Sando interview

[scakfan]

I sure miss his write ups. Any who, the stats he provided in an interview on 710 were pretty cool. No Seahawk team since 2001 has come back and won from a 21 point deficit. Okay, how about this one! Teams with a 95% chance of losing from the 2 qtr. of a game (Stats provided by HUH?? Help?) are 3-19 this year with RW winning 2 out of 2 ( my guess is Houstan and the Bucs) and Luck winning 1 out of 1 ( I think against us or Houstan, probable the later). The rest of the teams that faced that obstacle all had losses. Sorry about not having Sando's source, just lazy tonight I guess.

 

[telerion]

I sure miss his write ups. Any who, the stats he provided in an interview on 710 were pretty cool. No Seahawk team since 2001 has come back and won from a 21 point deficit. Okay, how about this one! Teams with a 95% chance of losing from the 2 qtr. of a game (Stats provided by HUH?? Help?) are 3-19 this year with RW winning 2 out of 2 ( my guess is Houstan and the Bucs) and Luck winning 1 out of 1 ( I think against us or Houstan, probable the later). The rest of the teams that faced that obstacle all had losses. Sorry about not having Sando's source, just lazy tonight I guess.

Uhm . . . what? Sounds like a cool stat (or maybe two) is in there, but I can't make out what you're saying.

 

[IcedHawk]

I sure miss his write ups. Any who, the stats he provided in an interview on 710 were pretty cool. No Seahawk team since 2001 has come back and won from a 21 point deficit. Okay, how about this one! Teams with a 95% chance of losing from the 2 qtr. of a game (Stats provided by HUH?? Help?) are 3-19 this year with RW winning 2 out of 2 ( my guess is Houstan and the Bucs) and Luck winning 1 out of 1 ( I think against us or Houstan, probable the later). The rest of the teams that faced that obstacle all had losses. Sorry about not having Sando's source, just lazy tonight I guess.

Uhm . . . what? Sounds like a cool stat (or maybe two) is in there, but I can't make out what you're saying.

Both Houston and Tampa Bay had a 95% chance at winning at halftime. Teams with 95% chance of winning are 19-3 this year. Both Houston and Tampa Bay lost, so we count for two of those losses. Luck was most likely the other loss, against either us or Houston (most likely Houston).

 

[sc85sis]

Short answer: Russell Wilson and this team are pretty special. And Luck ain't too bad either, in spite of those here who are sick of hearing about him.

 

[telerion]

I sure miss his write ups. Any who, the stats he provided in an interview on 710 were pretty cool. No Seahawk team since 2001 has come back and won from a 21 point deficit. Okay, how about this one! Teams with a 95% chance of losing from the 2 qtr. of a game (Stats provided by HUH?? Help?) are 3-19 this year with RW winning 2 out of 2 ( my guess is Houstan and the Bucs) and Luck winning 1 out of 1 ( I think against us or Houstan, probable the later). The rest of the teams that faced that obstacle all had losses. Sorry about not having Sando's source, just lazy tonight I guess.

Uhm . . . what? Sounds like a cool stat (or maybe two) is in there, but I can't make out what you're saying.

Both Houston and Tampa Bay had a 95% chance at winning at halftime. Teams with 95% chance of winning are 19-3 this year. Both Houston and Tampa Bay lost, so we count for two of those losses. Luck was most likely the other loss, against either us or Houston (most likely Houston).

I see. So luck and Wilson appear to be good at comebacks after halftime (although at 1 and 2 occurrences, respectively, there's no way that's statistically significant).

 

[MizzouHawkGal]

Short answer: Russell Wilson and this team are pretty special. And Luck ain't too bad either, in spite of those here who are sick of hearing about him.

Uninfortunately, before Wayne's injury I thought they were a lock for a February date in New York. Against us of course! Still possible but not totally likely.

 

[MizzouHawkGal]

Short answer: Russell Wilson and this team are pretty special. And Luck ain't too bad either, in spite of those here who are sick of hearing about him.

Uninfortunately, before Wayne's injury I thought they were a lock for a February date in New York. Against us of course!

 

[ClumsyLurk]

Sando link: http://mynorthwest.com/category/pod_pla ... nd%20Danny

Seeker's concerns duely noted, and he calls Djangruss "Assasin like"

Allegedly, in games where the probability of losing is 95% or greater - Russel is 2-0 (Im not sure where the Atlanta game ranked if this is so)

 

[themunn]

The reason that there's only 95% chance of losing is because there's a 5% chance that Russell Wilson is the QB

Really it's a 100% chance of losing unless you are Russell Wilson

 

[kidhawk]

I sure miss his write ups. Any who, the stats he provided in an interview on 710 were pretty cool. No Seahawk team since 2001 has come back and won from a 21 point deficit. Okay, how about this one! Teams with a 95% chance of losing from the 2 qtr. of a game (Stats provided by HUH?? Help?) are 3-19 this year with RW winning 2 out of 2 ( my guess is Houstan and the Bucs) and Luck winning 1 out of 1 ( I think against us or Houstan, probable the later). The rest of the teams that faced that obstacle all had losses. Sorry about not having Sando's source, just lazy tonight I guess.

Uhm . . . what? Sounds like a cool stat (or maybe two) is in there, but I can't make out what you're saying.

Both Houston and Tampa Bay had a 95% chance at winning at halftime. Teams with 95% chance of winning are 19-3 this year. Both Houston and Tampa Bay lost, so we count for two of those losses. Luck was most likely the other loss, against either us or Houston (most likely Houston).

I see. So luck and Wilson appear to be good at comebacks after halftime (although at 1 and 2 occurrences, respectively, there's no way that's statistically significant).

It's not just a few occurrences, it's over the course of 19 games in similar situations. that is not a bad base. Of course we don't know who the other teams were or their opponents, but to be 2-2 in a situation where other qbs are a combined 0-16 is a pretty solid indicator

 

[telerion]

It's not just a few occurrences, it's over the course of 19 games in similar situations. that is not a bad base. Of course we don't know who the other teams were or their opponents, but to be 2-2 in a situation where other qbs are a combined 0-16 is a pretty solid indicator

You're correct that there have been quite a few games where a team had a 95% of losing after halftime (identifying that probability requires some strong assumptions but we can ignore them ). What hasn't been identified by the statistic above is that RW or AL are the ones who have led these comebacks because they possess above average skills which permit them to overcome big halftime deficits and not because they are the lucky few that have pulled it off this year.

Now most of us here probably believe that RW has these sort of skills (in part because we've seen his great comebacks in other situations), but until we have more observations of RW in these so called "95% chance of losing at halftime" we really can't statistically identify that RW is better at these situations than other qb's. This is especially true if you want to identify characteristics distinct to RW from say cross-effects from particular players around him.

It's a nice stat. I just don't think it means much right now.

 

[ClumsyLurk]

It's not just a few occurrences, it's over the course of 19 games in similar situations. that is not a bad base. Of course we don't know who the other teams were or their opponents, but to be 2-2 in a situation where other qbs are a combined 0-16 is a pretty solid indicator

You're correct that there have been quite a few games where a team had a 95% of losing after halftime (identifying that probability requires some strong assumptions but we can ignore them ). What hasn't been identified by the statistic above is that RW or AL are the ones who have led these comebacks because they possess above average skills which permit them to overcome big halftime deficits and not because they are the lucky few that have pulled it off this year.

Now most of us here probably believe that RW has these sort of skills (in part because we've seen his great comebacks in other situations), but until we have more observations of RW in these so called "95% chance of losing at halftime" we really can't statistically identify that RW is better at these situations than other qb's. This is especially true if you want to identify characteristics distinct to RW from say cross-effects from particular players around him.

It's a nice stat. I just don't think it means much right now.

How much more and what do you need to see before you're willing to say that Russ has the "it" factor - the "clutch" factor. Becuase I believe it is obvious that he does and it's already been shown.

He did it in college as well.

 

[telerion]

It's not just a few occurrences, it's over the course of 19 games in similar situations. that is not a bad base. Of course we don't know who the other teams were or their opponents, but to be 2-2 in a situation where other qbs are a combined 0-16 is a pretty solid indicator

You're correct that there have been quite a few games where a team had a 95% of losing after halftime (identifying that probability requires some strong assumptions but we can ignore them ). What hasn't been identified by the statistic above is that RW or AL are the ones who have led these comebacks because they possess above average skills which permit them to overcome big halftime deficits and not because they are the lucky few that have pulled it off this year.

Now most of us here probably believe that RW has these sort of skills (in part because we've seen his great comebacks in other situations), but until we have more observations of RW in these so called "95% chance of losing at halftime" we really can't statistically identify that RW is better at these situations than other qb's. This is especially true if you want to identify characteristics distinct to RW from say cross-effects from particular players around him.

It's a nice stat. I just don't think it means much right now.

How much more and what do you need to see before you're willing to say that Russ has the "it" factor - the "clutch" factor. Becuase I believe it is obvious that he does and it's already been shown.

He did it in college as well.

I suspect he does have "the clutch factor" based on his winning in other situations (including cases with less than a 95% of losing). In addition we see him displaying the skills that we think are part of being clutch (e.g., self-control under stress, courage in the face of a defensive pressure, testimony of his leadership skills by teammates).

All I'm saying is that taking this one stat in isolation presents a cool example but doesn't add any more strength (statistically speaking) to the claim that he is clutch.

It would be like if RW and 19 of us each wrote our name on piece of paper and plopped them in a hat. Then some one draws a name from the hat. If Russ wins, we don't say "wow Russ must be better than us at games of chance!" We say "congrats, Russ, you got lucky."
If we repeat and he does it again then we probably still chalk it up to luck. If he does it 7 out of the next 10 times, we might want to interrogate the person drawing the names because RW probably has an unfair advantage over us.

So in summary. Russ has led the team to two very unlikely comebacks already this season. That's awesome. I suspect he's clutch and over his career we'll may see enough of these to say it with some mathematical precision.

 

[SalishHawkFan]

The reason that there's only 95% chance of losing is because there's a 5% chance that Russell Wilson is the QB

Really it's a 100% chance of losing unless you are Russell Wilson

Actually, there's only a 3.2% chance that Wilson is the QB (1 in 31). The other 1.8% chance is the 12th man.

 

[SalishHawkFan]

It's not just a few occurrences, it's over the course of 19 games in similar situations. that is not a bad base. Of course we don't know who the other teams were or their opponents, but to be 2-2 in a situation where other qbs are a combined 0-16 is a pretty solid indicator

You're correct that there have been quite a few games where a team had a 95% of losing after halftime (identifying that probability requires some strong assumptions but we can ignore them ). What hasn't been identified by the statistic above is that RW or AL are the ones who have led these comebacks because they possess above average skills which permit them to overcome big halftime deficits and not because they are the lucky few that have pulled it off this year.

Now most of us here probably believe that RW has these sort of skills (in part because we've seen his great comebacks in other situations), but until we have more observations of RW in these so called "95% chance of losing at halftime" we really can't statistically identify that RW is better at these situations than other qb's. This is especially true if you want to identify characteristics distinct to RW from say cross-effects from particular players around him.

It's a nice stat. I just don't think it means much right now.

How much more and what do you need to see before you're willing to say that Russ has the "it" factor - the "clutch" factor. Becuase I believe it is obvious that he does and it's already been shown.

He did it in college as well.

I suspect he does have "the clutch factor" based on his winning in other situations (including cases with less than a 95% of losing). In addition we see him displaying the skills that we think are part of being clutch (e.g., self-control under stress, courage in the face of a defensive pressure, testimony of his leadership skills by teammates).

All I'm saying is that taking this one stat in isolation presents a cool example but doesn't add any more strength (statistically speaking) to the claim that he is clutch.

It would be like if RW and 19 of us each wrote our name on piece of paper and plopped them in a hat. Then some one draws a name from the hat. If Russ wins, we don't say "wow Russ must be better than us at games of chance!" We say "congrats, Russ, you got lucky."
If we repeat and he does it again then we probably still chalk it up to luck. If he does it 7 out of the next 10 times, we might want to interrogate the person drawing the names because RW probably has an unfair advantage over us.

So in summary. Russ has led the team to two very unlikely comebacks already this season. That's awesome. I suspect he's clutch and over his career we'll may see enough of these to say it with some mathematical precision.

You're wrong. It adds an immense amount of strength. You don't take those two instances in a vacuum. You look at the other comeback clutch wins. Vs Chicago, Green Bay, New England, Washington, even Atlanta last year. You look at the 9 of 27 starts where he's led the team from behind. Then you look at his TWO chances to overcome the 95% odds against him at halftime and see he's won BOTH of those times. Now you factor in that only 3 of 19 attempts succeeded. Statistically, 1 of 20 should succeed. Variance allows for 3 of 19, of course, but for Wilson to be the person who did it TWICE?

No, you're wrong, that's a very strong indicator right there.

 

[telerion]

You're wrong. It adds an immense amount of strength. You don't take those two instances in a vacuum. You look at the other comeback clutch wins. Vs Chicago, Green Bay, New England, Washington, even Atlanta last year. You look at the 9 of 27 starts where he's led the team from behind. Then you look at his TWO chances to overcome the 95% odds against him at halftime and see he's won BOTH of those times. Now you factor in that only 3 of 19 attempts succeeded. Statistically, 1 of 20 should succeed. Variance allows for 3 of 19, of course, but for Wilson to be the person who did it TWICE?

No, you're wrong, that's a very strong indicator right there.

By your logic winning the Powerball lotto adds "a lot of strength" to the claim that the winner is more skilled at lotto playing.

To recap:
1) I agree with you that Russ is clutch
2) I also agree tha his games against Houston and Tampa Bay included with his other comeback wins (pro as well as college) offer some decent evidence of this.

Where I disagree:
3) Russell overcoming 95% odds of losing after halftime twice is is a "solid indicator" that he is clutch. To say that you can't take these stats in isolation and then claim that this is a "solid indicator" is a contradiction.

4) I am wrong.

Look, I'm acting on the assumption that the errors surrounding this 95% chance calculation approach normal in the limit with some modest variance. This is because I suspect that this figure came either by bootstrapping over some large subset of NFL game history or running Monte Carlo simulations with it. Typically when this is done the errors approach a normal distribution. Unless the variance is extremely small (and I doubt that) then it is not going to be statistically significant to do this twice.

Now if you can show me that the variance is extremely small or that there is a crazy amount of skewness so that the error distribution has very little mass in the right tail then please do so.

 

[Popeyejones]

I sure miss his write ups.

Me too.

How is the Hawks' new ESPN person?

Bill Williamson got assigned to the 9ers, and he really sucks.

 

[telerion]

I sure miss his write ups.

Me too.

How is the Hawks' new ESPN person?

Bill Williamson got assigned to the 9ers, and he really sucks.

I think it's Terry Blount, and I haven't been too impressed so far, but I'll give him time.

 

[kidhawk]

It's not just a few occurrences, it's over the course of 19 games in similar situations. that is not a bad base. Of course we don't know who the other teams were or their opponents, but to be 2-2 in a situation where other qbs are a combined 0-16 is a pretty solid indicator

You're correct that there have been quite a few games where a team had a 95% of losing after halftime (identifying that probability requires some strong assumptions but we can ignore them ). What hasn't been identified by the statistic above is that RW or AL are the ones who have led these comebacks because they possess above average skills which permit them to overcome big halftime deficits and not because they are the lucky few that have pulled it off this year.

Now most of us here probably believe that RW has these sort of skills (in part because we've seen his great comebacks in other situations), but until we have more observations of RW in these so called "95% chance of losing at halftime" we really can't statistically identify that RW is better at these situations than other qb's. This is especially true if you want to identify characteristics distinct to RW from say cross-effects from particular players around him.

It's a nice stat. I just don't think it means much right now.

The only SINGLE stat that means anything is W/L, the rest are meant to be compiled and analyzed as a group. When you look at this particular stat, you have to take into account historical data for the NFL and these quarterbacks in particular. Both Luck and Wilson have a high number of 4th quarter comebacks (Luck is a little higher and I believe has more than any other 2nd year quarterback in history), so when you add in that one extra stat it starts to form a pattern. There are so many stats out there that you really can't just ever take any single stat and use it as an entire basis for anything, but they are good indicators, especially when you can back them up with other statistical data.

 

[telerion]

It's not just a few occurrences, it's over the course of 19 games in similar situations. that is not a bad base. Of course we don't know who the other teams were or their opponents, but to be 2-2 in a situation where other qbs are a combined 0-16 is a pretty solid indicator

You're correct that there have been quite a few games where a team had a 95% of losing after halftime (identifying that probability requires some strong assumptions but we can ignore them ). What hasn't been identified by the statistic above is that RW or AL are the ones who have led these comebacks because they possess above average skills which permit them to overcome big halftime deficits and not because they are the lucky few that have pulled it off this year.

Now most of us here probably believe that RW has these sort of skills (in part because we've seen his great comebacks in other situations), but until we have more observations of RW in these so called "95% chance of losing at halftime" we really can't statistically identify that RW is better at these situations than other qb's. This is especially true if you want to identify characteristics distinct to RW from say cross-effects from particular players around him.

It's a nice stat. I just don't think it means much right now.

The only SINGLE stat that means anything is W/L, the rest are meant to be compiled and analyzed as a group. When you look at this particular stat, you have to take into account historical data for the NFL and these quarterbacks in particular. Both Luck and Wilson have a high number of 4th quarter comebacks (Luck is a little higher and I believe has more than any other 2nd year quarterback in history), so when you add in that one extra stat it starts to form a pattern. There are so many stats out there that you really can't just ever take any single stat and use it as an entire basis for anything, but they are good indicators, especially when you can back them up with other statistical data.

I think then that we basically agree.

Maybe in the case of kickers with fairly long careers you can get some significant statistics since they're taking 25-40 FG attempts per season for 7+ seasons and there isn't much meaningful variation in the attempts besides yardage, wind, and stress (probably a few others). But compared to a QB where there are 21 other active players on nearly every play, estimating "clutchness" of a FG kicker should be a "chip shot."

Sounds like a cool academic paper.