Stringer's Power Rankings and STL @ MIA Preview

shouright · Oct 16, 2012

Stringer Bell said:

Haven't really looked into that, but it certainly could be a valid theory. I'm not sure what mediating variables would necessarily be used however.
Click to expand...

Take interceptions for instance. You've labeled those as largely random, but how random are they when you introduce as a mediating variable, for example, a team's quarterback's number of seasons as a starter in the NFL? Of course I don't know the answer, but you may find that your model starts to more strongly correlate with winning percentage when you explore stuff like this.

Stringer Bell · Oct 16, 2012

shouright said:

Take interceptions for instance. You've labeled those as largely random, but how random are they when you introduce as a mediating variable, for example, a team's quarterback's number of seasons as a starter in the NFL? Of course I don't know the answer, but you may find that your model starts to more strongly correlate with winning percentage when you explore stuff like this.
Click to expand...

Wouldn't that reflect with some QBs having a correlation and others' not? There would have to be something that indicates the mediating variable to be significant. If you look at INT rates from year to year, there isn't any indication that experience is a factor. I've done distributions of real INT rates compared to what would be a complete random distribution of INT rates, and its essentially shown that 75-80% of INTs are random.

Either way, I'm not really concerned with correlating with past winning percentage as I am future winning percentage.

Desides · Oct 16, 2012

Stringer Bell said:

Right, but Tannehill's passer rating was because he threw for two TDs.
Click to expand...

I'm baffled by the implication here that throwing for more TDs than the other passer should somehow NOT count as a positive, or at all.

Miami had ~5.5 net yards per pass attempt, St. Louis had ~7.1 IIRC. I suppose it depends on which metrics you put more stock in.
Click to expand...

I don't put stock in metrics, I put stock in stats. Net YPA is still an excellent stat, but passer rating is more reflective of the outcome. I know that prior to week 6, passer rating had 65 wins and 11 losses.

Stringer Bell · Oct 16, 2012

Desides said:

I'm baffled by the implication here that throwing for more TDs than the other passer should somehow NOT count as a positive, or at all.
Click to expand...

Right, but it depends on the context. If Team A recovers a fumble on the 10 yd line, then has a 10 yd TD pass on the next play, the passer contributed a 10 yd pass to that TD, did he contribute to that TD as much as the fumble? Conversely, if a QB throws a 50 yard completion to the opponents 1 yd line, and on the next play the FB runs for a 1 yd TD, the QB is more responsible for that TD than the FB. Whomever gets credit for a TD is more reliant on extenuating factors than ability, at least in comparison to Net YPA.

Desides said:

I don't put stock in metrics, I put stock in stats. Net YPA is still an excellent stat, but passer rating is more reflective of the outcome. I know that prior to week 6, passer rating had 65 wins and 11 losses.
Click to expand...

Because the goal is to identify ability, not results. If we want to focus on the outcome of games, we can just look at point differential, or W/L record, or passer rating differential. Net YPA is more predictive of future wins than passer rating.

Stringer Bell · Oct 16, 2012

Also, to be clear, passer rating uses both ypa and TDs. But passing TDs are correlated to YPA. When you're using two separate statistics that correlate with one another, you will inevitably suffer from OVERFITTING.

shouright · Oct 16, 2012

Stringer Bell said:

Either way, I'm not really concerned with correlating with past winning percentage as I am future winning percentage.
Click to expand...

How is it working so far in that regard?

MonstBlitz · Oct 16, 2012

Based on what Stringer has told us about his model I still expect as time continues the teams whose records indicate they should be at the top should slowly float up there as the season continues.

I also think his model shows us which teams are "unlucky". As I said after week 1, the teams that were near the top that looked like they had no business being there had suffered from turnover issues. If the assumption is turnovers can be a result of bad luck or bad breaks, and aren't weighted heavily; it's the teams that gets these bad breaks more often, or maybe doesn't have enough of these bad breaks go their way that may be near the top but not have their record reflect that.

Stringer Bell · Oct 16, 2012

shouright said:

How is it working so far in that regard?
Click to expand...

Well that really raises the question of how to measure its effectiveness. I have a few different theories on how to accomplish that.

This past week the model didn't do a good job predicting who would win the games. It did do a good job of predicting the closing betting line. That is only one week though.

shouright · Oct 16, 2012

MonstBlitz said:

As I said after week 1, the teams that were near the top that looked like they had no business being there had suffered from turnover issues. If the assumption is turnovers can be a result of bad luck or bad breaks, and aren't weighted heavily; it's the teams that gets these bad breaks more often, or maybe doesn't have enough of these bad breaks go their way that may be near the top but not have their record reflect that.
Click to expand...

But this is where I would suspect that something like the starting experience of the QB would account for lots of those "breaks," of course in terms of the part of turnovers consisting of interceptions.

shouright · Oct 16, 2012

Stringer Bell said:

Wouldn't that reflect with some QBs having a correlation and others' not? There would have to be something that indicates the mediating variable to be significant. If you look at INT rates from year to year, there isn't any indication that experience is a factor. I've done distributions of real INT rates compared to what would be a complete random distribution of INT rates, and its essentially shown that 75-80% of INTs are random.
Click to expand...

Help me figure this out: you are determining that INTs are random by correlating them with themselves over time. Over how much time is that done, and how are you specifying numbers of INTs? By team? Are you looking at an individual team's number of INTs from year to year, or are you doing it some other way?

shouright · Oct 17, 2012

shouright said:

But this is where I would suspect that something like the starting experience of the QB would account for lots of those "breaks," of course in terms of the part of turnovers consisting of interceptions.
Click to expand...

I ran some statistics looking at this. Right now the correlation between Stringer's model and winning percentage is 0.61. When you partial out the variance associated with interceptions thrown by team, that correlation increases to 0.65.

ckparrothead · Oct 17, 2012

Stringer Bell said:

That doesn't really surprise me. Seven underdogs won outright. Pretty crazy week for linesmakers. Nice work with your system though
Click to expand...

Yeah for whatever reason, the instincts of decent bettors versus squares are really, really outperforming this year.

Stringer Bell · Oct 17, 2012

shouright said:

Help me figure this out: you are determining that INTs are random by correlating them with themselves over time. Over how much time is that done, and how are you specifying numbers of INTs? By team? Are you looking at an individual team's number of INTs from year to year, or are you doing it some other way?
Click to expand...

Split up a season into two completely randomly selected groups of 8, then find the correlation. I've also done distributions of what would essentially comprise a completely random distribution of interception rates for all 32 teams in one season, and compared that to an actual distribution of interception rates for all 32 teams.

shouright · Oct 17, 2012

Stringer Bell said:

Split up a season into two completely randomly selected groups of 8, then find the correlation.
Click to expand...

And you're doing that by team? And finding that teams don't correlate strongly with themselves in terms of their numbers of interceptions? In other words, the teams that end up with the fewest and most interceptions on the season show lots of variability in interceptions in those eight-game samples?

I've also done distributions of what would essentially comprise a completely random distribution of interception rates for all 32 teams in one season, and compared that to an actual distribution of interception rates for all 32 teams.
Click to expand...

And how does that differ from, let's say, YPA, or some other variable you've deemed to be more systematic?

Stringer Bell · Oct 17, 2012

shouright said:

And you're doing that by team? And finding that teams don't correlate strongly with themselves in terms of their numbers of interceptions? In other words, the teams that end up with the fewest and most interceptions on the season show lots of variability in interceptions in those eight-game samples?
Click to expand...

I do INT rate (INTs/dropbacks). But yes, there is a good amount of variability. You can also do it based on season-to-season, and you'll generally find similar results.

shouright said:

And how does that differ from, let's say, YPA, or some other variable you've deemed to be more systematic?
Click to expand...

The correlation of Net YPA with itself is ~.60. The correlation of INT rate with itself is ~.20.

I linked these earlier, but good info on this subject. Not exactly the same as I've done, but very much along the same lines:

http://www.advancednflstats.com/2010/11/how-random-are-interceptions.html

http://www.footballperspective.com/correlating-passing-stats-with-wins/

shouright · Oct 18, 2012

Stringer Bell said:

The correlation of Net YPA with itself is ~.60. The correlation of INT rate with itself is ~.20.

I linked these earlier, but good info on this subject. Not exactly the same as I've done, but very much along the same lines:

http://www.advancednflstats.com/2010/11/how-random-are-interceptions.html

http://www.footballperspective.com/correlating-passing-stats-with-wins/
Click to expand...

IMO there has to be some signal in there amidst the noise of interceptions, whether it be due to ability, situational variables in games (being down big and having to pass a lot, for example), or both.

IMO you have to do a lot more exploratory work in this regard in order to validly call your model "ability." Of course it isn't a problem if you want to use that label only heuristically, but I think there's a likely a good deal of "ability" in the variance in winning percentage it doesn't account for. It's just that the variables you would need to include require either more statistical exploration and/or a more sophisticated statistical procedure to capture the effect of them on winning percentage.

Stringer Bell · Oct 18, 2012

shouright said:

IMO there has to be some signal in there amidst the noise of interceptions, whether it be due to ability, situational variables in games (being down big and having to pass a lot, for example), or both.
Click to expand...

There is, and the model incorporates INT-rate. It just isn't a major factor.

shouright said:

IMO you have to do a lot more exploratory work in this regard in order to validly call your model "ability." Of course it isn't a problem if you want to use that label only heuristically, but I think there's a likely a good deal of "ability" in the variance in winning percentage it doesn't account for. It's just that the variables you would need to include require either more statistical exploration and/or a more sophisticated statistical procedure to capture the effect of them on winning percentage.
Click to expand...

I don't necessarily agree. If the game is ~50% luck, you'd be hard pressed to find anything with more than 70% predictivity.

ckparrothead · Oct 18, 2012

Stringer,

If you were to take your raw Rating & SOS numbers and apply them predictively to this weekend's matchups, how would you do it? Just curious.

Stringer Bell · Oct 18, 2012

ckparrothead said:

Stringer,

If you were to take your raw Rating & SOS numbers and apply them predictively to this weekend's matchups, how would you do it? Just curious.
Click to expand...

The numbers are the result of a logistic regression. I can easily come up with win probabilities. I'm not necessarily sure thats the path I want to go down right now.

Also, the ratings I post already incorporate the SOS numbers.

shouright · Oct 18, 2012

Stringer Bell said:

I don't necessarily agree. If the game is ~50% luck, you'd be hard pressed to find anything with more than 70% predictivity.
Click to expand...

Right, but what establishes that as fact?

shouright · Oct 18, 2012

Stringer Bell said:

The numbers are the result of a logistic regression. I can easily come up with win probabilities. I'm not necessarily sure thats the path I want to go down right now.
Click to expand...

Well, of course predicting the closing betting line is a pretty good path to go down, since that's often pretty discrepant from the opening one.

Stringer Bell · Oct 18, 2012

shouright said:

Well, of course predicting the closing betting line is a pretty good path to go down, since that's often pretty discrepant from the opening one.
Click to expand...

It is discrepant enough. Closing lines are going to be the most accurate assessment you'll find, simply because markets operate efficiently. If you can beat line movement you'll make money.

rafael · Oct 18, 2012

Stringer Bell said:

There is, and the model incorporates INT-rate. It just isn't a major factor.

I don't necessarily agree. If the game is ~50% luck, you'd be hard pressed to find anything with more than 70% predictivity.
Click to expand...

shouright said:

Right, but what establishes that as fact?
Click to expand...

It was years ago, but I posted a link to an article that examined that from a statistical perspective. I don't recall the specifics, but I do remember that they concluded that no statistical model would ever be above 75% b/c they felt that 25% of the outcomes were due to luck. Obviously it's just an opinion even though they supported it with facts, but I tend to think they were pretty close to the truth just based on my observations.

shouright · Oct 18, 2012

Stringer Bell said:

It is discrepant enough. Closing lines are going to be the most accurate assessment you'll find, simply because markets operate efficiently. If you can beat line movement you'll make money.
Click to expand...

It sounds like you may have heard my post as sarcastic. It wasn't. I'm with you on this.

Stringer Bell · Oct 23, 2012

After week 7:

[table="class: grid"]

[TR][td]RANK[/td][td]TEAM[/td][td]SOS[/td][td]RATING[/td][td]LAST[/td][td]CHANGE[/td][/tr] [tr][td]1[/td][td]Houston Texans[/td][td]-0.043818844[/td][td]1.29576854[/td][td]3[/td][td]2[/td][/tr] [tr][td]2[/td][td]San Francisco 49ers[/td][td]0.024272026[/td][td]1.267917913[/td][td]1[/td][td]-1[/td][/tr] [tr][td]3[/td][td]Denver Broncos[/td][td]0.104618158[/td][td]0.940204318[/td][td]2[/td][td]-1[/td][/tr] [tr][td]4[/td][td]Green Bay Packers[/td][td]0.098435971[/td][td]0.634115802[/td][td]6[/td][td]2[/td][/tr] [tr][td]5[/td][td]Pittsburgh Steelers[/td][td]-0.046937314[/td][td]0.634004475[/td][td]8[/td][td]3[/td][/tr] [tr][td]6[/td][td]Minnesota Vikings[/td][td]-0.063214098[/td][td]0.525334128[/td][td]4[/td][td]-2[/td][/tr] [tr][td]7[/td][td]Chicago Bears[/td][td]-0.221237484[/td][td]0.506425001[/td][td]5[/td][td]-2[/td][/tr] [tr][td]8[/td][td]Detroit Lions[/td][td]0.12372104[/td][td]0.43575936[/td][td]9[/td][td]1[/td][/tr] [tr][td]9[/td][td]Miami Dolphins[/td][td]0.017547151[/td][td]0.246029681[/td][td]12[/td][td]3[/td][/tr] [tr][td]10[/td][td]St. Louis Rams[/td][td]-0.05960324[/td][td]0.211816484[/td][td]7[/td][td]-3[/td][/tr] [tr][td]11[/td][td]New England Patriots[/td][td]-0.095593584[/td][td]0.165592643[/td][td]15[/td][td]4[/td][/tr] [tr][td]12[/td][td]Seattle Seahawks[/td][td]-0.098629343[/td][td]0.125944893[/td][td]10[/td][td]-2[/td][/tr] [tr][td]13[/td][td]Atlanta Falcons[/td][td]-0.108558706[/td][td]0.098564293[/td][td]16[/td][td]3[/td][/tr] [tr][td]14[/td][td]New York Giants[/td][td]-0.034538636[/td][td]0.071629252[/td][td]13[/td][td]-1[/td][/tr] [tr][td]15[/td][td]Oakland Raiders[/td][td]0.005394977[/td][td]0.024063541[/td][td]18[/td][td]3[/td][/tr] [tr][td]16[/td][td]Washington Redskins[/td][td]-0.190036472[/td][td]-0.029814209[/td][td]11[/td][td]-5[/td][/tr] [tr][td]17[/td][td]Dallas Cowboys[/td][td]-0.017382696[/td][td]-0.112432657[/td][td]20[/td][td]3[/td][/tr] [tr][td]18[/td][td]Philadelphia Eagles[/td][td]-0.016547081[/td][td]-0.122348309[/td][td]19[/td][td]1[/td][/tr] [tr][td]19[/td][td]Carolina Panthers[/td][td]-0.228128121[/td][td]-0.231451637[/td][td]17[/td][td]-2[/td][/tr] [tr][td]20[/td][td]Arizona Cardinals[/td][td]-0.144213545[/td][td]-0.279974621[/td][td]22[/td][td]2[/td][/tr] [tr][td]21[/td][td]Baltimore Ravens[/td][td]-0.310359422[/td][td]-0.332718608[/td][td]14[/td][td]-7[/td][/tr] [tr][td]22[/td][td]New York Jets[/td][td]-0.025160447[/td][td]-0.447236789[/td][td]23[/td][td]1[/td][/tr] [tr][td]23[/td][td]Tampa Bay Buccaneers[/td][td]-0.282884226[/td][td]-0.464095072[/td][td]24[/td][td]1[/td][/tr] [tr][td]24[/td][td]Cincinnati Bengals[/td][td]-0.310595189[/td][td]-0.472267765[/td][td]21[/td][td]-3[/td][/tr] [tr][td]25[/td][td]Tennessee Titans[/td][td]0.079812716[/td][td]-0.582369682[/td][td]28[/td][td]3[/td][/tr] [tr][td]26[/td][td]Indianapolis Colts[/td][td]-0.052301611[/td][td]-0.606033721[/td][td]25[/td][td]-1[/td][/tr] [tr][td]27[/td][td]San Diego Chargers[/td][td]-0.149000507[/td][td]-0.677447997[/td][td]26[/td][td]-1[/td][/tr] [tr][td]28[/td][td]Cleveland Browns[/td][td]-0.269198529[/td][td]-0.84928921[/td][td]27[/td][td]-1[/td][/tr] [tr][td]29[/td][td]Buffalo Bills[/td][td]-0.236165895[/td][td]-1.069174505[/td][td]30[/td][td]1[/td][/tr] [tr][td]30[/td][td]New Orleans Saints[/td][td]-0.162914338[/td][td]-1.076572832[/td][td]29[/td][td]-1[/td][/tr] [tr][td]31[/td][td]Jacksonville Jaguars[/td][td]-0.061915[/td][td]-1.304859169[/td][td]31[/td][td]0[/td][/tr] [tr][td]32[/td][td]Kansas City Chiefs[/td][td]-0.239914377[/td][td]-2.005087334[/td][td]32[/td][td]0[/td][/tr][/table]

ckparrothead · Oct 23, 2012

Just on a pure wins basis, last week's ratings were 10-3 if you just consider who is rated higher than whom, regardless of home/away venue.

Not sure if that means anything, though.

xphinfanx · Oct 23, 2012

ckparrothead said:

Just on a pure wins basis, last week's ratings were 10-3 if you just consider who is rated higher than whom, regardless of home/away venue.

Not sure if that means anything, though.
Click to expand...

Not straying off from the chart one pick meant something for me this week. Was going to go with home teams within five spots but stayed straight up and it did well here. Very nice work Stringer.

Edit: I did stray on one pick I picked Dallas over Carolina.

jdang307 · Oct 23, 2012

I've kind of missed this thread, so I ask in earnest. What's the reason for some really good teams being underneath us in the rankings?

MonstBlitz · Oct 23, 2012

Looks like it's starting to slowly match up with the records a bit. Only team that is a little shocking at this point is the Giants being at 14. I haven't seen enough of their games to realize if that's about where they should be or not according to your criteria, but I'd expect them to be higher since they do pretty well in all phases of the game.

Stringer Bell · Oct 23, 2012

jdang307 said:

I've kind of missed this thread, so I ask in earnest. What's the reason for some really good teams being underneath us in the rankings?
Click to expand...

Because they aren't as good as you think

Just looking at it team by team. The ones that stand out:

-The Giants and Patriots have a much worse pass defense than Miami.

-Atlanta and Miami are pretty much equal statistically in terms of efficiency.

-Baltimore is slightly worse than Miami in most of the statistics the model uses.

Miami is essentially slightly above average in most measures of efficiency, while being near the top of the league in a couple of others. They don't have any glaring weaknesses to bring them down.

Stringer Bell · Oct 23, 2012

MonstBlitz said:

Looks like it's starting to slowly match up with the records a bit. Only team that is a little shocking at this point is the Giants being at 14. I haven't seen enough of their games to realize if that's about where they should be or not according to your criteria, but I'd expect them to be higher since they do pretty well in all phases of the game.
Click to expand...

Their pass defense is pretty bad. They're giving up 7.5 net YPA.

I think the Giants are better than what I have them rated. But one thing about the Giants is that they generally have never been the type of team that plays very well in the regular season.

xphinfanx · Oct 23, 2012

I'm only stuck on one pick this week. Pats and Rams, with the Rams one ahead of the Pats and the Rams also home field? Everything saying Rams here but my gut says Pats.

Stringer Bell · Oct 30, 2012

Post Week-8:

[table="class: grid"] [TR][td]RANK[/td][td]TEAM[/td][td]SOS[/td][td]RATING[/td][td]LAST[/td][td]CHANGE[/td][/tr] [tr][td]1[/td][td]San Francisco 49ers[/td][td]0.147404097[/td][td]1.334969611[/td][td]2[/td][td]1[/td][/tr] [tr][td]2[/td][td]Houston Texans[/td][td]0.003849036[/td][td]1.287356046[/td][td]1[/td][td]-1[/td][/tr] [tr][td]3[/td][td]Denver Broncos[/td][td]0.080304999[/td][td]1.263146796[/td][td]3[/td][td]0[/td][/tr] [tr][td]4[/td][td]Pittsburgh Steelers[/td][td]-0.003339456[/td][td]0.827829482[/td][td]5[/td][td]1[/td][/tr] [tr][td]5[/td][td]Atlanta Falcons[/td][td]-0.046371089[/td][td]0.470067199[/td][td]13[/td][td]8[/td][/tr] [tr][td]6[/td][td]Green Bay Packers[/td][td]0.01864482[/td][td]0.44236549[/td][td]4[/td][td]-2[/td][/tr] [tr][td]7[/td][td]Minnesota Vikings[/td][td]-0.041481422[/td][td]0.394959908[/td][td]6[/td][td]-1[/td][/tr] [tr][td]8[/td][td]Seattle Seahawks[/td][td]0.112637486[/td][td]0.351986396[/td][td]12[/td][td]4[/td][/tr] [tr][td]9[/td][td]New England Patriots[/td][td]-0.027478913[/td][td]0.347570916[/td][td]11[/td][td]2[/td][/tr] [tr][td]10[/td][td]Miami Dolphins[/td][td]0.023329732[/td][td]0.339380849[/td][td]9[/td][td]-1[/td][/tr] [tr][td]11[/td][td]Chicago Bears[/td][td]-0.052932373[/td][td]0.264095243[/td][td]7[/td][td]-4[/td][/tr] [tr][td]12[/td][td]Detroit Lions[/td][td]0.08166423[/td][td]0.222591758[/td][td]8[/td][td]-4[/td][/tr] [tr][td]13[/td][td]Carolina Pantrers[/td][td]-0.021472322[/td][td]0.200301218[/td][td]19[/td][td]6[/td][/tr] [tr][td]14[/td][td]Oakland Raiders[/td][td]-0.007688371[/td][td]0.171167854[/td][td]15[/td][td]1[/td][/tr] [tr][td]15[/td][td]New York Giants[/td][td]-0.007501363[/td][td]0.161897565[/td][td]14[/td][td]-1[/td][/tr] [tr][td]16[/td][td]Washington Redskins[/td][td]0.007116322[/td][td]0.02211554[/td][td]16[/td][td]0[/td][/tr] [tr][td]17[/td][td]Arizona Cardinals[/td][td]0.213501328[/td][td]0.021659877[/td][td]20[/td][td]3[/td][/tr] [tr][td]18[/td][td]St. Louis Rams[/td][td]0.107532091[/td][td]-0.056185086[/td][td]10[/td][td]-8[/td][/tr] [tr][td]19[/td][td]Baltimore Ravens[/td][td]-0.084070154[/td][td]-0.162509713[/td][td]21[/td][td]2[/td][/tr] [tr][td]20[/td][td]Tampa Bay Buccaneers[/td][td]-0.130262448[/td][td]-0.237581041[/td][td]23[/td][td]3[/td][/tr] [tr][td]21[/td][td]Philadelphia Eagles[/td][td]0.057203806[/td][td]-0.237581912[/td][td]18[/td][td]-3[/td][/tr] [tr][td]22[/td][td]Dallas Cowboys[/td][td]0.045900275[/td][td]-0.24334077[/td][td]17[/td][td]-5[/td][/tr] [tr][td]23[/td][td]Cincinnati Bengals[/td][td]-0.059411804[/td][td]-0.277164754[/td][td]24[/td][td]1[/td][/tr] [tr][td]24[/td][td]New York Jets[/td][td]0.14305197[/td][td]-0.297776903[/td][td]22[/td][td]-2[/td][/tr] [tr][td]25[/td][td]San Diego Chargers[/td][td]-0.123464386[/td][td]-0.522035903[/td][td]27[/td][td]2[/td][/tr] [tr][td]26[/td][td]Tennessee Titans[/td][td]0.047667663[/td][td]-0.558055716[/td][td]25[/td][td]-1[/td][/tr] [tr][td]27[/td][td]Cleveland Browns[/td][td]-0.152328242[/td][td]-0.58193811[/td][td]28[/td][td]1[/td][/tr] [tr][td]28[/td][td]Indianapolis Colts[/td][td]-0.071055374[/td][td]-0.603659845[/td][td]26[/td][td]-2[/td][/tr] [tr][td]29[/td][td]Jacksonville Jaguars[/td][td]0.095407706[/td][td]-0.858001469[/td][td]31[/td][td]2[/td][/tr] [tr][td]30[/td][td]Buffalo Bills[/td][td]-0.087666649[/td][td]-0.976755633[/td][td]29[/td][td]-1[/td][/tr] [tr][td]31[/td][td]New Orleans Saints[/td][td]-0.029941406[/td][td]-1.24513707[/td][td]30[/td][td]-1[/td][/tr] [tr][td]32[/td][td]Kansas City Chiefs[/td][td]-0.142450973[/td][td]-1.974256474[/td][td]32[/td][td]0[/td][/tr][/table]

shouright · Oct 30, 2012

I guess we got lucky?

dWreck · Oct 30, 2012

how da fudge did we go down

Stringer Bell · Oct 30, 2012

dcafiero said:

how da fudge did we go down
Click to expand...

Short answer: we didn't pass the ball very efficiently, and the model doesn't really put much stock in blocked punts, onside kicks, etc.

xphinfanx · Oct 30, 2012

shouright said:

I guess we got lucky?
Click to expand...

Something we don't get very often.

Whoa the Rams took a hit.

dWreck · Oct 30, 2012

Stringer Bell said:

Short answer: we didn't pass the ball very efficiently, and the model doesn't really put much stock in blocked punts, onside kicks, etc.
Click to expand...

Gotcha.. kinda had a feeling that was the case.

shouright · Oct 30, 2012

Stringer do you realize that in your formulation the Dolphins are four standard deviations above the mean in team rating?

If the ability of the 11th-ranked team in the league is truly four standard deviations above the mean, I'd say you have the data to argue that there really isn't all that much parity in the NFL!

Stringer Bell · Oct 30, 2012

shouright said:

Stringer do you realize that in your formulation the Dolphins are four standard deviations above the mean in team rating?
Click to expand...

Didn't realize that, no. What do you make of it?

Log in / Register

Useful Searches

Stringer's Power Rankings and STL @ MIA Preview

shouright Banned

Stringer Bell Post Hard, Post Often Club Member

Desides Well-Known Member

Stringer Bell Post Hard, Post Often Club Member

Stringer Bell Post Hard, Post Often Club Member

shouright Banned

MonstBlitz Nobody's Fart Catcher

Stringer Bell Post Hard, Post Often Club Member

shouright Banned

shouright Banned

shouright Banned

ckparrothead Draft Forum Moderator Luxury Box

Stringer Bell Post Hard, Post Often Club Member

shouright Banned

Stringer Bell Post Hard, Post Often Club Member

shouright Banned

Stringer Bell Post Hard, Post Often Club Member

ckparrothead Draft Forum Moderator Luxury Box

Stringer Bell Post Hard, Post Often Club Member

shouright Banned

shouright Banned

Stringer Bell Post Hard, Post Often Club Member

rafael Well-Known Member

shouright Banned

Stringer Bell Post Hard, Post Often Club Member

ckparrothead Draft Forum Moderator Luxury Box

xphinfanx Stay strong my friends.

jdang307 Season Ticket Holder Club Member

MonstBlitz Nobody's Fart Catcher

Stringer Bell Post Hard, Post Often Club Member

Stringer Bell Post Hard, Post Often Club Member

xphinfanx Stay strong my friends.

Stringer Bell Post Hard, Post Often Club Member

shouright Banned

dWreck formerly dcaf

Stringer Bell Post Hard, Post Often Club Member

xphinfanx Stay strong my friends.

dWreck formerly dcaf

shouright Banned

Stringer Bell Post Hard, Post Often Club Member

Share This Page