What's new
Fantasy Football - Footballguys Forums

Welcome to Our Forums. Once you've registered and logged in, you're primed to talk football, among other topics, with the sharpest and most experienced fantasy players on the internet.

Who is the clutchest QB in the NFL since 1996? (1 Viewer)

SSOG

SSOG

Moderator
http://www.footballoutsiders.com/2006/06/2...-analysis/3978/

This article counted all the times a QB was down by 8 or fewer points in the 4th quarter and had the ball, tracked that QB's won/loss percentage, and then ran the numbers to see who, statistically, was the best 4th-quarter comeback QB in the league since 1996.

As you could probably guess, Brady ranks pretty high (4th overall, actually). As for who beat him out... well, let's just say that the answer is going to surprise everyone who ever said that Denver/Indianapolis will never win a Superbowl with Jake Plummer/Peyton Manning at QB.

As an added bonus, it also runs a similar test to see who the best coaches at holding a 4th quarter lead are.

:popcorn:

 
The percentage of win/loss to me is more important...

That 13-5 is much more impressive to me than 19-29 held by Plummer/Manning.

However, I am very surprised to see Bulger /Martz up there.... I suppose I shouldn't be, but I am.

Anyone find that stat somewhat alarming?

 
Any QB rating criteria that makes Kerry Collins and Vinny Testaverde comparable with Peyton Manning is highly credible.

 
Last edited by a moderator:
Snake at the top doesn't surprise me at all. The guy is downright brilliant at times, but at others he is attrocious. Inconsistent as they come.

 
Any QB rating criteria that makes Kerry Collins and Vinny Testaverde comparable with Peyton Manning is highly credible.
Click to expand...
Manning,Peyton 33189 yardsCollins,Kerry 33635 yards

Testaverde,Vinny 45252 yards

:thumbup:

 
Yeah but Brady is 13-8 while Peyton is 19-29. So Brady IS the QB you want to have when you're trailing in the 4th quarter.

 
Yeah but Brady is 13-8 while Peyton is 19-29. So Brady IS the QB you want to have when you're trailing in the 4th quarter.
Click to expand...
<======= Why so you can pick him off for 101 yards??? :D
 
Yeah but Brady is 13-8 while Peyton is 19-29. So Brady IS the QB you want to have when you're trailing in the 4th quarter.
Click to expand...
Well, that depends. Really, what the t-test does is it says that it's more likely that Peyton Manning is actually an above-average comeback QB than Brady.Think of it this way. Say you flip one coin twice and get heads both times. Now you flip a second coin a billion times and get heads 52% of the time. Which coin is more likely to be weighted? That's what this T-Test is saying- it's more likely that Brady is really an average late-game QB and his heroics were a fluke than that Manning is really an average late-game QB and his heroics were a fluke.

I don't understand why Brady ranks 4th. I read the article, but it still doesn't make much sense to me.
Click to expand...
Read the comments. I don't actually agree with the article's methodology, I was just posting it to play Devil's Advocate. With that said, there's some really good number crunching in the comments field, as well. It gives a couple of other ways to look at the numbers (most of which do, in fact, wind up ranking Brady ahead of Manning and Plummer again).
 
Also, everyone who is throwing out the study because of the rankings of Kitna, Collins, and Testeverde is a little bit off base. Look at their numbers. They clearly are good 4th-quarter QBs, regardless of what you think of them the rest of the time.

 
Yeah but Brady is 13-8 while Peyton is 19-29. So Brady IS the QB you want to have when you're trailing in the 4th quarter.
Click to expand...
Well, that depends. Really, what the t-test does is it says that it's more likely that Peyton Manning is actually an above-average comeback QB than Brady.Think of it this way. Say you flip one coin twice and get heads both times. Now you flip a second coin a billion times and get heads 52% of the time. Which coin is more likely to be weighted? That's what this T-Test is saying- it's more likely that Brady is really an average late-game QB and his heroics were a fluke than that Manning is really an average late-game QB and his heroics were a fluke.

I don't understand why Brady ranks 4th. I read the article, but it still doesn't make much sense to me.
Click to expand...
Read the comments. I don't actually agree with the article's methodology, I was just posting it to play Devil's Advocate. With that said, there's some really good number crunching in the comments field, as well. It gives a couple of other ways to look at the numbers (most of which do, in fact, wind up ranking Brady ahead of Manning and Plummer again).
Click to expand...
I didn't read the comments, but how you described the T-Test above is how I interpreted it. It just seems that 13 out of 21 is harder to do than 19 out of 48. Brady would need to go 5 out of 27 to do worse than Manning. The second worst QB in the study went 5 out of 28. Therefore, 5 out of 28 is really horrible. So how can Brady be behind Manning?
 
One of the funny things about this information is regarding Plummer himself, at least from his Cardinals days. One of the main reason why he was scrambling like mad to come back in the 4th quarter was that his team was trailing due to his boneheaded mistakes that resulted in INT's. I fail to see how that makes him a good QB.

 
One of the funny things about this information is regarding Plummer himself, at least from his Cardinals days. One of the main reason why he was scrambling like mad to come back in the 4th quarter was that his team was trailing due to his boneheaded mistakes that resulted in INT's. I fail to see how that makes him a good QB.
Click to expand...
My thoughts exactly. Most of those guys( esp Plummer) give it up early causing the so called comeback.
 
Based on t-tests, 21 quarterbacks have been significantly “good” at fourth-quarter comebacks since 1996 and 18 quarterbacks have been significantly “not good” by the same test (all at the .95 level). The top 10 and bottom 10 are listed above in Table 1 and Table 2. Here are the others:Significantly good: Aaron Brooks, Tim Couch, Trent Green, Kent Graham, Steve Young, Ben Roethlisberger, John Elway, Kordell Stewart, Daunte Culpepper, Elvis Grbac, Rich Gannon.Significantly bad: Neil O’Donnell, Doug Flutie, Frank Reich (yes, the same Frank Reich who led the greatest postseason comeback in NFL history in
Click to expand...
Man, I don't know what it is, but I've always hated Neil O'donnell
 
Last edited by a moderator:
One of the funny things about this information is regarding Plummer himself, at least from his Cardinals days.  One of the main reason why he was scrambling like mad to come back in the 4th quarter was that his team was trailing due to his boneheaded mistakes that resulted in INT's.  I fail to see how that makes him a good QB.
Click to expand...
My thoughts exactly. Most of those guys( esp Plummer) give it up early causing the so called comeback.
Click to expand...
:lmao: that's so dead on! :lmao:
 
http://www.footballoutsiders.com/2006/06/2...-analysis/3978/

This article counted all the times a QB was down by 8 or fewer points in the 4th quarter and had the ball, tracked that QB's won/loss percentage, and then ran the numbers to see who, statistically, was the best 4th-quarter comeback QB in the league since 1996.

As you could probably guess, Brady ranks pretty high (4th overall, actually). As for who beat him out... well, let's just say that the answer is going to surprise everyone who ever said that Denver/Indianapolis will never win a Superbowl with Jake Plummer/Peyton Manning at QB.

As an added bonus, it also runs a similar test to see who the best coaches at holding a 4th quarter lead are.

:popcorn:
Click to expand...
I hope footballoutsiders pays you well
 
One of the funny things about this information is regarding Plummer himself, at least from his Cardinals days.  One of the main reason why he was scrambling like mad to come back in the 4th quarter was that his team was trailing due to his boneheaded mistakes that resulted in INT's.  I fail to see how that makes him a good QB.
Click to expand...
My thoughts exactly. Most of those guys( esp Plummer) give it up early causing the so called comeback.
Click to expand...
:lmao: that's so dead on! :lmao:
Click to expand...
Since he's been in Denver and hasn't been down by 48 points at half-time like he was in Arizona, Plummer has had EXCELLENT TD/INT ratios all three years.
 
That's the dumbest way of defining "clutchness" ever :wall:

How about we do it by winning PERCENTAGE in the 4th quarter, instead of the number of wins? Brady's winning percentage in the 4th quarter >>>>>>> Jake Plummer/Peyton Manning. I'm not even suprised that Bulger is 10-5, he's a great clutch QB imo as well. That list is downright awful.

I'll take Bradys almost 2:1 win percentage and Bulgers actual 2:1 over any of those QBs any day.

 
One of the funny things about this information is regarding Plummer himself, at least from his Cardinals days. One of the main reason why he was scrambling like mad to come back in the 4th quarter was that his team was trailing due to his boneheaded mistakes that resulted in INT's. I fail to see how that makes him a good QB.
Click to expand...
The study never claimed he was a good QB, just that if you were down by 8 or less in the 4th quarter he was one of the most likely guys in the NFL to come up with the win.He is a very good QB, though. The reason he made those boneheaded mistakes was because he knew his team was so horrible they were going to lose if he didn't do something heroic.

http://www.footballoutsiders.com/2006/06/2...-analysis/3978/

This article counted all the times a QB was down by 8 or fewer points in the 4th quarter and had the ball, tracked that QB's won/loss percentage, and then ran the numbers to see who, statistically, was the best 4th-quarter comeback QB in the league since 1996.

As you could probably guess, Brady ranks pretty high (4th overall, actually). As for who beat him out... well, let's just say that the answer is going to surprise everyone who ever said that Denver/Indianapolis will never win a Superbowl with Jake Plummer/Peyton Manning at QB.

As an added bonus, it also runs a similar test to see who the best coaches at holding a 4th quarter lead are.

:popcorn:
Click to expand...
I hope footballoutsiders pays you well
Click to expand...
Why, because I posted an interesting article from there to spark a discussion, or because I openly disagree with their methodology?
That's the dumbest way of defining "clutchness" ever :wall:

How about we do it by winning PERCENTAGE in the 4th quarter, instead of the number of wins? Brady's winning percentage in the 4th quarter >>>>>>> Jake Plummer/Peyton Manning. I'm not even suprised that Bulger is 10-5, he's a great clutch QB imo as well. That list is downright awful.

I'll take Bradys almost 2:1 win percentage and Bulgers actual 2:1 over any of those QBs any day.
Click to expand...
Here's the problem with winning percentage. Let's say QBA is 1-0 when trailing by 8 in the fourth quarter, and QBB is 19-1 when trailing by 8 in the fourth quarter. Who would you rather have, QBA who has a higher winning percentage, or QBB who has done it often enough to prove it wasn't just a fluke?That's the point of the T-Test. It takes sample size into consideration in order to determine what the odds are that it was just a fluke.

I agree that there are better ways to do it, but I certainly don't think that straight winning percentage is it.

 
That's the dumbest way of defining "clutchness" ever :wall:

How about we do it by winning PERCENTAGE in the 4th quarter, instead of the number of wins? Brady's winning percentage in the 4th quarter >>>>>>> Jake Plummer/Peyton Manning. I'm not even suprised that Bulger is 10-5, he's a great clutch QB imo as well. That list is downright awful.

I'll take Bradys almost 2:1 win percentage and Bulgers actual 2:1 over any of those QBs any day.
Click to expand...
:goodposting:
 
Bottom Line it's Brady for anyone who has really been paying attention in last 5 years. It's obvious enough that anyone setting up a "studies" parameters (such as making it Ten years and more about total comebacks than % of time you did it) that changes it from Brady needs to be questioned about said methods (and why they chose them) of determining who it is. Clutch also means making the big plays in Tied games or games you have a slight lead in. Not giving the other team extra chances is important too. Or in a Tied game having the sack to mount a game winning drive despite what the World and Madden thinks about you just playing it "safe" running out the clock and playing overtime.

 
Or in a Tied game having the sack to mount a game winning drive despite what the World and Madden thinks about you just playing it "safe" running out the clock and playing overtime.
Click to expand...
Does it really take a sack to go 5/7 for 53 yards and 0 TDs? I think I see that just about every week of the NFL season. I really doubt that three screen passes to J.R. Redmond is what makes a QB great.
 
Executing under pressure is. And going 5 for 7 in that situation in that important of a game is something many QBs would fail at ....just the time management messes up some QBs/Teams. Squeezing a winning drive in with cool headed plays and excellent clock management is very clutch. Joe Montana keeping his head under pressure is why many think he was so clutch. It's about making plays AND not screwing things up in a tight game.

And the 0 TDs on that drive is meaniless when a TD isn't necessary...a clutch QB is aware of the situation and what is needed. Doesn't force it until he has no other options. If a FG will do it that's good enough.

 
Last edited by a moderator:
Or in a Tied game having the sack to mount a game winning drive despite what the World and Madden thinks about you just playing it "safe" running out the clock and playing overtime.
Click to expand...
Does it really take a sack to go 5/7 for 53 yards and 0 TDs? I think I see that just about every week of the NFL season. I really doubt that three screen passes to J.R. Redmond is what makes a QB great.
Click to expand...
So you'd think a QB who forced it downfield to a WR who was covered would be better? That's why many aren't clutch..they panic thinking they need to do it all too fast. Brett Favres quote on "just picking up first downs" comes to mind...and he's a guy who loves going downfield (and often forces an Interception doing so) but credits his 2 minute successes to when he just tries to get first downs. The QB that takes what's there and moves down the field at the pace required is the guy I want on the comeback last minute drive.
 
Executing under pressure is. And going 5 for 7 in that situation in that important of a game is something many QBs would fail at ....just the time management messes up some QBs/Teams. Squeezing a winning drive in with cool headed plays and excellent clock management is very clutch. Joe Montana keeping his head under pressure is why many think he was so clutch. It's about making plays AND not screwing things up in a tight game.

And the 0 TDs on that drive is meaniless when a TD isn't necessary...a clutch QB is aware of the situation and what is needed. Doesn't force it until he has no other options. If a FG will do it that's good enough.
Click to expand...
You seem fixated on this clutch thing. Can you explain it a bit to me? Apparently you think Tom Brady is clutch, so let's use him as an example.Saying Brady is clutch, does that mean he can't fail? If we ran that final drive over 100 times, would he lead the Pats to 100 field goals (or better)?

Is NeedABigDriveBrady just X% better than RegularOldBrady? Does X = 10? 20? 50?

Is Brady's clutch factor consistent? If NeedABigDriveBrady is 30% better than RegularOldBrady, is that going to hold true in five years? In 10 years?

What makes a QB clutch? Is it how good the NeedABigDrivePlayerX is, or is it how much better NeedABigDrivePlayerX is than RegularOldPlayerX?

I've got a few follow up questions as well. Thanks.

 
"Saying Brady is clutch, does that mean he can't fail? "

Obviously you're lacking something. Clutch means good under pressure to me and I'll guess majority. Dictionary has "a crucial moment in a critical situation" Good in a Crucial Moments in a critical situation doesn't mean Guaranteed.

Jumping to absolutes? It would take too long to educate you from the starting point you are beginning with. I'll pass on that. Checked on that 1999 Tennessee run defense yet?

 
Last edited by a moderator:
"Saying Brady is clutch, does that mean he can't fail? "

Obviously you're lacking something. Clutch means good under pressure to me and I'll guess majority. Dictionary has "a crucial moment in a critical situation" Good in a Crucial Moments in a critical situation doesn't mean Guaranteed.

Jumping to absolutes? It would take too long to educate you from the starting point you are beginning with. I'll pass on that. Checked on that 1999 Tennessee run defense yet?
Click to expand...
Thanks for dodging all my questions. Here I was hoping you could explain what you mean by clutch.I have no idea where you're going with this 1999 Tennessee run defense, since that comes out of left field. That year the Titans D stunk, and the run D was only marginally better than the horrific pass D.

Pass D ranked 27th in YPA allowed and 27th in Passing TDs allowed.

Rush D ranked 22nd in YPC allowed.

Total D ranked 18th in total yards allowed.

Tenn ranked 15th in points allowed.

The Titans D wasn't very good in 1999, wherever you're going with this.

 
Yeah but Brady is 13-8 while Peyton is 19-29. So Brady IS the QB you want to have when you're trailing in the 4th quarter.
Click to expand...
Well, that depends. Really, what the t-test does is it says that it's more likely that Peyton Manning is actually an above-average comeback QB than Brady.Think of it this way. Say you flip one coin twice and get heads both times. Now you flip a second coin a billion times and get heads 52% of the time. Which coin is more likely to be weighted? That's what this T-Test is saying- it's more likely that Brady is really an average late-game QB and his heroics were a fluke than that Manning is really an average late-game QB and his heroics were a fluke.

I don't understand why Brady ranks 4th. I read the article, but it still doesn't make much sense to me.
Click to expand...
Read the comments. I don't actually agree with the article's methodology, I was just posting it to play Devil's Advocate. With that said, there's some really good number crunching in the comments field, as well. It gives a couple of other ways to look at the numbers (most of which do, in fact, wind up ranking Brady ahead of Manning and Plummer again).
Click to expand...
I didn't read the comments, but how you described the T-Test above is how I interpreted it. It just seems that 13 out of 21 is harder to do than 19 out of 48. Brady would need to go 5 out of 27 to do worse than Manning. The second worst QB in the study went 5 out of 28. Therefore, 5 out of 28 is really horrible. So how can Brady be behind Manning?
Click to expand...
I don't like the way they did this, but it is better than just using straight win percentages.The problem with their rankings is that they are only ranking the players by how likely it is that they are above average. Manning can be more likely than Brady to be above average while being less likely than Brady to be in the top five percent. Which guy is more likely to be merely above average does not tell us which guy is more likely to be better than the other one.

They should have used Bayesian inference analysis instead of t-tests.

 
By the way, where it says "Full Results" in green, is that supposed to be a link to the full results? Because it's not click-able as far as I can tell. (I mean, you can click on it, but nothing happens.)

 
I don't like the way they did this, but it is better than just using straight win percentages.

The problem with their rankings is that they are only ranking the players by how likely it is that they are above average. Manning can be more likely than Brady to be above average while being less likely than Brady to be in the top five percent. Which guy is more likely to be merely above average does not tell us which guy is more likely to be better than the other one.

They should have used Bayesian inference analysis instead of t-tests.
Click to expand...
Thanks. That makes a lot of sense. I was really confused as to why Manning ranked ahead of Brady.
 
Yards not Yards per Carry. Tennessee made Superbowl in part because teams could not consistently run on them and they played in a Running Division. They were not Soft as you stated in a previous argumentative post.

I believe this type of more questions later stuff is just boring argumentative stuff again. As for dodging...I don't feel like answering a laundry list with a promise from you of another laundry list to come (when I think we all know what Clutch is..and if someone doesn't that's more posting than I'm willing to explain to them beyond a short definition/reply). I answered what I thought Clutch meant....and I have no unusual fixation with it beyond that it is the Title of the Thread. Good under pressure or Good in Crucial or critical situations means just that. Not Guaranteed...not assigning % to it. ..just Good. In Bradys case I 'd say even a step beyond Good. A Clutch player = Good under pressure. Great Clutch player Better than Good Clutch Player. Choker means opposite. I don't believe many fans don't already know what Clutch and Choker means.

Do I get a list of questions about definitions of Choker now? Or do you realize I don't care to try to slap a % or any guarantees to a term that is not a statistical term. It's a relative term just like Clutch is.

Great Clutch > Good Clutch> Normal> Choker> Big Time Choker

No % with these terms they are relative to each other and in some ways always going to be subjective when broken down too far. They have a place in describing how players react to "crucial and or Critical situations" in a general way. But when it gets down to someone like Brady who has shown a pattern by now his classification is pretty much revealed. He handles pressure/crucial way better than a normal QB..he'll still fail occasionally...but not nearly as often as some ranked higher on that list.

 
Yeah but Brady is 13-8 while Peyton is 19-29. So Brady IS the QB you want to have when you're trailing in the 4th quarter.
Click to expand...
Well, that depends. Really, what the t-test does is it says that it's more likely that Peyton Manning is actually an above-average comeback QB than Brady.Think of it this way. Say you flip one coin twice and get heads both times. Now you flip a second coin a billion times and get heads 52% of the time. Which coin is more likely to be weighted? That's what this T-Test is saying- it's more likely that Brady is really an average late-game QB and his heroics were a fluke than that Manning is really an average late-game QB and his heroics were a fluke.

I don't understand why Brady ranks 4th. I read the article, but it still doesn't make much sense to me.
Click to expand...
Read the comments. I don't actually agree with the article's methodology, I was just posting it to play Devil's Advocate. With that said, there's some really good number crunching in the comments field, as well. It gives a couple of other ways to look at the numbers (most of which do, in fact, wind up ranking Brady ahead of Manning and Plummer again).
Click to expand...
I didn't read the comments, but how you described the T-Test above is how I interpreted it. It just seems that 13 out of 21 is harder to do than 19 out of 48. Brady would need to go 5 out of 27 to do worse than Manning. The second worst QB in the study went 5 out of 28. Therefore, 5 out of 28 is really horrible. So how can Brady be behind Manning?
Click to expand...
I don't like the way they did this, but it is better than just using straight win percentages.The problem with their rankings is that they are only ranking the players by how likely it is that they are above average. Manning can be more likely than Brady to be above average while being less likely than Brady to be in the top five percent. Which guy is more likely to be merely above average does not tell us which guy is more likely to be better than the other one.

They should have used Bayesian inference analysis instead of t-tests.
Click to expand...
Thanks. That's a good explanation why this ranking is not right at the edges of it. Top and Bottom would be skewed while closer you get to the average it becomes more accurate. Is that what you're saying?
 
Last edited by a moderator:
Yards not Yards per Carry. Tennessee made Superbowl in part because teams could not consistently run on them and they played in a Running Division. They were not Soft as you stated in a previous argumentative post.
Click to expand...
The Tennessee Titans were 13-3. They ran a ball controlled offense. As a result of those two things, opponents didn't rush much against them. When they did, they did pretty well:James Stewart: 25/98/1

Marshall Faulk: 16/90/1

Charlie Garner: 12/69

Priest Holmes: 9/100

Jerome Bettis: 14/88

Napolean Kaufman and Tyrone Wheatley: 18/100

Richard Huntley: 11/78

Antowain Smith: 14/79

Fred Taylor: 19/110

I think you're confusing the 1999 Titans with the 2000 Titans. No shame in that. The 2000 Titans were very tough against the run, ranking in the top five in both rushing yards allowed and rushing yards per carry allowed.

Tennesee made the Super Bowl for lots of reasons, but their run defense wasn't one of them. They very easily could have been bounced in the first round if not for a miracle. By nearly any measure the 2000 Titans were a much better team, which as I stated before, was probably the source of the confusion.

... I think we all know what Clutch is..and if someone doesn't that's more posting than I'm willing to explain to them beyond a short definition/reply...
Click to expand...
Alright then LFW. Have a good one.
 
Yards not Yards per Carry. Tennessee made Superbowl in part because teams could not consistently run on them and they played in a Running Division. They were not Soft as you stated in a previous argumentative post.
Click to expand...
The Tennessee Titans were 13-3. They ran a ball controlled offense. As a result of those two things, opponents didn't rush much against them. When they did, they did pretty well:James Stewart: 25/98/1

Marshall Faulk: 16/90/1

Charlie Garner: 12/69

Priest Holmes: 9/100

Jerome Bettis: 14/88

Napolean Kaufman and Tyrone Wheatley: 18/100

Richard Huntley: 11/78

Antowain Smith: 14/79

Fred Taylor: 19/110

I think you're confusing the 1999 Titans with the 2000 Titans. No shame in that. The 2000 Titans were very tough against the run, ranking in the top five in both rushing yards allowed and rushing yards per carry allowed.

Tennesee made the Super Bowl for lots of reasons, but their run defense wasn't one of them. They very easily could have been bounced in the first round if not for a miracle. By nearly any measure the 2000 Titans were a much better team, which as I stated before, was probably the source of the confusion.

... I think we all know what Clutch is..and if someone doesn't that's more posting than I'm willing to explain to them beyond a short definition/reply...
Click to expand...
Alright then LFW. Have a good one.
Click to expand...
That List does make 1999 Tenn Run Defense look worse than I remembered. I had remembered them as using some of Buddy Ryans old Tatics for most of Fishers tenure and being solid vs Run until the last few years. I'd still not call them soft myself as no one was really ripping them apart as only Priest and Fred Taylor went over 100 and barely (not like soft run defenses that give up 150 to good RB and 100 to mediocre ones) but I can see why you think that given guys like Huntley going 11-78
 
http://www.footballoutsiders.com/2006/06/2...-analysis/3978/

This article counted all the times a QB was down by 8 or fewer points in the 4th quarter and had the ball, tracked that QB's won/loss percentage, and then ran the numbers to see who, statistically, was the best 4th-quarter comeback QB in the league since 1996.

As you could probably guess, Brady ranks pretty high (4th overall, actually). As for who beat him out... well, let's just say that the answer is going to surprise everyone who ever said that Denver/Indianapolis will never win a Superbowl with Jake Plummer/Peyton Manning at QB.

As an added bonus, it also runs a similar test to see who the best coaches at holding a 4th quarter lead are.

:popcorn:
Click to expand...
They don't include playoff games in this study either. I think when you try ranking the upper end of this then playoff games should definately be included. Especially if someone is using this as a playoff argument for Manning/Plummer.And with playoffs being a one loss and out format instead of best of seven. I think were not all that interested in average for a baseline on winning these types. I think we're more interested in who has shown the least failures in these types of situations. For instance if Peyton or Plummer is up against Brady or Big Ben in next years playoffs are we interested in who is more certain to be above average clutch/comeback capable QB/Team...or are we more interested in who we think is possibly the Top clutch QB/Team. I know I'd go with the High Side Guys in Big Ben and Brady (that we have less certainty on just what they might be) even if their sample size isn't what Plummers/Mannings are. Is Brady/Big Ben A Montana,or Johnny Unitas,Otto Graham or even Elway level(who always had knack for comebacks even when team wasn't good enough), ? we don't have certainty on that yet..but I'd feel more comfortable saying they have a shot at it still than guys who have failed at comebacks as often and with frequency of Testaverde,Plummer,Peyton. And in Playoffs we want Top end ..not mearly better than league average with more certainty.

It actually makes an argument for Peyton to beat average Teams/QBs in regular season more often than not (which he has). But you don't face the average team /average QB in playoffs as likely as you are to face an above average one. Side note- Take it for what you will...Two of Peytons 3 Playoff wins are vs Plummer. Peyton 1-6 not facing his buddy at Top of this list in playoffs.

 
Last edited by a moderator:
I don't like the way they did this, but it is better than just using straight win percentages.

The problem with their rankings is that they are only ranking the players by how likely it is that they are above average. Manning can be more likely than Brady to be above average while being less likely than Brady to be in the top five percent. Which guy is more likely to be merely above average does not tell us which guy is more likely to be better than the other one.

They should have used Bayesian inference analysis instead of t-tests.
Click to expand...
Thanks. That's a good explanation why this ranking is not right at the edges of it. Top and Bottom would be skewed while closer you get to the average it becomes more accurate. Is that what you're saying?
Click to expand...
Nope, that's not how it works. A t-test is designed to tell how likely it is that something is significantly above or below average.Allow me to explain again with coinflips. Let's say I flip a coin 5 times, and it comes up heads 100% of the time. Now let's say I flip a different coin a gajillion times, and it comes up heads 52% of the time. Now, I could run a t-test, and the second coin would score higher. The test is basically saying how likely it is that a coin is "weighted" (i.e. actually more likely to land on one side than the other). Now, even though Coin #2 only landed on Heads 52% of the time, we flipped it so much that we can use that data to determine that there is, in fact, a significantly better chance of landing on heads than tails. Meanwhile, with coin #1, the coin might be weighted... or it might have just been dumb luck. We don't know, because we only flipped it 5 times.

Now what Maurile is saying is that the t-test only tells us how likely it is that the coin is weighted- it doesn't really tell us how weighted the coin is. Coin #1 could have an 80% "true" chance to land on heads, which would be much better than coin #2's chance... the T-Test doesn't tell us that. It doesn't tell us which is MORE above average, it just tells us which is MORE LIKELY TO BE above average (to some degree or another).

I don't really know any other way to explain it, so I hope this cleared it up for you.

 
Bottom Line it's Brady for anyone who has really been paying attention in last 5 years. It's obvious enough that anyone setting up a "studies" parameters (such as making it Ten years and more about total comebacks than % of time you did it) that changes it from Brady needs to be questioned about said methods (and why they chose them) of determining who it is. Clutch also means making the big plays in Tied games or games you have a slight lead in. Not giving the other team extra chances is important too. Or in a Tied game having the sack to mount a game winning drive despite what the World and Madden thinks about you just playing it "safe" running out the clock and playing overtime.
Click to expand...
There's no hocus pocus. They didn't skew the study's parameters to try to get a specific result. They only have play-by-play data from the 1996 season to present, so they could only possibly run this study from 1996 to present. Also, you can question their motives all you want, but the site is owned and run by a diehard Patriots fan, and Patriots fans are also the largest posting demographic, as well.Anyway, the study wasn't to find out who was the "clutchest", that was just a word I used to describe it (since titles have to be reasonably succinct). The study was just supposed to tell who was the best at 4th-quarter comebacks. I agree that there would be a lot more parameters that would have to be examined if you wanted to find out who was the "clutchest", such as how many scores the defense allowed during that span, as well.

 
Bottom Line  it's Brady for anyone who has really been paying attention in last 5 years. It's obvious enough that anyone setting up a "studies" parameters (such as making it Ten years and more about total comebacks than % of time you did it) that changes it from Brady needs to be questioned about said methods (and why they chose them) of determining who it is. Clutch also means making the big plays in Tied games or games you have a slight lead in. Not giving the other team extra chances is important too. Or in a Tied game having the sack to mount a game winning drive despite what the World and Madden thinks about you just playing it "safe" running out the clock and playing overtime.
Click to expand...
There's no hocus pocus. They didn't skew the study's parameters to try to get a specific result. They only have play-by-play data from the 1996 season to present, so they could only possibly run this study from 1996 to present. Also, you can question their motives all you want, but the site is owned and run by a diehard Patriots fan, and Patriots fans are also the largest posting demographic, as well.Anyway, the study wasn't to find out who was the "clutchest", that was just a word I used to describe it (since titles have to be reasonably succinct). The study was just supposed to tell who was the best at 4th-quarter comebacks. I agree that there would be a lot more parameters that would have to be examined if you wanted to find out who was the "clutchest", such as how many scores the defense allowed during that span, as well.
Click to expand...
OK , but it doesn't really say who is the best at comebacks either. Just who is most certain to be above the league average. So not really about most Clutch or Best at Comebacks. I think with data available over long careers we all would say Peyton is certainly above the league average. And most would say Testaverde and Plummer are above average QBs as well....but Teste and Plummer extremely high on certainty of being above the leagues average QB doesn't install any confidence to me that they are a good QB to have in a playoff run nor should it imo. Or in a close game vs someone lower on the list like Brady,Bulger(if healthy) or Big Ben.
 
They don't include playoff games in this study either. I think when you try ranking the upper end of this then playoff games should definately be included. Especially if someone is using this as a playoff argument for Manning/Plummer.

And with playoffs being a one loss and out format instead of best of seven. I think were not all that interested in average for a baseline on winning these types. I think we're more interested in who has shown the least failures in these types of situations. For instance if Peyton or Plummer is up against Brady or Big Ben in next years playoffs are we interested in who is more certain to be above average clutch/comeback capable QB/Team...or are we more interested in who we think is possibly the Top clutch QB/Team. I know I'd go with the High Side Guys in Big Ben and Brady (that we have less certainty on just what they might be) even if their sample size isn't what Plummers/Mannings are. Is Brady/Big Ben A Montana,or Johnny Unitas,Otto Graham or even Elway level(who always had knack for comebacks even when team wasn't good enough), ? we don't have certainty on that yet..but I'd feel more comfortable saying they have a shot at it still than guys who have failed at comebacks as often and with frequency of Testaverde,Plummer,Peyton. And in Playoffs we want Top end ..not mearly better than league average with more certainty.

It actually makes an argument for Peyton to beat average Teams/QBs in regular season more often than not (which he has). But you don't face the average team /average QB in playoffs as likely as you are to face an above average one. Side note- Take it for what you will...Two of Peytons 3 Playoff wins are vs Plummer. Peyton 1-6 not facing his buddy at Top of this list in playoffs.
Click to expand...
They did include playoff games in the study.
OK , but it doesn't really say who is the best at comebacks either. Just who is most certain to be above the league average. So not really about most Clutch or Best at Comebacks. I think with data available over long careers we all would say Peyton is certainly above the league average. And most would say Testaverde and Plummer are above average QBs as well....but Teste and Plummer extremely high on certainty of being above the leagues average QB doesn't install any confidence to me that they are a good QB to have in a playoff run nor should it imo. Or in a close game vs someone lower on the list like Brady,Bulger(if healthy) or Big Ben.
Click to expand...
Read the comments section. They ran several other methods to get other statistics such as "comeback wins over expected average" and "binomial p values". There's lots of other stuff in there.Anyway, I feel like you're approaching this all wrong. I feel like you already have your mind made up before you even read the article. You've decided that Manning's not very clutch, and that Brady is the clutchest, and you're just searching for points or arguements that can support your prejudice, without reading what the article and comments really have to say. I don't think this article says the first thing about how Peyton Manning can only make comebacks against average teams, or how good a QB is to have in a playoff run. Neither of those points is within the scope of the study.

Also, I definitely think that the fact that a QB is among the top-10 in "clutch" performance- combined with the fact that that QB in general is really good- should definitely inspire confidence that they are a good QB to have in a playoff run. I mean, do you not think that the Niners fans and Broncos fans in the late 80s were extremely confident once their teams got to the playoffs because they had Montana and Elway, two miracle-workers with a knack for huge comebacks?

 
SSOG- I must have been too tired last night. It does clearly state it Does include playoff games. My bad.

As for thinking actually winning % is not as good a predictor as adding a T-Test to it. From the article

"Tom Brady is next on the list and, remember, these numbers do include the postseason. Brady is one of only a handful of quarterbacks with a winning record in more than three games with a fourth-quarter deficit. The others are Marc Bulger (10-5), Ben Roethlisberger (7-2), Steve Young (7-4) and John Elway (7-6). Bulger, who ranks eighth by this metric, is the only one in that group that has yet to win a Super Bowl."

I think 4 of the 5 with actually winning percentages in these situations having a Superbowl ring indicates winning % isn't that bad a starting point and that just being a little better than league average isn't as important in playoffs as actually coming through more often than not.

Anyways you are correct that I've already made up my mind that Brady, is the better QB to have in a close game over Manning. I believe there has been plenty of opportunities for each of them to come through in tight games and Brady has ALREADY shown to be the better bet to come through at a High level and not just a better than average level. I do look at a study that has Plummer,Manning,Testaverde at top of it's comeback ratings with scepticism and think tweaking it with a T-test or anything else that weights it towards rewarding barely above average performance vs higher rates of success in less opportunities as wrong. If arguing guys that both come through more often than not ...say Big Bens 7-2 vs Bradys 13-8 then is Bradys 6-6 in more opportunities better than Big Bens 7-2 without 12 more opps? That's a good debate imo. But saying Mannings 19-29 is better than guys who have actually come through more often than failing? That's a leap I'm not willing to take with the study as I feel it's a flaw to think Big Ben would only win less than 12 of his next 39 tries in this scenario...or that Brady would only win less than 6 times in next 27 of these situations. Study rewards failing more often than not YET it even points out that those who don't fail more often than not have a high correlation to being successful in playoffs against tougher competition and becoming Superbowl Champs.

 
SSOG- I must have been too tired last night. It does clearly state it Does include playoff games. My bad.

As for thinking actually winning % is not as good a predictor as adding a T-Test to it. From the article

"Tom Brady is next on the list and, remember, these numbers do include the postseason. Brady is one of only a handful of quarterbacks with a winning record in more than three games with a fourth-quarter deficit. The others are Marc Bulger (10-5), Ben Roethlisberger (7-2), Steve Young (7-4) and John Elway (7-6). Bulger, who ranks eighth by this metric, is the only one in that group that has yet to win a Super Bowl."

I think 4 of the 5 with actually winning percentages in these situations having a Superbowl ring indicates winning % isn't that bad a starting point and that just being a little better than league average isn't as important in playoffs as actually coming through more often than not.

Anyways you are correct that I've already made up my mind that Brady, is the better QB to have in a close game over Manning. I believe there has been plenty of opportunities for each of them to come through in tight games and Brady has ALREADY shown to be the better bet to come through at a High level and not just a better than average level. I do look at a study that has Plummer,Manning,Testaverde at top of it's comeback ratings with scepticism and think tweaking it with a T-test or anything else that weights it towards rewarding barely above average performance vs higher rates of success in less opportunities as wrong. If arguing guys that both come through more often than not ...say Big Bens 7-2 vs Bradys 13-8 then is Bradys 6-6 in more opportunities better than Big Bens 7-2 without 12 more opps? That's a good debate imo. But saying Mannings 19-29 is better than guys who have actually come through more often than failing? That's a leap I'm not willing to take with the study as I feel it's a flaw to think Big Ben would only win less than 12 of his next 39 tries in this scenario...or that Brady would only win less than 6 times in next 27 of these situations. Study rewards failing more often than not YET it even points out that those who don't fail more often than not have a high correlation to being successful in playoffs against tougher competition and becoming Superbowl Champs.
Click to expand...
First off, I disagree with your premise that Superbowl Wins somehow validate or invalidate statistics. Our very own Doug Drinen did a statistical analysis of how likely the best team is to win the superbowl in any given season (it's a 4-part series, click the links at the top for the other 3). The results might surprise you. Basically, if we replayed last season 10,000 times, then every single team in the league other than San Fran, Houston, and New Orleans would have won the SB at least once, and that if SD was in the NFC South, their chances of winning the SB would more than double. Would some of the more surprising or unsurprising outcomes really validate or invalidate certain statistics?Heck, even in the 4th-quarter comeback article... Kurt Warner has a superbowl ring, and he ranks in the bottom-10 in 4th quarter comebacks. So Brady and Elway having superbowl rings shows how important winning percentage is, while Kurt Warner having a superbowl ring doesn't show anything because that would totally invalidate your arguement?

In my experience, it's usually not a good idea to read an article with your mind already made up. Keep an open mind, actually LISTEN TO and ASSIMILATE the opposing viewpoint, and who knows... you might actually learn something, or even change your mind entirely.

 
Executing under pressure is. And going 5 for 7 in that situation in that important of a game is something many QBs would fail at ....just the time management messes up some QBs/Teams. Squeezing a winning drive in with cool headed plays and excellent clock management is very clutch. Joe Montana keeping his head under pressure is why many think he was so clutch. It's about making plays AND not screwing things up in a tight game.

And the 0 TDs on that drive is meaniless when a TD isn't necessary...a clutch QB is aware of the situation and what is needed. Doesn't force it until he has no other options. If a FG will do it that's good enough.
Click to expand...
Yeah and Manning can't win the big one because his FG kicker can't make the same kick the Venitieri made EVERY time. How clutch would Brady be if Ven. missed like a normal kicker? He probably would have 2 less SB's.As for SSOG posting this article, it would be nice if people did try and understand it rather than rip it first. Outside of SSOG's blind love for Elway over Marino :hey: he presented this only as a reference point to people without his own bias put in...and isn't that what we want?

I for one will tell you that almost everyone you meet has their own bias based on what they see on a VERY limited basis. This is even more true when it comes to clutch or being unclutch. I have major studies on this from baseball and people have VERY selective memory. It would shock people that ARod is actually more clutch than Derek Jeter. This is a fact and I could prove it to anyone over a 3 year period and over 200 AB's in close and late situations. The dreaded selective memory is the albatross of logical thinking.

 
Last edited by a moderator:
That's the dumbest way of defining "clutchness" ever :wall:

How about we do it by winning PERCENTAGE in the 4th quarter, instead of the number of wins? Brady's winning percentage in the 4th quarter >>>>>>> Jake Plummer/Peyton Manning. I'm not even suprised that Bulger is 10-5, he's a great clutch QB imo as well. That list is downright awful.

I'll take Bradys almost 2:1 win percentage and Bulgers actual 2:1 over any of those QBs any day.
Click to expand...
Here's the problem with winning percentage. Let's say QBA is 1-0 when trailing by 8 in the fourth quarter, and QBB is 19-1 when trailing by 8 in the fourth quarter. Who would you rather have, QBA who has a higher winning percentage, or QBB who has done it often enough to prove it wasn't just a fluke?That's the point of the T-Test. It takes sample size into consideration in order to determine what the odds are that it was just a fluke.

I agree that there are better ways to do it, but I certainly don't think that straight winning percentage is it.
Click to expand...
Hmm, you bring up a valid point. I'd say that the best way to rectify that issue would to be subtract 4th quarter losses from wins, and whoever has more after that is the more clutch QB.
 
"Heck, even in the 4th-quarter comeback article... Kurt Warner has a superbowl ring, and he ranks in the bottom-10 in 4th quarter comebacks. So Brady and Elway having superbowl rings shows how important winning percentage is, while Kurt Warner having a superbowl ring doesn't show anything because that would totally invalidate your arguement?"

No, an exception doesn't totally invalidate a pattern or argument. I'm not claiming a winning % QB from this scenario will win the Title every year. Just saying they have a better likelihood than someone who is barely above average,. 4 of the 5 Winning % QBS in the 8 points back scenario in last 10 years (with 3 or more opps) have a Title. They have 7 titles total. They don't win everytime...they just have a higher probability of winning. An exception to any rule or pattern doesn't necessarily invalidate that there is a pattern. Winning percent comeback QBs have a clearer pattern of playoff success than The Top 3 on that Comeback list. A fluke? I don't believe so.

To take from that study that Jake or Peyton have some Top level Clutch play or great close contest play in them that just hasn't yet manifested itself in the playoffs yet is a stretch. Less of a stretch is saying that the Top QBs who actually succeed in these siituations more than they fail have something about them or at least a belief from their team mates that they will get it done in a tight game more often than not.

---------------------------------------

Title "Who is the clutchest QB in the NFL since 1996? "

In First post

"well, let's just say that the answer is going to surprise everyone who ever said that Denver/Indianapolis will never win a Superbowl with Jake Plummer/Peyton Manning at QB."

You did introduce the word clutch and try to connect this to an argument about why doubters should believe in Plummer and Peyton in playoff time. Sure in a 10,000 sample Plummer and Peyton would win more than the average QB/Teams would...but they aren't as likely to do it as the guys who succeed more often than fail in tight comeback type games. If going to extrapolate this list to playoff success the way it is ranked is just wrong. QBs who are far above average in tight games are better in playoffs than guys who are a little above average with long track records.

 
If Brady and Elway and Montana are not tops on the list it's a flawed system. Maybe Marino too, not sure. I think Elway broke Marino's record for most wins in the final 2 minutes.

 
You must log in or register to reply here.

Similar threads

Ministry of Pain
Atlanta Falcons Offense '24- WR/RB/TE/QB and the OL
Replies
3
Views
175
zamboni
zamboni
TheDirtyWord
Going Against The Grain - Who You Got?
2 3
Replies
112
Views
5K
Chaka
Chaka
Tau837
The NFL 100: The Athletic’s pursuit of the best players in football history
Replies
25
Views
790
Tau837
Tau837

Users who are viewing this thread

Total: 2 (members: 0, guests: 2)
Top