Wins and Point Diff Around the League

[Hapless Bills Fan]

I've tucked this data from previous weeks into various threads.

Thought I'd give it a shot with its own thread and see what folks make of it.

 

Around the league, the average of points scored/points given up per game is holding steady at 22.6 with a standard deviation of 5.3.

That means a "typical" NFL team scores or gives up between 17-28 points. 

Our D, giving up now an average of 16.6 ppg is still above average.

Our O, scoring19.3 ppg is below average, but not horrid - it's basically 1 score per game below average.  (3.3 points to be specific, or just over 1 FG per game).

 

Data are sorted by Wins.  From NFL.com

With Week 10 in the books, we can see that things are starting to sort themselves out, but the wins still don't totally align with the point differential per game. 

 

For example, the 4th best team on point differential is Dallas.  Beware of the 'Boys.

 

Green Bay and Seattle, on the other hand, are running significantly lower net points/game not only than NE and SF, but a number of teams that have fewer wins including Baltimore, Minnesota, Houston, and (for Seattle) KC, LA, and Buffalo.  That says to me that they're overperforming - their defense in the case of Seattle.

 

Buffalo has the lowest point differential of the 6 win teams, but better than all but two 5 win teams and one each of the 7 and 8 win teams.  So we are certainly not the "worst 6 win team in the history of the league" but we're verging on overperforming our offense.

 

OK hope this is interesting to someone, let me know how you see it.  Will answer any questions, but I think it's pretty stump simple in terms of what I did.

 

I think the bottom line is that most NFL games are a lot closer in score than people think.

 

 

 

Edited November 12, 2019 by Hapless Bills Fan
correct mistake in difference between league average and Bills scoring

[The Frankish Reich]

Very good points. 

Over the long haul point differential says a lot: at the end of the season, the teams with the greatest positive point differential are almost always the objectively “best” teams. 

Right now we’re overperforming our point differential considerably, and even though 9 games is not an adequate sample, it’s suggestive of the Bills being a .500 team at best that may be very fortunate to get to 10 wins and a playoff spot. It’s to Beane’s and McDemott’s credit that they saw this and held back on making a significant deadline trade deal. We are not “one player away” from anything valuable right now. 

[SlimShady'sSpaceForce]

I dunno what to make of it overall.  

 

KC 4.5,  Seattle 2.1,  GB 4.5, NO 2.4  ALL good teams compared to   "crappy"   Buffalo 2.7.  

sounds more like they play crappy defenses and score more points

 

on both sides 

[Hapless Bills Fan]

I'll bump my own thread, but hopefully legit because I'll put more data out there.

It also came up a while back, someone was expressing that the top-scoring teams are the top-winning teams. 

I put out raw data from NFL.com or pro-football-reference that I thought made the point "ain't necessarily so" but someone objected strongly that it was invalid because not corrected for games played (I thought it could be seen it still was, and the objector didn't want to grind the corrected data).

 

Since I have a spreadsheet rocking, here is the corrected data sorted by points for/game and wins.

What you can see, I think, is that while there's a correlation, and generally the top scoring teams are the top winning teams, it's not strong or predictive.

The #3 scoring team right now has 3 wins. 

If you tell me a team scores on average, 20-25 PPG (league average), I can tell you they have somewhere between 2 and 8 wins^^

 

^^ the same can be said of point differential/game, actually, but it's a better correlation because it's taking into account both sides of the ball
Since I was asked: R2 on point diff vs wins 0.759 vs R2 on points scored vs wins 0.544

Edited November 12, 2019 by Hapless Bills Fan
added r2

[section122]

3 minutes ago, SlimShady'sGhost said:

I dunno what to make of it overall.  

 

KC 4.5,  Seattle 2.1,  GB 4.5, NO 2.4  ALL good teams compared to   "crappy"   Buffalo 2.7.  

sounds more like they play crappy defenses and score more points

 

on both sides 

 

They have crappier defenses but better offenses.  I opined this in another thread but it is because winning with defense is not sexy that people think the Bills are bad.  The Bills have a very good defense and as the numbers show a below average offense.  

 

I was just talking yesterday that if the Bills scored 21 points in each game they would only have 1 loss.  That is what they need to improve on and it really isn't that far from where they are now.  It works out to be an extra field goal per game.  They haven't arrived yet but they are not as far away as some people think.

 

I don't know if this is the place for it but the defense isn't getting enough love imo.  Last year they were the #2 defense by yards but people were mad that they were 18th in ppg against and that was the stat that really mattered.  Now they are 21st in rush defense and 3rd in ppg against and people are still blaming them for the loss against Cleveland.  In my mind if the offense doesn't score 24 they aren't doing their jobs.  If the defense gives up more than 21 they aren't doing theirs.  Are we still doing heat maps?

 

Win against Jets - Defense

Win against Giants - Team

Win against Bengals - Defense

Loss against Patriots - Offense

Win against Titans - Defense

Win against Fins - Team

Loss against the Eagles - Team

Win against Redskins - Team

Loss against Browns - Offense

 

So I have the defense responsible for 3 wins on their own, 4 more games where they did their job, and 1 game they didn't

The Offense I have responsible for 0 wins on their own, 2 losses on their own, 4 games where they did their job.

 

Defensively we are a 7-1 team.

Offensively we are a 4-4 team.

 

Sounds about right to me.  

[SlimShady'sSpaceForce]

4 minutes ago, section122 said:

They have crappier defenses but better offenses.  I opined this in another thread but it is because winning with defense is not sexy that people think the Bills are bad.  The Bills have a very good defense and as the numbers show a below average offense.  

 

I was just talking yesterday that if the Bills scored 21 points in each game they would only have 1 loss.  That is what they need to improve on and it really isn't that far from where they are now.  It works out to be an extra field goal per game.  They haven't arrived yet but they are not as far away as some people think.

 

I don't know if this is the place for it but the defense isn't getting enough love imo.  Last year they were the #2 defense by yards but people were mad that they were 18th in ppg against and that was the stat that really mattered.  Now they are 21st in rush defense and 3rd in ppg against and people are still blaming them for the loss against Cleveland.  In my mind if the offense doesn't score 24 they aren't doing their jobs.  If the defense gives up more than 21 they aren't doing theirs.  Are we still doing heat maps?

 

This is kind of what I was getting at ....

 

is the offense better because they play crappier defenses? 

 

I am not saying Dabolls offensive play calling ISN'T offensive ....  but all those things matter.  

[section122]

Just now, SlimShady'sGhost said:

 

This is kind of what I was getting at ....

 

is the offense better because they play crappier defenses? 

 

I am not saying Dabolls offensive play calling ISN'T offensive ....  but all those things matter.  

 

No the 4 teams you mentioned (KC, Seattle, GB, and NO) all have elite play at the QB position.  The Bills don't have that (yet).  QBs are sexy and teams with those qbs will always be favored or thought higher of because of it.

 

I really don't think Daboll's play calling is as bad as people think.  I think the execution is poor.  I think Frank Gore is washed so not getting Singletary more involved is a sore spot for me.  However I think the Bills need a #1 WR to bump everyone down a spot.  John Brown has played great and exceeded my expectations outside of him though there are no threats at the position.  Beasley is a dink and dunk guy, there is no "big WR", and they need to improve their RB room. 

 

I keep going back to the Bills have no game breakers.  The Browns have OBJ, Landry, Hunt, and Chubb at the skill position who are all better than their Bills counterpart.  Eagles got Wentz Alshon Jeffery and spent a first on wr (agholor), Rams got Goff Cooks and Robert Woods, KC has an obscene plethora of weapons for Mahomes.  The Bills need to get Josh some talent to take some of the pressure off of him.  They need to get Daboll some better talent to work with.  Hopefully this off season those things happen.

 

On a side note if Daboll is fired I will be hoping and praying for Jay Gruden but fear we will see Ken Whisenhunt.

 

[row_33]

51 minutes ago, The Frankish Reich said:

 

Over the long haul point differential says a lot: at the end of the season, the teams with the greatest positive point differential are almost always the objectively “best” teams. 

 

 

duh....

 

 

when you win you have more points

 

and if you win 13 or 14 times it would take a heck of a lopsided set of losses to upset standing for point differential

 

the stat adds NOTHING of use.

 

 

Edited November 12, 2019 by row_33

[ticketssince61]

Two things that need to be kept in mind when reviewing these "statistics"

 

1. Quality of opposition - scoring 21 points against NE is much better than 21 against the Bengals

 

2. The points per game can get skewed over 1 or 2 blow out wins (or losses). e.g. The Ravens' 59 points against the Dolphins can make up for a few below average scoring games

 

The more important measure is consistency (i.e. Standard Deviation) in the points (scored or allowed). 

 

As an extreme case, a team that alternately scores either 0 or 50 may have a vastly different record than a a team that scores 25 every game

 

Also, important to take into account TD's on turnovers and by special teams (which we are sorely lacking)

 

To show points #1 and #2, The Cowboys should have a really good record, but their points were mostly piled up in a few blow out wins against very inferior opponents

    Giants       35 and 37

    Redskins  31

    Dolphins   31

 

 

[Big Turk]

55 minutes ago, section122 said:

 

They have crappier defenses but better offenses.  I opined this in another thread but it is because winning with defense is not sexy that people think the Bills are bad.  The Bills have a very good defense and as the numbers show a below average offense.  

 

I was just talking yesterday that if the Bills scored 21 points in each game they would only have 1 loss.  That is what they need to improve on and it really isn't that far from where they are now.  It works out to be an extra field goal per game.  They haven't arrived yet but they are not as far away as some people think.

 

I don't know if this is the place for it but the defense isn't getting enough love imo.  Last year they were the #2 defense by yards but people were mad that they were 18th in ppg against and that was the stat that really mattered.  Now they are 21st in rush defense and 3rd in ppg against and people are still blaming them for the loss against Cleveland.  In my mind if the offense doesn't score 24 they aren't doing their jobs.  If the defense gives up more than 21 they aren't doing theirs.  Are we still doing heat maps?

 

Win against Jets - Defense

Win against Giants - Team

Win against Bengals - Defense

Loss against Patriots - Offense

Win against Titans - Defense

Win against Fins - Team

Loss against the Eagles - Team

Win against Redskins - Team

Loss against Browns - Offense

 

So I have the defense responsible for 3 wins on their own, 4 more games where they did their job, and 1 game they didn't

The Offense I have responsible for 0 wins on their own, 2 losses on their own, 4 games where they did their job.

 

Defensively we are a 7-1 team.

Offensively we are a 4-4 team.

 

Sounds about right to me.  

 

Would help if their kicker had enough leg to hit NFL FGs still and didn't force the coach to go for it when presented with a 53 yard FG during the game....that is a pretty basic FG distance these days, and if your kicker can't reach that he is really hurting you.

Edited November 12, 2019 by matter2003

[Hapless Bills Fan]

56 minutes ago, section122 said:

 

They have crappier defenses but better offenses.  I opined this in another thread but it is because winning with defense is not sexy that people think the Bills are bad.  The Bills have a very good defense and as the numbers show a below average offense.  

 

I was just talking yesterday that if the Bills scored 21 points in each game they would only have 1 loss.  That is what they need to improve on and it really isn't that far from where they are now.  It works out to be an extra field goal per game.  They haven't arrived yet but they are not as far away as some people think.

 

I don't know if this is the place for it but the defense isn't getting enough love imo.  Last year they were the #2 defense by yards but people were mad that they were 18th in ppg against and that was the stat that really mattered.  Now they are 21st in rush defense and 3rd in ppg against and people are still blaming them for the loss against Cleveland.  In my mind if the offense doesn't score 24 they aren't doing their jobs.  If the defense gives up more than 21 they aren't doing theirs.  Are we still doing heat maps?

 

Win against Jets - Defense

Win against Giants - Team

Win against Bengals - Defense

Loss against Patriots - Offense

Win against Titans - Defense

Win against Fins - Team

Loss against the Eagles - Team

Win against Redskins - Team

Loss against Browns - Offense

 

So I have the defense responsible for 3 wins on their own, 4 more games where they did their job, and 1 game they didn't

The Offense I have responsible for 0 wins on their own, 2 losses on their own, 4 games where they did their job.

 

Defensively we are a 7-1 team.

Offensively we are a 4-4 team.

 

Sounds about right to me.  

 

Good post.

 

That's exactly what the data show.  We are 3.3 points below league average as an offense.  Just over 1 FG per game.

[SlimShady'sSpaceForce]

37 minutes ago, section122 said:

 

No the 4 teams you mentioned (KC, Seattle, GB, and NO) all have elite play at the QB position.  The Bills don't have that (yet).  QBs are sexy and teams with those qbs will always be favored or thought higher of because of it.

 

I really don't think Daboll's play calling is as bad as people think.  I think the execution is poor.  I think Frank Gore is washed so not getting Singletary more involved is a sore spot for me.  However I think the Bills need a #1 WR to bump everyone down a spot.  John Brown has played great and exceeded my expectations outside of him though there are no threats at the position.  Beasley is a dink and dunk guy, there is no "big WR", and they need to improve their RB room. 

 

I keep going back to the Bills have no game breakers.  The Browns have OBJ, Landry, Hunt, and Chubb at the skill position who are all better than their Bills counterpart.  Eagles got Wentz Alshon Jeffery and spent a first on wr (agholor), Rams got Goff Cooks and Robert Woods, KC has an obscene plethora of weapons for Mahomes.  The Bills need to get Josh some talent to take some of the pressure off of him.  They need to get Daboll some better talent to work with.  Hopefully this off season those things happen.

 

On a side note if Daboll is fired I will be hoping and praying for Jay Gruden but fear we will see Ken Whisenhunt.

 

 

1. the O Line isn't protecting Josh's blind side 
2. Elite QB's are scoring points.  20 game Josh is not.  Nothing shocking there. 
3.  Execution is poor and there have been too many dropped balls.  
4. ONLY 20 total rushes vs Cleveland in NOT poor execution but crappy play calling. 
5. Not as many but still DUMB penalties.  

[Hapless Bills Fan]

31 minutes ago, ticketssince61 said:

Two things that need to be kept in mind when reviewing these "statistics"

 

1. Quality of opposition - scoring 21 points against NE is much better than 21 against the Bengals

 

2. The points per game can get skewed over 1 or 2 blow out wins (or losses). e.g. The Ravens' 59 points against the Dolphins can make up for a few below average scoring games

 

The more important measure is consistency (i.e. Standard Deviation) in the points (scored or allowed). 

 

As an extreme case, a team that alternately scores either 0 or 50 may have a vastly different record than a a team that scores 25 every game

 

Also, important to take into account TD's on turnovers and by special teams (which we are sorely lacking)

 

To show points #1 and #2, The Cowboys should have a really good record, but their points were mostly piled up in a few blow out wins against very inferior opponents

    Giants       35 and 37

    Redskins  31

    Dolphins   31

 

Valid points, except what's up with putting "statistics" in quotes - is that supposed to indicate these are not really statistics or something? 

They are.  Seems extra.

 

I don't know about standard deviation being "more important".  I can't think of an example where standard deviation is interpretable without the context provided by the mean and the values relative to mean. I grant you it would provide additional insight.

 

The PF/PA on NFL.com actually include all points scored, including by ST and on turnovers.

 

Of course, that's the fundamental genesis of the Frankenstats like DVOA and etc.    Folks look at something simple like "points scored" and say "well, no, that doesn't tell the whole story, what about points given up?"  So someone like myself comes along and says "Let's just look at the difference between points scored and points given up and see what that tells us".  And it tells us stuff, but it's clearly not predictive, since teams with the same record have the same diff.

 

Then someone thoughtful like yourself comes along and tries to capture all that stuff - quality of opposition, standard deviation of the PF and PA, points scored on turnovers or off turnovers as a special category - whatever they argue may influence the result. 

 

Then you wind up with a special sauce statistic that is said to be predictive, but gets continually (in private) revamped and reweighted because it isn't quite quite

 

My key points from all this are:

1) this is a reasonable, simple and straightforward way to look at the relative quality of offense and defense on a team

2) nope it's not perfect

3) the Bills are in fact not the worst winning team in the league when both offensive and defensive performance are considered.

4) we are the equivalent of a bit over a FG per game away from being an "average" offense.  1 extra TD every 2 games would do it.

 

 

[SoCal Deek]

You can over analyze it to death. The Bills have a winning record because they’ve played a very weak schedule....but because they’ve played a weak schedule their point differential should be much higher. In other words they’re just barely getting by now. Wait until they play Dallas, Pittsburgh, New England, etc

[DaBillsFanSince1973]

19 minutes ago, Hapless Bills Fan said:

That's exactly what the data show.  We are 3.3 points below league average as an offense.  Just over 1 FG per game.

not good enough, below average is going to continue to bite them in close games. if they don't pick it up pronto, like sunday pronto, they will not win many out of the next 7 games.

 

the browns game showed me they just can't get it done. 16 points

[Boatdrinks]

9 minutes ago, SoCal Deek said:

You can over analyze it to death. The Bills have a winning record because they’ve played a very weak schedule....but because they’ve played a weak schedule their point differential should be much higher. In other words they’re just barely getting by now. Wait until they play Dallas, Pittsburgh, New England, etc

Yes, in a nutshell the Bills are half a football team. The defense is above average, and the offense is below. In today’s NFL, it seems better to be strong on offense and weaker on D than the other way around. The game favors offense so much now, that most assets are best spent on that side of the ball. An elite pass rusher and a top CB / cover LB should be the major investments on D. 20 years ago a top D could carry a team. 

[saundena]

46 minutes ago, ticketssince61 said:

Two things that need to be kept in mind when reviewing these "statistics"

 

1. Quality of opposition - scoring 21 points against NE is much better than 21 against the Bengals

 

 

This is the key point.  It would be interesting to plot this against opponents win loss record

Edited November 12, 2019 by saundena

[Hapless Bills Fan]

Just now, DaBillsFanSince1973 said:

not good enough, below average is going to continue to bite them in close games. if they don't pick it up pronto, like sunday pronto, they will not win many out of the next 7 games.

 

the browns game showed me they just can't get it done. 16 points

 

They did miss a FG from chip shot XP distance.

 

I'm not saying the offense doesn't need to improve, nor that below average offense won't bite us

But I'm puzzled at the "just can't get it done" conclusion.

 

The facts are, the Bills at Browns game was very close statistically.  The Browns had ~3.5m edge on TOP, which is big.  44 more offensive yards.  A missed FG from makeable distance, 0-2 on 4th down conversions vs 1-2.  No turnovers.

 

We lost.

 

But the roots of how we could have won are there, and they don't seem out of reach.

2 minutes ago, saundena said:

 

This is the key point.  It would be interesting to plot this against opponents win loss record

 

Go For It!  I can send you my point diff spreadsheet to start with if that helps out.

[SlimShady'sSpaceForce]

21-20 is a guaranteed Win 

 

the rest is ALL noise 

you think NE gave a shite winning the SB 13-3?

 

you think they are still worried they only had 3 points by the half? 

 

[whatdrought]

Which really makes you wonder what happens if our defense keeps playing with low % but our offense finds itself. Could be good. Could never happen, but could be really good.