79, 82, 91, 92, 93, 98, 04
List of NFC champions - Wikipedia, the free encyclopedia -71, 74, 78, 83, 89, 93, 97

I have no problem with your prediction, I was just pointing out that your method of showing something statistically was inaccurate.

 

[NinerSickness]

If you are going to factor odds using real life occurrences then you need that occurrence more than one time. Otherwise you risk basing your theory on a statistical anomaly.

Totally agree. That's why I supplemented the 1/27 odds with the 13/92 times 13/92 odds (but you're right that it's 17 not 13).

The second bolded sentence is also incorrect. One would predict that on average a repeat super bowl would happen every 28 years, not that it would happen precisely on a 28 year cycle.

27 year cycle, but yes. I know that. That's why I said "around' SB 55 it should happen again. It could be this year, or it could happen 100 years from now & then happen again 101 year from now. My point wasn't to say I could predict WHEN it would happen; my point was that the years following SB 28 (using the both-teams method) can't be recorded as part of the odds until it happens again because that would fall under the category of an arbitrary number of tests. Like I said, one must record the denominator every time the test shows the result you're testing, not based on some arbitrary number.

I never put quotation marks around the word math.

That part was directed at Southieinnic. I should have been more clear on that. And you're right that 1/27 is far too small of a sample size, which is why I went with your method. You're right; there were 17 repeat teams. I was only looking at winners & losers, but some teams won & lost in consecutive years. Thanks for the correction. So the real odds using the conference method is 17/92 X 17/92, which is 3.41%.

Weird; that's really close to 1/27 or 3.70%.

Either way, historically, repeat teams occurred much more often than randomly selected playoff teams from the previous year. 3.41% > 1.15%.

You down for my bet Southieinic?

 

[France_Steve]

Totally agree. That's why I supplemented the 1/27 odds with the 13/92 times 13/92 odds (but you're right that it's 17 not 13).



27 year cycle, but yes. I know that. That's why I said "around' SB 55 it should happen again. It could be this year, or it could happen 100 years from now & then happen again 101 year from now. My point wasn't to say I could predict WHEN it would happen; my point was that the years following SB 28 (using the both-teams method) can't be recorded as part of the odds until it happens again because that would fall under the category of an arbitrary number of tests. Like I said, one must record the denominator every time the test shows the result you're testing, not based on some arbitrary number.



That part was directed at Southieinnic. I should have been more clear on that. And you're right that 1/27 is far too small of a sample size, which is why I went with your method. You're right; there were 17 repeat teams. I was only looking at winners & losers, but some teams won & lost in consecutive years. Thanks for the correction. So the real odds using the conference method is 17/92 X 17/92, which is 3.41%.

Weird; that's really close to 1/27 or 3.70%.

Either way, historically, repeat teams occurred much more often than randomly selected playoff teams from the previous year. 3.41% > 1.15%.

You down for my bet Southieinic?

Did not mean for this to get this long. I was being the equivalent of a grammar Nazi. 1/27 gets you really close to what I got, I maintain that it was coincidence and that 1/27 should not be referenced because it is too small a sample size. Agreed with most everything you said otherwise.

 

[Southieinnc]

Totally agree. That's why I supplemented the 1/27 odds with the 13/92 times 13/92 odds (but you're right that it's 17 not 13).



27 year cycle, but yes. I know that. That's why I said "around' SB 55 it should happen again. It could be this year, or it could happen 100 years from now & then happen again 101 year from now. My point wasn't to say I could predict WHEN it would happen; my point was that the years following SB 28 (using the both-teams method) can't be recorded as part of the odds until it happens again because that would fall under the category of an arbitrary number of tests. Like I said, one must record the denominator every time the test shows the result you're testing, not based on some arbitrary number.



That part was directed at Southieinnic. I should have been more clear on that. And you're right that 1/27 is far too small of a sample size, which is why I went with your method. You're right; there were 17 repeat teams. I was only looking at winners & losers, but some teams won & lost in consecutive years. Thanks for the correction. So the real odds using the conference method is 17/92 X 17/92, which is 3.41%.

Weird; that's really close to 1/27 or 3.70%.

Either way, historically, repeat teams occurred much more often than randomly selected playoff teams from the previous year. 3.41% > 1.15%.

You down for my bet Southieinic?

Did not mean for this to get this long. I was being the equivalent of a grammar Nazi. 1/27 gets you really close to what I got, I maintain that it was coincidence and that 1/27 should not be referenced because it is too small a sample size. Agreed with most everything you said otherwise.

Wow, you guys can type!

1) Ninersickness, Name the bet. ( you were saying the 9ers and Ravens will repeat?)

2) France_Steve, I put quotes around "math" rather than say " this is a total crock of shit" I'm have now clarified?

 

[Brees#1]

My prediction is Denver or San Diego in the AFC................San Diego I believe gets a wild card and it depends how Denver treats that week 15 game against them. If chargers play Houston or any other AFC South winner they are winning that game. If New England, its not impossible but they will probably lose. The broncos could have a rematch that has even more knowledge than the ravens did last year. They need to be careful.

In the AFC I hope to Heavens it's my saints. I believe either the saints or falcons come out of the NFC. It is the NFC South's turn but if the NFC South is skipped over again like in 07 after the three other divisions represented the last three years then Chicago will finally get back there.

So its Denver or San Diego versus New Orleans, Atlanta, or Chicago.

 

[NinerSickness]

Wow, you guys can type!

1) Ninersickness, Name the bet. ( you were saying the 9ers and Ravens will repeat?)

I'll take the Niners & Ravens, you take 2 other teams to make it to the SB next year. If your teams go, I'll use a pre-approved picture of your choice as an avatar for 1 year (just make sure I clear it first; nothing lewd). If neither of us get it right, the bet is void. If I get it right, you have to use this picture as an avatar for 1 year (I'll clean up the corners before then):

Deal?

2) France_Steve, I put quotes around "math" rather than say " this is a total crock of shit" I'm have now clarified?

I think the quotation marks were pretty clear, but the numbers we produced are still right.