My thread versus downundermike

[jsb235]

This is the war you have been waiting for - arguably the biggest defender of Jalen Hurts versus arguably his biggest hater.

The battle is over qb rating.

And off we go.

The basics - my contention is that Jalen Hurts had his best qb ratings over the final 3 games of the season, when he registered marks of 110, 103 and 91. Downundermike claims his ratings were better over a three-game stretch earlier in the season when he registered marks of 86, 115 and 103.

His argument is that you have to use raw data from those three games (attempts, completions, yardage, TDs and INTs) to get an accurate overall number for those three games.

I disagree.

Here is my reasoning.

The definition of a quarterback rating is "a statistic that represents a quarterback's overall performance."

After each game, a player gets a number. That number represents how that player did in that particular game.

Therefore, if a player scores a 75, a 100 and a 125, it is assumed his performance was about a 100 over three games.

That is accurate, because the whole point was to create a snapshot of how a player did in a particular point of time. It was introduced to eliminate outliers and also to compare qb stats in real time.

So, for example, if a qb scored a 75, a 75 and a 150 over three games, the qb rating would reflect that he had two bad outings and one great one, and not overinflate the one great outing. His rating would be the average of those three performances, or 100. Again, the rating is a number that reflects performance based on stats, not simply the stats themselves.

But Downundermike wants to change the rules. Let's say, in the above example, the qb had two games where he attempted 20 passes, completed 10, for 180 yards, had 2 TDs and 2 INTs. In those two games he would register a rating of 75. And then he had a game where he registered an almost perfect passer rating. He went 20 of 30 for 400 yards and four TDs with no interceptions. That is a passer rating of 150.

The way I calculated his rating (based on performance and not pure stats) is that he had a somewhat decent rating of 100. But under Downundermike's calculation, his rating would be 109.2, a significant increase.

But enough hypotheticals, let's examine Jalen Hurts.

In the last three games of the season, he had the aforementioned ratings of 110, 103 and 91.

Over that period he completed 54 passes out of 81 attempts for 709 yards and three TDs and one interception.

In the period where Downundermike claims he was better, he completed 36 passes out of 54 attempts for 443 yards, with three TDs and one INT.

Downundermike wants to just plug raw data to generate one number for those three games. But remember, the whole point is that the rating is supposed to be a reflection of a qb's performance in a set time frame, in this case one game. If we are comparing his performance on a per-game basis over three games, we have to use each rating as a reflection of that performance.

Passer rating is not a perfect statistic. But it becomes less perfect when you use it in a way that it wasn't intended to be used. It is a reflection of how a qb performs over a set period of time. In this case, the parameters were his per game performance over three games. Therefore, he was clearly as good over the last three games of the season as he was in the three-game stretch cited by Downundermike, and not significantly worse, as our antagonist has claimed.

As an aside, anyone agreeing with downundermike needs a basic math refresher course. 

 

[Guest]

You should add a poll as to who is winning.  I bet I take the lead before i actually respond.

I am on the recliner with my sleeping granddaughter, so you will have to wait to catch these hands.

[jsb235]

Just now, downundermike said:

You should add a poll as to who is winning.  I bet I take the lead before i actually respond.

I am on the recliner with my sleeping granddaughter, so you will have to wait to catch these hands.

A poll? Of what, how many morons post here?

Good luck. You need it.

[Guest]

15 minutes ago, jsb235 said:

The way I calculated his rating (based on performance and not pure stats) is that he had a somewhat decent rating of 100. But under Downundermike's calculation, his rating would be 109.2, a significant increase.

I will start with this gem.

We are debating how you calculated the rating vs how the NFL calculates the rating.

I will dive in once I can get to my PC, but it is basically over right here.

[ilikepargo]

5 minutes ago, downundermike said:

You should add a poll as to who is winning.  I bet I take the lead before i actually respond.

I am on the recliner with my sleeping granddaughter, so you will have to wait to catch these hands.

So you're old, huh?  (Me too.)

I"m not a Hurts fan.  I hated the pick from the moment BigEd told us that it was about to be announced.  What I've seen of Hurts so far is that he's been unimpressive as a reader of defenses and inconsistent as a passer.

But I want him to do well.  He's the Eagles QB, so I want him to develop and to succeed.  I want him to prove me wrong.

[Guest]

22 minutes ago, jsb235 said:

As an aside, anyone agreeing with downundermike needs a basic math refresher course

This is your opinion, and those who agree with me know how QB rating is calculated, and also understand to get the full average of multiple groups of data, you can’t average the averages of those groups of data, you have to compile the data as a whole and then calculate the average.

[jsb235]

2 minutes ago, downundermike said:

I will start with this gem.

We are debating how you calculated the rating vs how the NFL calculates the rating.

I will dive in once I can get to my PC, but it is basically over right here.

I thought it was over once we got to the  point in the discussion where your best argument was "Jalen Hurts was only really good at the end of the season, but actually played slightly better earlier in the season," but I guess here we are. 

[Guest]

3 minutes ago, jsb235 said:

I thought it was over once we got to the  point in the discussion where your best argument was "Jalen Hurts was only really good at the end of the season, but actually played slightly better earlier in the season," but I guess here we are. 

That is not what I said.  You said his best 3 game stretch was his last 3 games, I have proven that is his 3rd best 3 game stretch.

Your entire post is weak and full of holes, and I am going to rain hell fire once I get to my PC.

[jsb235]

Just now, downundermike said:

That is not what I said.  You said his best 3 game stretch was his last 3 games, I have proven that is his 3rd best 3 game stretch.

Your entire post is weaker  and full of holes, and I am going to rain hell fire once I get to my PC.

Good luck. Just remember, qb rating is a measure of performance, not a raw stat. So don't make the mistake of saying "averages of averages is not average" because that is a fundamental error.

[Aerolithe_Lion]

Passer rating is the most antiquated and misinformed analytic stat in all of American sports. A quarterback could have a superior day passing the football over his opposing QB, and yet still end up with a worse passer rating. It ignores so much, and puts improper weight on a limited number of areas.

Deshaun Watson has the second greatest passer rating of all time, while Tony Romo, Dak Prescott, and Kirk Cousins round out the top 10. arguably none were even top 5 QBs at any point in their careers, much less all time.

The only thing it helps track is who has trouble discerning QB talent without using it.

[Guest]

6 minutes ago, ilikepargo said:

So you're old, huh?  (Me too.)

I"m not a Hurts fan.  I hated the pick from the moment BigEd told us that it was about to be announced.  What I've seen of Hurts so far is that he's been unimpressive as a reader of defenses and inconsistent as a passer.

But I want him to do well.  He's the Eagles QB, so I want him to develop and to succeed.  I want him to prove me wrong.

48, 4 year old granddaughter.

This entire topic is not about whether Hurts is good or bad, this is a discussion about his best 3 game stretch last year.  When someone pointed out that teams had figured out how to shut him down, jsb said his best 3 game stretch in regards to QB rating was the last 3 games, to which I have already proven that Hurts had two other 3 game stretches that where better.

The rest of this is jsb doing mental gymnastics and changing how the NFL calculates statistics.

[jsb235]

Just now, downundermike said:

jsb said his best 3 game stretch in regards to QB rating was the last 3 games, to which I have already proven that Hurts had two other 3 game stretches that where better.

Again, you fail to understand the qb rating is a measure of performance, not a raw stat.

It's a snapshot on how a qb did in a certain timeframe. So if a qb had ratings of 86, 110 and 90, you wouldn't take the raw stats from each game and generate a new rating, but use the ratings from each game to generate an average performance over that time frame.

You want to simply calculate a new rating based on raw stats, but that's not the way qb rating works. Once the number is calculated, it becomes the number. You can't undue it simply to win an argument.

Also, as an aside, I didn't want to use qb rating, but it was in response to someone else who demanded it be used. So I only used it to keep the discussion on a level playing field.

[ilikepargo]

45 minutes ago, downundermike said:

48, 4 year old granddaughter.

This entire topic is not about whether Hurts is good or bad, this is a discussion about his best 3 game stretch last year.  When someone pointed out that teams had figured out how to shut him down, jsb said his best 3 game stretch in regards to QB rating was the last 3 games, to which I have already proven that Hurts had two other 3 game stretches that where better.

The rest of this is jsb doing mental gymnastics and changing how the NFL calculates statistics.

Hmmm...  So basically I wasn't paying attention.  Not the first time.  Won't be the last either.

And this thread is less meaningful than I thought.

And you're a young punk!!!  

[LeanMeanGM]

52 minutes ago, jsb235 said:

Again, you fail to understand the qb rating is a measure of performance, not a raw stat.

It's a snapshot on how a qb did in a certain timeframe. So if a qb had ratings of 86, 110 and 90, you wouldn't take the raw stats from each game and generate a new rating, but use the ratings from each game to generate an average performance over that time frame.

You want to simply calculate a new rating based on raw stats, but that's not the way qb rating works. Once the number is calculated, it becomes the number. You can't undue it simply to win an argument.

Also, as an aside, I didn't want to use qb rating, but it was in response to someone else who demanded it be used. So I only used it to keep the discussion on a level playing field.

How do you think each QB’s passer rating is calculated every season? 

[Aerolithe_Lion]

10 minutes ago, LeanMeanGM said:

How do you think each QB’s passer rating is calculated every season? 

[jsb235]

27 minutes ago, LeanMeanGM said:

How do you think each QB’s passer rating is calculated every season? 

Is that the same as comparing two 3-game stretches during a season? Because it seems a lot different to me. 

But, okay, if i have seasonal qb ratings of 75, 75 and 150, what is my average seasonal qb rating? If you can answer that question, you proved my point. 

[opa-opa]

Tldr

[Swoop]

I am now inspired to now make a thread calling @LeanMeanGM out for... something

[EagleJoe8]

So, if I’m understanding the argument correctly, it appears the discrepancy between Mike and JSB is based on how the calculation is done when looking over a multiple game stretch.

JSB is saying the passer ratings are calculated by adding the ratings and dividing them out by the number of ratings in question. In other words, the same way you would find an average in any other scenario. 

Mike is saying you have to tally the qualifying statistics for the stretch of games you’re looking at, and re-calculate. 
 

If I’m wrong, ignore anything else beyond this point, and tell me to go post more boobs in the boob thread. 
 

If I’m correct, then it seemed to me the most simple way to confirm this, would be to look at their ratings over a full season and see if they match up over a full season using the normal method of finding averages. Using the method claimed by Mike would require more leg work, but could still be done of course. 
 

I tried this out for 5 different QBs. I added up the passer ratings over a full season, and divided by the number of games they played. Hurts of course played 15 games, the others played all 17. The info I used for all 5 was from espn.com. I wrote down their calculated ratings vs their listed ratings. Here’s the screen shot:

As shown, there is a slight discrepancy between the calculated rating vs the listed rating. That could be because I’m using a much larger sample size than 3 games, and the numbers get closer with having more data. I did notice with only 3 games listed in the Blog, the numbers seemed a bit further apart. The bottom line is the numbers didn’t match exact. Some calculated higher than their listed rating, some listed lower. Unless I’m missing something, I’m curious to know why they don’t match exact. 

[NCTANK]

i think we all want Hurts to succeed but most of us have serious doubts if he ever will. no argument or rating is gonna prove anything to me. imo, he has this year to EARN it, on the field, with his arm. 

period

[time2rock]

Yawn

[EaglePhan1986]

Oh yeah I’m really going to read all of that. You need laid, bad… 

[AngryBird]

Someone get Dana White on the phone so you two can settle this like real men on the undercard. Mandalay Bay Events Center.

 

We areee LIVVVEEEEEEE!

[HazletonEagle]

If anyone knows Mike they know he didn't even sleep last night because he was preparing his rebuttal.

OP is going down when mikes granddaughter is done pooping on his lap.

[Bwestbrook36]

Lame