Bangarang Posted June 29, 2019 Posted June 29, 2019 Just now, oldmanfan said: I get it. The people that misinterpret stats don't. No, you clearly don’t get it. If we’re talking about Allen’s low completion percentage in relation to the rest of the league then you can’t simply use his adjusted completion percentage to prove everything is fine without comparing it to the adjusted completion percentages of his peers. It’s not misinterpretation if you’re asking for an equal comparison.
oldmanfan Posted June 29, 2019 Posted June 29, 2019 22 minutes ago, Foxx said: talk about not addressing a question presented. wow. I don't think adjusted completion rate is particularly useful to assess a QB. There are many variables such as protection that can lead to hurried throws, throwaways etc. that have nothing to do with a QBs performance. In fact, if you have a QB that throws it away instead of taking a sack his adjusted rate could be lower for doing the right thing. 2 minutes ago, Bangarang said: No, you clearly don’t get it. If we’re talking about Allen’s low completion percentage in relation to the rest of the league then you can’t simply use his adjusted completion percentage to prove everything is fine without comparing it to the adjusted completion percentages of his peers. It’s not misinterpretation if you’re asking for an equal comparison. Where did I say it's fine? I have indicated where Allen needs to improve. And I can do that watching him without the benefit of an overinflated poorly constructed statistic.
Cripple Creek Posted June 29, 2019 Posted June 29, 2019 8 minutes ago, oldmanfan said: Explain to me how comparing a first year QBs stats to a HOF QB with vastly more experience, with different receivers, different O line, coaching and all the other variables makes sense. I can watch Allen on the Bills last year, and I watched every play over his last three games, and I can tell you what he needs. He needs to make his reads more quickly, to get the game to slow down. He needs to improve his ball placement (which is more precision than accuracy) especially on shorter throws. He needs to improve his footwork throwing to the right. He needs better receivers and better protection. Prett much all these are what you'd expect from a young QB. Comparing him to a HOFer as if he should be at that level today is pointless and achieves nothing. You’re the one setting parameters, not me. How about comparing the other rookie QBs? See any value in that? What about first year starters? Any value there? And, while we’re at it WTF is wrong with comparing to the best that are out there? How else do you know how far you have to go? If you don’t think Allen studies other QBs then you’re sadly mistaken. i think there’s someone for you to chase off your lawn
Foxx Posted June 29, 2019 Posted June 29, 2019 (edited) 10 minutes ago, oldmanfan said: I don't think adjusted completion rate is particularly useful to assess a QB. There are many variables such as protection that can lead to hurried throws, throwaways etc. that have nothing to do with a QBs performance. In fact, if you have a QB that throws it away instead of taking a sack his adjusted rate could be lower for doing the right thing. ... your replies to and accounting from my post where it was said that without something to measure against, you can't possibly know how any stat rates. i would think that you of all people should understand this. that you are avoiding answering such an obvious question is absurd. Edited June 29, 2019 by Foxx
oldmanfan Posted June 29, 2019 Posted June 29, 2019 1 minute ago, Foxx said: your replies to and accounting from my post where it was said that without something to measure against, you can't possibly know how any stat rates. i would think that you of all people should understand this. that you are avoiding answering such an obvious question is absurd. The questio was how does one know the value of adjusted completion rate if you don't compare between individuals. And my answer is simple: if the star has limited value if any, as I would suggest, then comparisons are meaningless. Yiu are correct that stats give you a sense of trends in a group and a basis for comparison, but when you oversimplify a stat and don't take into account the myriad of variables that affects said statistic then it becomes of little consequence. Have you ever designed a study? Figuring out and controlling variables is one of the key things you have to do before determining the value of what you're going to measure
Bangarang Posted June 29, 2019 Posted June 29, 2019 15 minutes ago, oldmanfan said: Where did I say it's fine? I have indicated where Allen needs to improve. And I can do that watching him without the benefit of an overinflated poorly constructed statistic. Are you even paying attention to the discussion right now? I don’t think you are 9 minutes ago, Foxx said: your replies to and accounting from my post where it was said that without something to measure against, you can't possibly know how any stat rates. i would think that you of all people should understand this. that you are avoiding answering such an obvious question is absurd. If Allen’s adjusted completion percentage was among the best, it would matter. Since it isn’t, it’s irrelevant. That’s how these things tend to work around here. 2
oldmanfan Posted June 29, 2019 Posted June 29, 2019 (edited) 3 minutes ago, Bangarang said: Are you even paying attention to the discussion right now? I don’t think you are Yes I am. I don't think you understand the basis of statistical analysis correctly. You claimed I am using my argument to say everything is fine with Allen. Which I never said, and in fact I pointed out areas he needs to improve. You are the one being disingenuous here. Edited June 29, 2019 by oldmanfan
Bangarang Posted June 29, 2019 Posted June 29, 2019 (edited) 1 hour ago, oldmanfan said: Yes I am. I don't think you understand the basis of statistical analysis correctly. You’ve come to the conclusion that I lack an understanding because I want to see an equal comparison? It’s shocking how dense you’re being right now. Edited June 29, 2019 by Bangarang
Foxx Posted June 29, 2019 Posted June 29, 2019 3 hours ago, Foxx said: most of this thread is, 'unwatchable'. i reiterate. 1
Cripple Creek Posted June 29, 2019 Posted June 29, 2019 1 minute ago, Foxx said: i reiterate. I concur and I am ashamed to have added to it.
oldmanfan Posted June 29, 2019 Posted June 29, 2019 17 minutes ago, Bangarang said: You’ve come to the conclusion that I lack an understanding because I want to see an equal comparison? It’s socking how dense you’re being right now. 17 minutes ago, Bangarang said: You’ve come to the conclusion that I lack an understanding because I want to see an equal comparison? It’s socking how dense you’re being right now. But it's not an equal comparison because there are a lot of variables between teams that in turn can affect the stat you're looking at, independent of the QBs actual play. I have had graduate level course work in stats. If you have as well then let's have an intelligent discourse on this. The fact that you have not responded to the difficulties with this stat despite my pointing them out several times indicates you don't understand statistical analysis. Prove to me you do and we can talk.
Bangarang Posted June 29, 2019 Posted June 29, 2019 33 minutes ago, oldmanfan said: You claimed I am using my argument to say everything is fine with Allen. Which I never said, and in fact I pointed out areas he needs to improve. You are the one being disingenuous here. No, I’m not. That was in reference to the person who brought up Allen’s adjusted completion percentage. You keep proving you aren’t paying attention.
Foxx Posted June 29, 2019 Posted June 29, 2019 37 minutes ago, oldmanfan said: The questio was how does one know the value of adjusted completion rate if you don't compare between individuals. And my answer is simple: if the star has limited value if any, as I would suggest, then comparisons are meaningless. Yiu are correct that stats give you a sense of trends in a group and a basis for comparison, but when you oversimplify a stat and don't take into account the myriad of variables that affects said statistic then it becomes of little consequence. Have you ever designed a study? Figuring out and controlling variables is one of the key things you have to do before determining the value of what you're going to measure one can pretty much tailor any statistical polling to get it to say what they are wanting, depending upon the variables used. however and this is my last word on the subject for now unless this takes a turn for the more intelligent discussion, without something to measure any stat against, said stat is completely useless for purposes of having an understanding of what that stat means. in the 70's a 50% completion percentage, when weighed against peers was considered good. you take that same stat and today it is garbage because of the current context of other peer stats. 1
Bangarang Posted June 29, 2019 Posted June 29, 2019 15 minutes ago, oldmanfan said: But it's not an equal comparison because there are a lot of variables between teams that in turn can affect the stat you're looking at, independent of the QBs actual play. I have had graduate level course work in stats. If you have as well then let's have an intelligent discourse on this. The fact that you have not responded to the difficulties with this stat despite my pointing them out several times indicates you don't understand statistical analysis. Prove to me you do and we can talk. Can you spot a receiver dropping the ball? Or a QB throwing the ball away because he was pressured or spiking it to stop the clock? if you understood what went into figuring out adjusted completion percentage you would know it doesn’t take an advanced understanding of each team’s specific scheme, play call or route concepts. Are you trying to discredit such a simple concept because it doesn’t yield the result you want it to? Genuinely asking.
oldmanfan Posted June 29, 2019 Posted June 29, 2019 (edited) 1 hour ago, Foxx said: one can pretty much tailor any statistical polling to get it to say what they are wanting, depending upon the variables used. however and this is my last word on the subject for now unless this takes a turn for the more intelligent discussion, without something to measure any stat against, said stat is completely useless for purposes of having an understanding of what that stat means. in the 70's a 50% completion percentage, when weighed against peers was considered good. you take that same stat and today it is garbage because of the current context of other peer stats. True. The question though is whether a stat is worth comparing. 1 hour ago, Bangarang said: Can you spot a receiver dropping the ball? Or a QB throwing the ball away because he was pressured or spiking it to stop the clock? if you understood what went into figuring out adjusted completion percentage you would know it doesn’t take an advanced understanding of each team’s specific scheme, play call or route concepts. Are you trying to discredit such a simple concept because it doesn’t yield the result you want it to? Genuinely asking. No. I think I understand what goes into the calculation. And for me some of the variables are more team related vs. QB related which is why I don't give it much relevance to assessing QBs. And quit telling me I'm looking for a specific result. I'm not. I would argue it's the opposite. Edited June 29, 2019 by oldmanfan
Bangarang Posted June 29, 2019 Posted June 29, 2019 (edited) 31 minutes ago, oldmanfan said: No. I think I understand what goes into the calculation. And for me some of the variables are more team related vs. QB related which is why I don't give it much relevance to assessing QBs. Can you elaborate on what you mean? Quote And quit telling me I'm looking for a specific result. I'm not. I would argue it's the opposite. Well, I never said you were. I merely asked given how opposed you are to such a comparison. Edited June 29, 2019 by Bangarang
oldmanfan Posted June 29, 2019 Posted June 29, 2019 15 minutes ago, Bangarang said: Can you elaborate on what you mean? Well, I never said you were. I merely asked given how opposed you are to such a comparison. First, your quote: Are you trying to discredit such a simple concept because it doesn’t yield the result you want it to? . So yes you did say I'm looking for a result. Yield the result you want it to. Your words. Second if you look at the denominator for the adjusted completion percentage it has things like throwaways, drops, and such. If you have WRs that can't get open as well as the next team, or a line that doesn't protect as well, then the denominator increases and the percentage decreases. And those are more on different parts of the offense other than the QB. So you could have a QB that has to throw the ball away more because of poor line play, and his ACP would be lower. In such a case the QB is making the smart play but his stat suffers. Thus the ACP to me may have usefulness as a team stat, but not a QB stat.
transplantbillsfan Posted June 29, 2019 Author Posted June 29, 2019 6 hours ago, Bangarang said: No, you clearly don’t get it. If we’re talking about Allen’s low completion percentage in relation to the rest of the league then you can’t simply use his adjusted completion percentage to prove everything is fine without comparing it to the adjusted completion percentages of his peers. It’s not misinterpretation if you’re asking for an equal comparison. Which was precisely the reason I compared him with 6 other rookie QBs rather than 6 veteran HOF QBs 1
transplantbillsfan Posted June 29, 2019 Author Posted June 29, 2019 6 hours ago, Bangarang said: If Allen’s adjusted completion percentage was among the best, it would matter. Since it isn’t, it’s irrelevant. That’s how these things tend to work around here. Well, Allen threw the 2nd highest percentage of catchable passes excluding Throwaways among the 7 rookie QBs I compared.
Recommended Posts