Jump to content
Carolina Huddle
  • Hey There!

    Please register to see fewer ads and a better viewing experience:100_Emoji_42x42:

1of10Charnatives

Anyone here have experience with statistical hypothesis testing?

Recommended Posts

I’ve got a question I wanna look at using hypothesis testing but although I had some exposure to it for my degree that was many moons ago and tbh my grade in business statistics reflected that it was one of the subjects I had the weakest grasp of any I studied. If anybody’s even decent at that stuff I could use a hand not screwing it up. Post here or pm me.

Oh and before you ask the question relates to running qb’s as they age, so not remotely relevant to the Panthers.

  • Pie 2

Share this post


Link to post
Share on other sites


Basically a lot of people tend to go with the assumption that when running qb’s stop running or run a lot less, they become less effective overall. When I looked at running qb’s career statistics, those that played long enough to make it into their 30’s did run markedly less from about 30 onward, but they were also able to keep playing into their mid 30’s. (a fact I found mildly  surprising and has caused me to reevaluate how viable I thought keeping Cam is as an option going forward.)

What I want look at beyond just eye balling numbers is whether the decrease in running impacted their passer rating negatively or not. I think it’s possible that as these qb’s matured, their experience enabled them to largely compensate for the need to shift away from running.

statistical hypothesis testing is a math based way to answer that question. A question that it might be useful to know about what it would be reasonable to expect from Cam going forward barring injury.

 

  • Pie 2

Share this post


Link to post
Share on other sites

I was just dangerous enough to help maybe 2 decades ago.

But I would start running some liner regressions and see what pops out.

Share this post


Link to post
Share on other sites
41 minutes ago, 1of10Charnatives said:

Basically a lot of people tend to go with the assumption that when running qb’s stop running or run a lot less, they become less effective overall. When I looked at running qb’s career statistics, those that played long enough to make it into their 30’s did run markedly less from about 30 onward, but they were also able to keep playing into their mid 30’s. (a fact I found mildly  surprising and has caused me to reevaluate how viable I thought keeping Cam is as an option going forward.)

What I want look at beyond just eye balling numbers is whether the decrease in running impacted their passer rating negatively or not. I think it’s possible that as these qb’s matured, their experience enabled them to largely compensate for the need to shift away from running.

statistical hypothesis testing is a math based way to answer that question. A question that it might be useful to know about what it would be reasonable to expect from Cam going forward barring injury.

 

giphy.gif

  • Flames 1

Share this post


Link to post
Share on other sites
4 minutes ago, Wes21 said:

giphy.gif

We should expect Cam to run less from this point on based on other successful running qb’s. Math offers a way to determine whether we should expect that to make him less effective as a passer instead of just guessing.

Share this post


Link to post
Share on other sites

You’re not going to have enough data to get a statistically significant result.

  • Pie 3

Share this post


Link to post
Share on other sites
3 minutes ago, Tbe said:

You’re not going to have enough data to get a statistically significant result.

Do you remember how big a sample size is needed? I think there’s about 8 to 10 guys you could use as a sample, but I don’t recall what the necessary threshold for statistical significance would be.

Share this post


Link to post
Share on other sites

I do. I am probably a bit rusty since a most of my work does not involve this sort of detail. However, you have proved time and time again to disregard facts and have personally insulted me several times so I will not be offering you any help. Are you one of those dook fans who didn’t go to dook?

cheers! 

  • Poo 1

Share this post


Link to post
Share on other sites
29 minutes ago, 1of10Charnatives said:

Do you remember how big a sample size is needed? I think there’s about 8 to 10 guys you could use as a sample, but I don’t recall what the necessary threshold for statistical significance would be.

But what's the criteria to judge a QB on?  Yards/TD/INTs?  Rating?  QBR?  Wins?  Completion %?  Some abstract efficiency rating from a statistics site?  How do we decide a guy is a "running QB" or not?  What's the criteria for that?

I think its trying to add math and science to something that just needs to be discussed with fluid criteria, simply taking it player by player.  Start with Russell Wilson...he's been more effective as a QB.  BUT, his Super Bowls came back when he was less effective and cost less.

 

  • Poo 1

Share this post


Link to post
Share on other sites
48 minutes ago, 1of10Charnatives said:

Do you remember how big a sample size is needed? I think there’s about 8 to 10 guys you could use as a sample, but I don’t recall what the necessary threshold for statistical significance would be.

More than 8-10. You are looking at age and QBR I guess? At those low numbers one guy with a great or terrible record after 30 will have a major impact on the entire model. You need hundreds of QB’s to make it significant.

Share this post


Link to post
Share on other sites
58 minutes ago, Wes21 said:

But what's the criteria to judge a QB on?  Yards/TD/INTs?  Rating?  QBR?  Wins?  Completion %?  Some abstract efficiency rating from a statistics site?  How do we decide a guy is a "running QB" or not?  What's the criteria for that?

I think its trying to add math and science to something that just needs to be discussed with fluid criteria, simply taking it player by player.  Start with Russell Wilson...he's been more effective as a QB.  BUT, his Super Bowls came back when he was less effective and cost less.

 

Every thread bro - you say a lot, but you don’t really say anything at all. Bunch of words that don’t really mean anything. Bet you think you are super smart. 

  • Beer 1
  • Poo 2

Share this post


Link to post
Share on other sites
2 minutes ago, 4Corners said:

Every thread bro - you say a lot, but you don’t really say anything at all. Bunch of words that don’t really mean anything. Bet you think you are super smart. 

Cat meet Kettle.  LOL.  

Share this post


Link to post
Share on other sites
15 minutes ago, Shocker said:

Cat meet Kettle.  LOL.  

That is not the expression, Einstein. 

Share this post


Link to post
Share on other sites
1 hour ago, Tbe said:

More than 8-10. You are looking at age and QBR I guess? At those low numbers one guy with a great or terrible record after 30 will have a major impact on the entire model. You need hundreds of QB’s to make it significant.

Significant is a big word for studies like this come one on man it can still be done and interesting stuff can be found. Tons of studies with viable information weren’t enough to be ‘significant’.

We can still find trends which is something.

Sorry I always hated the word significant when crunching numbers. To me it’s the statistical ‘black and white’ the ‘perfect answer to causation’. But in reality it’s just too complex for that and you can always rip apart the numbers with just one variable or another if you really want to find an explanation or prove an agenda.

Just run the numbers and show the trends and explain possible causations and the reasons behind it all including possible fallacies and let the audience decide what they want to think and where their brain goes with it.

It will still be interesting.

Edited by onmyown

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.




×
×
  • Create New...