I'm working on two analytical articles for next week... would welcome questions

[KB_fan]

Hi all,

I wanted to give you a heads' up that with some free time I've had the past week, I've been doing some Panthers data analysis, and I'm working on two analytical articles that I plan to post here early next week once the dust from roster cuts settles a little bit.   I thought I'd give you a chance to ask questions about both topics.  Your questions might help me sharpen my analysis.

 

1) Topic 1:  Week 1 wins & losses - do they matter?  A look at NFL history for all teams since 1995.   (I've been tweeting a bunch of stuff on this today, some of you may have seen those charts and comments.  My twitter feed:  https://twitter.com/PanthersFan_Afr)  Panthers have a pretty lousy week 1 record ... I'll take a closer look at how our week 1 record may have shaped our team history.

 

2) Topic 2:  Strength of Schedule analysis - do all games matter equally?  Are some wins or losses bigger than others?  How have we done against our most difficult opponents over the years and does it matter?  We've all heard the comments about our easy schedule in 2015.  Was it true?  Was our 2016 schedule really hard?  What might be our most important games in 2017?

[electro's horse]

do you think either of these have not been done before?

[KB_fan]

2 minutes ago, electro's horse said:

do you think either of these have not been done before?

No, but they may not be current, or may not be Panthers-specific.

If you have links to anything of interest on either topic, I'd love to see them.

Never claimed my analysis pieces are unique or never been done.  I do hope they prove interesting to current Panthers fans, that's all.

[electro's horse]

yeah, FO, 538, PFF have all done regression analysis like this

week 1 wins kinda matter, which is what you'd expect in a league that only plays 16 games a year

SOS isn't predictive at all

[rayzor]

5 minutes ago, electro's horse said:

do you think either of these have not been done before?

I know its hard, but don't be such an ass.

[stirs]

1 minute ago, electro's horse said:

yeah, FO, 538, PFF have all done regression analysis like this

week 1 wins kinda matter, which is what you'd expect in a league that only plays 16 games a year

SOS isn't predictive at all

Are wet blankets your specialty?

[rayzor]

Back on topic...considering the league likes to start us off on the road so often it would be good to see how we fair vs. national average and just how much a factor being away plays into it. 

[KB_fan]

2 minutes ago, electro's horse said:

yeah, FO, 538, PFF have all done regression analysis like this

week 1 wins kinda matter, which is what you'd expect in a league that only plays 16 games a year

Not for all teams.  Some teams seem to really be affected by week 1 losses, notably the Panthers.  Other teams, not so much, including a few teams who made the playoffs MORE often in years with week one losses...

 

2 minutes ago, electro's horse said:

SOS isn't predictive at all

I agree that "pre-season" SoS (based on teams' prior season records) means very little.   I'm talking more ACTUAL Strength of Schedule.... how we've done against the winningest teams each season.  That seems to matter A LOT for Panthers...  When we do well against our toughest opponents we have a better season record.  That's not rocket science necessarily, but I found it pretty interesting.  Psychology matters in football.  I think I can prove numerically that wins vs. SF in 2013 and Seattle in 2015 made a big difference in the outcome of both seasons...  It's common sense, but it's fun to see data to back up what one would expect.

[KB_fan]

2 minutes ago, rayzor said:

Back on topic...considering the league likes to start us off on the road so often it would be good to see how we fair vs. national average and just how much a factor being away plays into it. 

Yup.  Good question.  I'm definitely looking into this.

Just for now, we've played a total of 10 home week 1 games, and 12 away., but only 1 home game since 2010. (in 2013).

[electro's horse]

1 minute ago, KB_fan said:

Not for all teams.  Some teams seem to really be affected by week 1 losses, notably the Panthers.  Other teams, not so much, including a few teams who made the playoffs MORE often in years with week one losses...

 

I agree that "pre-season" SoS (based on teams' prior season records) means very little.   I'm talking more ACTUAL Strength of Schedule.... how we've done against the winningest teams each season.  That seems to matter A LOT for Panthers...  When we do well against our toughest opponents we have a better season record.  That's not rocket science necessarily, but I found it pretty interesting.  Psychology matters in football.  I think I can prove numerically that wins vs. SF in 2013 and Seattle in 2015 made a big difference in the outcome of both seasons...  It's common sense, but it's fun to see data to back up what one would expect.

I sincerely doubt your first point. like i bet week 1 losses v. playoff appearances correlates much stronger with sos during the season than any specific team. to predict what you're going to say, a team like hte patriots has 7 byes a year. 

As to your second, yes, you'd expect better teams to beat teams with better w/l at the end of the season, as you indicate. but if you want to do a regression analysis, divest yourself from emotional arguments. 

for example, that win in SF? SF poo housed them in the playoffs. and that win against Seattle? they very nearly gave up the 2nd largest playoff lead of all time. 

[KB_fan]

3 minutes ago, electro's horse said:

I sincerely doubt your first point. like i bet week 1 losses v. playoff appearances correlates much stronger with sos during the season than any specific team. to predict what you're going to say, a team like hte patriots has 7 byes a year. 

I'll try to take SoS into account.  That's a valid point.

3 minutes ago, electro's horse said:

As to your second, yes, you'd expect better teams to beat teams with better w/l at the end of the season, as you indicate. but if you want to do a regression analysis, divest yourself from emotional arguments. 

for example, that win in SF? SF poo housed them in the playoffs. and that win against Seattle? they very nearly gave up the 2nd largest playoff lead of all time. 

I started with data analysis and not any emotional argument.  My comments about the SF win and Seattle wins came AFTER I'd done the data analysis.  And I'm not factoring in playoff games at all.  Just final regular season W-L record compared to the quality of our opponents and our results against them each season. 

Just how there are "BIG Plays" made or allowed that can contribute to a W or L in a given game, I think the data show that there are BIG WINS or BIG LOSSES that swing seasons.... again, common sense, but fun to see some numbers to support the theory.

 

[electro's horse]

so if youre not taking playoff appearances into consideration, look at the panthers before those SF and Seattle games, respectively.

before the niners, they were 5-3

before the seahawks, they were 4-0

they were good teams already.

[KB_fan]

9 minutes ago, electro's horse said:

so if youre not taking playoff appearances into consideration, look at the panthers before those SF and Seattle games, respectively.

before the niners, they were 5-3

before the seahawks, they were 4-0

they were good teams already.

In 2013 Panthers were 0-2 against teams that finished the season with a winning record (losses to SEA & ARIZONA) prior to playing SF.  In beating SF we went on to finish 4-0 against teams with a winning record.

In 2015, Panthers only faced one team that finished with a winning record (HOU) in our 1st 4 games.  With SEATTLE win we went on to beat 3 more strong teams (IND, GB, WAS).  Past Panthers history following losses to good teams suggests we may not have won those additional games if we had lost to Seattle.  Certainly the Indy & GB games proved VERY close.  We almost gave them away.  The momentum from the Seattle win, the confidence it gave, may have been the intangible factor in helping the team pull out those key wins.

[electro's horse]

both seattle and arizona were better than the panthers that year.

honestly if you're talking about momentum the panthers probably dont need overtime to beat a poo colts team. 

like i feel like you're trying to find evidence that big wins help the team the next week, or bad losses hurt them. do whateve ryou want in your free time, but i sincerely doubt you're going to find any that would indicate the panthers are more or less susceptible to "big wins" than other teams.

trying to attribute meaning(big wins) to complicated things to very complicated things (football) is like as old a human trait since staring at the stars. 

everyone in the nfl has been playing at the top level of football available to them since they were 14. they dont need "big wins" or whatever.