My personal opinion would be something along the lines of what Pauly said, where we start with Tannehill’s ability to produce a season next year that’s significantly better than his typical one has been historically. In response to someone else’s question, it’s entirely possible that could happen; I’m certainly not ruling that out. But at the same time realize that we are dealing with a player who experienced multiple sets of surroundings during his career, prior to the present time, and who never played this way in any of those surroundings. For me that begs the question of whether these particular surroundings can be sustained to the degree that they produce a new norm in Tannehill’s performance.
So, then when a poster points out Thill is about to top the list of historically great seasons in terms of QB rating, that poster is to be held to higher standards than the league of that sport and the worldwide leader in sports, all because you decided they should?
Yet you can't seem to tease apart that despite the different staffs, they were all terrible? Which HC he had is still a head coach today? Zach Taylor is awful in Cinci. How many of his OCs are still OCs in the league? How many players left Miami over Tannehill's tenure, and had success with other teams that they didn't experience here? Just saying he had multiple sets of surroundings doesn't actually tell anyone anything. You're not providing any actual proof. You say he had multiple sets of surroundings, so that means he's at fault. I'm saying, despite different regimes, Tannehill had one surrounding:. BAD surroundings.
It takes time for more accurate statistics (as in statistics that more accurately reflects player ability) to make its way into common usage. Best example I can think of in any team sport is WAR = Wins Above Replacement, which has totally transformed how players are compared across eras in baseball and is the reason a player like Mike Trout can repeatedly win the MVP (including this year) even when on a losing team, which was nearly impossible before. So it takes time, and someday no one will be using raw passer rating to compare. Yes, but you want to determine that through a stat that isn't inherently biased due to passer rating inflation. I have no idea, but you guessed wrong. I'd say I had no particular emotion when I typed that. Absolutely. Things don't improve if you don't try to improve on things.
Getting people to use z-scores instead of raw ratings is definitely improving on the statistical analysis. And none of that would have happened without me constantly pushing this. So yeah.. it's me "making something happen" that is quite useful for comparing across eras.
No it isn't. The point the poster made was the point they wanted to make. You want them to make a different point. The point you want them to make is not their responsibility. - They point they wanted to make is Thill playing so good his numbers this year are extremely high. - The point you want him to make is that Thill's numbers aren't the best ever based on modified historical data because you've always found ways to temper any enthusiasm around Thill. This is normal human interaction stuff and you want it be graphed, and therein lies the problem.
If I were his agent I would’ve advised he stayed in Pitt for slightly less money. When you are (at the time) the best deep Wr in the game? You can’t take for granted the guy getting you the ball. Deep guy’s and longevity aren’t very common, unless you are Ted Ginn of course.
It's pretty normal human interaction stuff here.. I mean I am a human. Regardless, there's value to improving on statistical analysis and when a post either explicitly or implicitly suggests a comparison across eras using a misleading statistic I'll seriously consider pointing it out. Like I said, I don't do this all the time, but it needs to be done often enough to slowly change things. Anyway.. no point arguing dude. You're not going to convince me to stop trying to improve the statistical analysis used here. I'll keep posting the way I post whether you like it or not.
What is the correct historical statistics to back up the bold? Gonna need that graphed or someone will tell you were wrong and you meant something different and that will be your responsibility, as I've just learned.
It proves both points. He still has to catch it. Wallace was also the worst I have ever seen at keeping track of the sideline and not needlessly stepping out of bounds.
Dude #1 - It's raining. Dude #2 - It can't be raining, my statistical model of the weather for the last 2000 years clearly says that there is a 0.0000000001% chance of rain today. Dude#1 - Stick your head out of the window, it's raining. Dude #2 - (sticking his head out the window) The z score is so low that is a statistically insignificant amount of rain, so statistically, it is not raining. Dude #1 - Your HEAD is WET! Dude #2 - But this isn't a historic amount of rain.
Just be blunt, what are you trying to say? Obviously he did, what are you suggesting? Ahhh, you are referring to Ginn. Yes, yes he did. And it is a great example of how 90% of players need systems to succeed. He would’ve never been even serviceable if he had been with the Dolphins. Kind of the same with Tannehill, he wouldn’t have been able to be successful here the way the team has been constructed. Good for all parties involved to change the scene.
Almost all of them bad.... when they weren't bad, he produced passer rating that averaged over 100 and winning % high enough to make the playoffs.
You and another stat guy on here love that stuff. Most of us disagree, but have stopped discussing it with you guys because you simply ignore what we say, and repeat your stats based conclusions that agree with the premise you started with.
This is the key, especially with members of the OL. Clabo, McKinney, Thomas, Douglas, Martin, etc etc etc
Ah.. but resnor that z-score stuff is just a descriptive statistic. There's no "premise" involved in using z-scores. It's just literally describing to you how many standard deviations above/below the mean that person was. Now.. whether someone is using stats to just support a premise they previously had is another story. I don't think you'll find many posts of mine where I do that. Most of what I'm posting is data-driven analysis, not hypothesis-driven. In fact, most of the hypothesis-driven usage of stats comes from the pro-Tannehill crowd.
In that case they don't need to. But talking about (potentially) having the highest passer rating for any year IS comparing across eras. However.. there is still a utility to z-scores even if you look at a single year because that allows you to calculate the probability of that z-score occurring while raw passer rating won't allow that.
No it isn't. Not if you don't make any claim that setting the record proves they are better than QBs in other eras.
The claim doesn't have to be made explicit. The comparison is across eras. That is.. where that person ranks is a comparison across eras. Your analogy doesn't even apply to the current discussion. If it's raining no statistician would say it's not raining. You created a strawman.
This isn't a question of opinion. It is a fact that no statistician would be successful (in statistics) at arguing "it's not raining" if it is raining regardless of the predicted probability of rain. And it's also a fact that the comparison of ratings was across eras.
His intent is beside the point. When the comparison itself is across eras you shouldn't use a statistic biased by era.