Pats to win out?

[BradyFTW!]

Don't think i know that?

Well, if you know that then I'm not sure why you made your last post, because it makes no sense given that fact.

 

[Oswlek]

San Diego is 50/50

They are? Is that the odds that SD keeps it close or wins outright?

 

[Rob0729]

Where are you getting that they're a 7 point favorite this weekend? It's 3.5 everywhere I've seen. As for the rest, I think that's just a fundamental misunderstanding of how often the team that's supposed to win does. Historically, 7-1o point underdogs win outright about a quarter of the time (364 out of 1,383 in 2011, which was the most recent data I could find off hand). So even if the Pats are favored by 7-10 points in each of their remaining games, their expected record based on that spread is still 'just' 3-1.

Bottom line: to have a >50/50 shot at going 4-0 for the rest of the season, you would have to believe that they have about a 90% change of winning each game, more or less. And that's just not the reality of the NFL. The Broncos were favored to win every game they played this season, and somehow they've lost three games. In the first two games they lost, the Pats were 3 (vs. KC) and 4.5 (vs. Miami) point favorites.

I looked at the wrong odds. But even so, the Pats are almost a TD favorite on a neutral field.

And the Pats are arguably one of the two best teams in the league and you arguing they are a 50/50 odds of going on a 4 game winning streak after they just finished a 7 game winning streak with four of the team during that streak having as good or better records than the teams left on their schedule. It just doesn't make sense. You might not want to be a statistician.

I am sorry, but if you are a 9-3 team. You are not a 50/50 odds of going on a four game winning streak unless you are facing four teams with 8-4 to 10-2 records. The Pats may not have a 80-20 odds, but it certainly isn't 50-50.

 

[BradyFTW!]

I looked at the wrong odds. But even so, the Pats are almost a TD favorite on a neutral field.

So what? The game isn't being played on a neutral field.

And the Pats are arguably one of the two best teams in the league and you arguing they are a 50/50 odds of going on a 4 game winning streak after they just finished a 7 game winning streak with four of the team during that streak having as good or better records than the teams left on their schedule. It just doesn't make sense. You might not want to be a statistician.

Yes, that's exactly what I'm arguing, and I even brought the numbers to support it. Did you not read the content of my post, or are you jumping over to ad hominem attacks specifically because you can't actually address the content?

Here's a tip: maybe you should reacquaint yourself with sixth grade albegra before questioning other people's skills. Unless you wish to object to the very concept of percentages, you're objectively wrong about the odds here. Your argument isn't even with me, at this point: it's with math. Incredibly basic math, at that.

In case there are some people who are actually curious about the numbers and aren't just looking for an argument: for the Pats to win out, they would have to be on average 84.08% favorites to win each individual game, since .8408 is the fourth root of .5. Even if the Pats were 10 point favorites in every remaining game

 

[Brady_to_Moss]

Well, if you know that then I'm not sure why you made your last post, because it makes no sense given that fact.

2013: 2nd game: 30-27 jets on second second FG
2012 2nd game: 49-19 pats
2011 2nd game: 37-16 pats
2010 2nd game: 45-3 pats
2009 2nd game: 31-14 pats
2008 2nd game: 34-31 OT loss jets
2007 2nd game: 20-10 pats

Jets are 2-5 since 2007 in 2nd games

Average score: Pats 32.7 PPG
Jets 18 PPG

 

[Rob0729]

So what? The game isn't being played on a neutral field.

And the Pats are favored.




Yes, that's exactly what I'm arguing, and I even brought the numbers to support it. Did you not read the

content of my post, or are you jumping over to ad hominem attacks specifically because you can't actually address the content?

Here's a tip: maybe you should reacquaint yourself with sixth grade albegra before questioning other people's math skills.

You are arguing it is a coin flip that the Pats go on a four game winning streak against inferior teams. Your argument is wrong. You argue you have to have at least a 90% chance of winning every game to be more than 50/50 to run the table and that doesn't make sense. If the Pats have an 80% chance to win every game, their odds of running the table are still far greater chance than 50-50 to run the table. If you have an 80% chance to win each of the next four games, that means you very likely to win every one of those games. The odds are probably something like 70-30 that they win out. You math quite frankly sucks.

For the odds to be around 50-50, the Pats would have to have something like 60-65% chance of winning each game.

The Pats are likely to win the next four games. I don't know the exact odds, but they are clearly better than 50-50.

 

[BradyFTW!]

And the Pats are favored.




Yes, that's exactly what I'm arguing, and I even brought the numbers to support it. Did you not read the

You are arguing it is a coin flip that the Pats go on a four game winning streak against inferior teams. Your argument is wrong. You argue you have to have at least a 90% chance of winning every game to be more than 50/50 to run the table and that doesn't make sense. If the Pats have an 80% chance to win every game, their odds of running the table are still far greater chance than 50-50 to run the table. If you have an 80% chance to win each of the next four games, that means you very likely to win every one of those games. The odds are probably something like 70-30 that they win out. You math quite frankly sucks.

For the odds to be around 50-50, the Pats would have to have something like 60-65% chance of winning each game.

The Pats are likely to win the next four games. I don't know the exact odds, but they are clearly better than 50-50.

Are you serious? I'm not sure what's worse; that you're this bad at extremely basic math, or that you're so aggressively criticizing mine. Do you have a kid? If so, go ask your kid what a conditional probability is; as long as they're older than 10, they will have covered this in school and will be able to help you out. If not, go find a calculator and type out .8 ^4. That's the odds of winning four consecutive games if you have an 80% chance of winning each. It comes out to 41%. So, if you believe that the Patriots have an 80% chance of winning each game left on their schedule, then by definition you believe that they have a 41% chance of winning out.

Keep arguing this point if you want, but you're just making yourself look really dumb. I'm not even making an argument, I'm just explaining how sixth grade math works. It's fine to be ignorant, but ignorant + aggressive = stupid, and it's not fine to be stupid.

 

[Oswlek]

If the Pats have an 80% chance to win every game, their odds of running the table are still far greater chance than 50-50 to run the table. If you have an 80% chance to win each of the next four games, that means you very likely to win every one of those games

No, it is simple arithmetic.

80% * 80% *80% * 80% = 41% chance of winning them all.

Where his statement falls is that these aren't just data points in a vacuum. NE is substantially better than SD and is coming off a loss. Miami's only win in NE since 2001 was against Matt Cassel in 2008, Buffalo is still looking for theirs.

The only game where NE doesn't have a remarkable outside edge is against Jets, who are conveniently the worst remaining opponent.

In reality, assuming good health and the need to win every game for seeding purposes (it is conceivable that NE has the #1 seed locked up heading into week 17), NE probably wins all four 70% of the time in a Monte Carlo-esque simulation.

 

[Rob0729]

Are you serious? I'm not sure what's worse; that you're this bad at simple math, or that you're so aggressively criticizing mine. Do you have a kid? If so, go ask your kid how to do this; as long as they're older than 10, they will have covered this in school and will be able to help you out. If not, go find a calculator and type out .8 ^4. That's the odds of winning four consecutive games if you have an 80% chance of winning each. It comes out to 41%. So, if you believe that the Patriots have an 80% chance of winning each game left on their schedule, then by definition you believe that they have a 41% chance of winning out.

Keep arguing this point if you want, but you're just making yourself look incredibly stupid here. I'm not even making an argument here, I'm just explaining how sixth grade math works. It's fine to be ignorant, but ignorant + aggressive = stupid, and it's not fine to be stupid.

How does that make any sense. You have an 80% chance of winning a game. That means if you play 10 times, you win 8. That means if you have four game like that you play those teams 40 times, you win 36 times. Stats don't work exactly that way, but to say that you have four games that you have an 80% chance of winning them that you have a 41% chance of winning all four doesn't make sense. I don't know how you even come with that math.

You have an 80% chance of beating a team you are most likely going to beat them. You have four games of teams that you have an 80% chance of beating each of those teams, you should beat all four of those teams.

 

[Rob0729]

No, it is simple arithmetic.

80% * 80% *80% * 80% = 41% chance of winning them all.

Where his statement falls is that these aren't just data points in a vacuum. NE is substantially better than SD and is coming off a loss. Miami's only win in NE since 2001 was against Matt Cassel in 2008, Buffalo is still looking for theirs.

The only game where NE doesn't have a remarkable outside edge is against Jets, who are conveniently the worst remaining opponent.

In reality, assuming good health and the need to win every game for seeding purposes (it is conceivable that NE has the #1 seed locked up heading into week 17), NE probably wins all four 70% of the time in a Monte Carlo-esque simulation.

The odds do go down with multiple games (I can't remember the statistical math off hand) . But if you are heavily favored to win every game in a four game stretch like 80% each game, you are still heavily favored to win all of them. My guess if the Pats have an 80% chance of winning each of the next four games, they have a 70-75% chance of winning them all. It certainly does not go down to 41%.

 

[Oswlek]

The odds do go down with multiple games (I can't remember the statistical math off hand) . But if you are heavily favored to win every game in a four game stretch like 80% each game, you are still heavily favored to win all of them. My guess if the Pats have an 80% chance of winning each of the next four games, they have a 70-75% chance of winning them all. It certainly does not go down to 41%.

Rob, it is simple statistical probability. To calculate the odds of concurrent events all happening in a particular sequence, you multiply the odds of each individual event. It isn't inaccurate in any way.

It maybe best illustrated by a two flip coin sequence. If you flip a coin twice, what are the odds it will be heads both times? Well, we know it is 50/50 for each flip, so 50% * 50% = 25%.

Now, let's prove it. Here are all the possibilities.

Heads, heads - winner!
Heads, tails
Tails, heads
Tails, tails

See? Four possible outcomes with one being the target, ergo 25%.

 

[Joker]

That means if you play 10 times, you win 8. That means if you have four game like that you play those teams 40 times, you win 36 times.

32 times...not 36

 

[BradyFTW!]

How does that make any sense. You have an 80% chance of winning a game. That means if you play 10 times, you win 8. That means if you have four game like that you play those teams 40 times, you win 36 times. Stats don't work exactly that way, but to say that you have four games that you have an 80% chance of winning them that you have a 41% chance of winning all four doesn't make sense. I don't know how you even come with that math.

You have an 80% chance of beating a team you are most likely going to beat them. You have four games of teams that you have an 80% chance of beating each of those teams, you should beat all four of those teams.

Yes, in a vacuum you should beat all four. But the odds of all four likely outcomes happening concurrently with one another are a whole lot lower than that.

There's a sequence that you're not taking into account: there's an 80% chance that you win the first game.
Then, to go 2-0, you have to take the 80% chance that you won the first game, and then multiply it by another 80% chance that you win the second game as well. Now you're at a 64% chance that you won both games. While there's an 80% chance that you win the second game, it's only 64% that you win the second game given that you also already won the first.

For the third game, there's an 80% chance that you win this one, but you have to multiply that 80% chance by 64%, to get the probability that you win the third game given that you also won the first two. You're down to a 51.2% chance that you all three games.

Lastly, for the fourth game, you again have an 80% chance of winning, but only a 40.96% (51.2% * 80%) chance of winning that game AND having also won the three games before it.

If you're interested in learning about this stuff, just look up conditional probabilities. If it worked the way that you're suggesting it does, 16-0 (and 0-16) seasons would be a routine occurrence. It would be the expected outcome for the best team in the league, almost every year. But that's not how probability works, and that's what makes stuff like the 2007 Patriots so crazy. Even a team with a 95% chance of winning every game will has a less than 50-50 shot at going 16-0.

 

[Rob0729]

Yes, in a vacuum you should beat all four. But the odds of all four likely outcomes happening concurrently with one another are a whole lot lower than that.

There's a sequence that you're not taking into account: there's an 80% chance that you win the first game.
Then, to go 2-0, you have to take the 80% chance that you won the first game, and then multiply it by another 80% chance that you win the second game as well. Now you're at a 64% chance that you won both games. While there's an 80% chance that you win the second game, it's only 64% that you win the second game given that you also already won the first.

For the third game, there's an 80% chance that you win this one, but you have to multiply that 80% chance by 64%, to get the probability that you win the third game given that you also won the first two. You're down to a 51.2% chance that you all three games.

Lastly, for the fourth game, you again have an 80% chance of winning, but only a 40.96% (51.2% * 80%) chance of winning that game AND having also won the three games before it.

If you're interested in learning about this stuff, just look up conditional probabilities. If it worked the way that you're suggesting it does, 16-0 (and 0-16) seasons would be a routine occurrence. It would be the expected outcome for the best team in the league, almost every year. But that's not how probability works, and that's what makes stuff like the 2007 Patriots so crazy. Even a 10 to 1 favorite will only go 16-0 about 18% of the time.

If I am wrong about the odds, I apologize. I haven't taken stats in over a decade and it is late. I may be wrong.

But even so, we are arguing odds when football isn't about odds. Football is not flipping a coin. There is so much more that goes into it. Odds does not calculate match ups, quality of each teams, etc.

The Pats should win all their last games. They could lose any of the games. But it is better than a coin flip chance they will because you cannot quantify odds like that because the Pats are not flipping coins to beat the opponent.

 

[BradyFTW!]

No, it is simple arithmetic.

80% * 80% *80% * 80% = 41% chance of winning them all.

Where his statement falls is that these aren't just data points in a vacuum. NE is substantially better than SD and is coming off a loss. Miami's only win in NE since 2001 was against Matt Cassel in 2008, Buffalo is still looking for theirs.

The only game where NE doesn't have a remarkable outside edge is against Jets, who are conveniently the worst remaining opponent.

In reality, assuming good health and the need to win every game for seeding purposes (it is conceivable that NE has the #1 seed locked up heading into week 17), NE probably wins all four 70% of the time in a Monte Carlo-esque simulation.

FWIW, I didn't claim that the Pats had a 41% chance of winning out, and if I was making that claim, that math definitely wouldn't be the reason why. I was using Rob's estimates to point out where his math was leading him astray.

 

[Joker]

I understand your "conditional probability" explanation but I do not see the next game or the one after that or the one after that as dependent on the preceding one. If the next game there's an 80% chance of the Pats winning ,then THAT is where it starts and ends. The next game is a totally different island universe unto itself. One can deem an 80% chance of winning, 70, 60, 95, 100% THAT game given an entirely different set of variables that apply to THAT specific game. For example,if, as you put forth, that the 4th game is a 41% probability without any question, then you are assuming that the loss of a starting QB, WR's, D linemen or corners play no part in that specific probability..doesn't pass the "make sense" test.

 

[Oswlek]

I understand your "conditional probability" explanation but I do not see the next game or the one after that or the one after that as dependent on the preceding one. If the next game there's an 80% chance of the Pas winning ,then THAT is where it starts and ends. The next game is a totally different island universe unto itself. One can deem an 80% chance of winning, 70, 60, 95, 100% THAT game given an entirely different set of variables that apply to THAT specific game. For example,if, as you put forth, that the 4th game is a 41% probability without any question, then you are assuming that the loss of a starting QB, WR's, D linemen or corners play no part in that specific probability..doesn't pass the "make sense" test.

That is an entirely different matter altogether.

Your example is like saying, "now that we've gotten 15 heads in a row, what are the odds that the next flip will be heads again?" which is, of course, still 50/50. The preceding flips don't impact the odds of the next in any way. But you still have to stack the odds when projecting from the first flip, even though each event is entirely independent.

 

[BradyFTW!]

If I am wrong about the odds, I apologize. I haven't taken stats in over a decade and it is late. I may be wrong.

But even so, we are arguing odds when football isn't about odds. Football is not flipping a coin. There is so much more that goes into it. Odds does not calculate match ups, quality of each teams, etc.

The Pats should win all their last games. They could lose any of the games. But it is better than a coin flip chance they will because you cannot quantify odds like that because the Pats are not flipping coins to beat the opponent.

I feel like the Pats should win every time they play. I go into every game expecting a win. Simultaneously, I don't expect them to go 16-0 every year. The reason why is the same reason why I don't think it's a near-slam dunk that they win out from here.

We're definitely not flipping coins here, and I think it would be a mistake to write this all off as meaningless abstraction that doesn't apply to 'real' stuff. What I am acknowledging and accounting for is that football is a weird sport, and that the favorite doesn't always win. In fact, the favorite very, very frequently doesn't win. Even 7-10 point underdogs win about a quarter of the time. **** happens, and **** happening is the main reason why Brady doesn't have as many rings as he does fingers to wear them on.

 

[captain stone]

Here are some more odds:

If the Pats don't win every single one of their last 4 remaining games, the odds of their getting to the SB
is 0.0%.

 

[Oswlek]

Here are some ********:

If the Pats don't win every single one of their last 4 remaining games, the odds of their getting to the SB
is 0.0%.

FTFY.