Einstein

You’re confusing variables and suggestions. The data absolutely suggests what I wrote, but you’re correct that there are other variables that go into it. That being said, the fact that there are so many variables and the results remain so consistent over 40 years is what makes it remarkable. Five Thirty Eight (the polling website) did a similar study that concluded nearly the exact conclusion that I did. They showed that HC/QB combos have 5 years maximum to win their first SB.

Feel free to point out a single instance of statistical wrongness.

A rather solid argument could be made that Kelly would not be a top 5 QB in todays NFL.

Have to admit, I laughed at how pathetic that is.

Thus why I said (in my first response to you) that my opinion on Anarumo is not based on statistics, but my gut feeling watching him. No, this did not happen. I can only give you the information. I can not make you understand it.

Apologies for getting 2 letters in his name wrong lol.

It’s really not a hard concept. Reid got more of a leash because he got much closer to the ultimate goal, numerous times, than other coaches have. Very simple. I can only give you the information. I can not make you understand it.

He was attempting to be deceptive. ”Well, KC kept Reid despite not going to the championship game… ignore the FIVE championship games he was in before. They weren’t “afc” championship game”. And then Augie liked the post lol. If you’re getting that close to the ultimate goal, that many times, teams tend to give more leeway than if, say, keep getting bounced in the divisional round and struggle against 3rd string QB’s. .

Mahomes vs playoff teams without Anuromo: 33.1ppg Mahomes vs playoff teams WITH Anuromo: 23.5ppg That is a MASSIVE difference. Visit the Chiefs forum and you will see a fan base that shudders at the mention of Anuromo’s name. They fear him like we feared Belichick for 2 decades.

NFL teams hire mathematicians and data analysts to run their analysis and statistics. Football professionals (like coaches, GM’s, scouts) know jack squat about data analytics. The Bills data analyst for example, Malcolm Charles, came straight to the team out of college (Clemson and Marquette). Evan Weiss, the teams football analyst, came from PFF (who also has no peer reviewed equations) and before working at PFF was a marketing intern. You seem to have this idea that NFL teams employ “football experts” for data analytics but in reality they are just mathematically inclined data engineers, sometimes straight out of college, that know about as much about football as the people on this forum. Great data analysts can seamlessly move between sectors - from technology, to oil, to manufacturing - because, it doesn’t matter! They don’t need to be an expert to interpret data.

Sometimes you just have to laugh at a post. That was one of them. He wants basic calculus to be peer reviewed because he doesn’t like the results 😂

My desire (with no calculation provided just gut feeling) is Lou Anorumo. 1) I believe it helps the Bills. Anorumo has stopped the Chiefs offense twice in the playoffs. Something we have never done. 2) It hurts the Chiefs. 3) It hurts the Bengals (they lose their best coordinator). It is a win/win/win. Ps, I believe the percentage to be higher than 1% once the replacements get added back in and you look at it on a year by year basis.

Calculus and maths are me treating it like science? Oofta. This is college-level calculus. Not calculating the wave length of light passing into a black hole. You don’t like the results and you are attempting to discredit them by acting as if (mostly) basic math needs to be peer reviewed. It’s hilariously stupid. What has the world come to. We are so stupid that we think the presence of maths needs SME validation. I have applied common, age-old formulas and rules to a data set.

This is a thread on a football forum. Attempting to discredit it because it’s not relevant for a dissertation is just silly. But sure, feel free to send my equations to your math professor friend and ask him if i’m wrong. As for developing a measurement tool and assuring its validity - what you’re describing is the peer-review process. To say that some difficult calculus needs SME’s to peer review the work is literally laughable lol. I wonder how anyone makes it out of college math nowadays. .

Yes we were very lucky in that game. Mostly because Frank Reich is not a great head coach. His decision to go for 4th and goal cost them the game.

It depends on how you define success. The only goal of the NFL is to win a Super Bowl. There is no trophy for making the playoffs or winning the division.

It is actually a very difficult calculation. I have spent the last 2 hours trying to calculate it and there are many bottlenecks but I believe that I am almost there. Here is where I am right now, and I am still refining it: H denotes head coach L denotes Playoffs B denotes Super Bowl S denotes more than 7 years of coaching 𝑛(𝐿) = 14 𝑛(𝐻) = 32 𝑛(𝐵) = 2 Probability of having more than 7 years coaching and making a super bowl is given by: Since the probability of having more than 7 years of coaching experience is 1 (because we are considering only the case of having more than 7 years’ experience), in temporary conclusion we have: But there is a flaw and that is the duplicates. They need to be accounted for and that will drive the number up. .

Lol so you were purposefully being deceptive by talking about when Reid made his first afc championship game on his second team? So… not his first championship game? And completely irrelevant to our discussions in this thread. He had been in FIVE CHAMPIONSHIP games prior to that second team. While we are at, it took Tom Brady over 20 years to make his first NFC Championship game! That totally proves that QB’s should be given 20 years to prove their worth! Makes sense. This is embarrassing for both you and @Augie

I will be honest and say that I worded my response to you very poorly. That’s not your fault. I have since edited it to make it more

But but but… you’re cherry picking stats. But but but… there are other variables. (Just warming you up to the upcoming excuses. Completely invalid excuses but excuses nonetheless). McDermott is also the longest tenured coach without a Super Bowl appearance. Because in the modern era, teams typically fire coaches who haven’t made a Super Bowl by this point in their career (especially with a franchise QB). Not exactly how it works. McDermott does not have a 10% chance. That’s the historical percentage among the 48 coaches. But some made it 1 time and others made it numerous times. .

In my opinion Allen became a franchise QB in year 2, when he had 29 TD’s. But keep in mind (since I sense that this is where you’re going) that the data reflects total years. Not years since a coach gets a franchise QB. My Belichick comment was simply explaining the variable.

It’s sad that math is so elusive nowadays that it’s seen as foreign and copied. We know the historical percentage of coaches who made it past year 7. My post above spelled this out for you already, but given that only 5 coaches in the last 40 years have made it to the SB past year 7 of their coaching career, the simple probability historically = (Number of coaches past year 7 who made a Super Bowl) / (Number of Super Bowl coaches). 5 / 48 or approximately 0.1042 (10.42%). but this doesn’t take into account some coaches made it several times, and other coaches made it one time and there are more than 48 opportunities but because of the duplicate coaching data it doesn’t take that into account.

And the data shows that McDermott likely never will.

Actually, that’s exactly what i’m doing (on a simple level). Do you not understand how probability works? I’ll teach you. It’s simple math. For example, the probability of landing a heads or a tails on a two sided coin in 50%. Because there are two faces and 1/2 = 0.5. Or how about the probability of rolling a 6 on a six-sided die? The total number of outcomes is 6 (the die can land on 1, 2, 3, 4, 5, or 6). The number of favorable outcomes is 1. So the probability is P = 1 / 6 ≈ 0.1667. I’m sure you’ll say “but variables”, and you can find conditional probability with variables as well. For example… say Josh Allen completes a pass 65% of the time (making it up). However, his completion rate increases to 75% when the Bills are in the lead and we lead in 40% of our games. So, how can we figure out the probability the Bills are winning the game if Allen completes a pass? Well, we can define L as the Bills currently having the lead, P as Allen just completing a pass, and we want to find P(L | P)… (the probability the Bills are leading when Allen just completed the pass). We know from our data set that P(L) = 0.40 ... P(~L) = 1 - P(L) = 0.60 (probability the Bills are losing). P(P | L) = 0.75, the probability of Allen just completing a pass given the Bills are in the lead. P(P | ~L) = 0.65, the probability of Allen completing a pass given the Bills are losing or tied. Then using the law of total probability we can find Allen’s total probability of completing a pass in ANY scenario (winning or losing): P(P) = P(P and L) + P(P and ~L) = P(P | L) * P(L) + P(P | ~L) * P(~L) = 0.75 * 0.40 + 0.65 * 0.60 = 0.30 + 0.39 = 0.69. Lastly you can use Bayes theorem to find the probability the Bills are leading when Allen just completed that pass: P(L | P) = [P(P | L) * P(L)] / P(P) = (0.75 * 0.40) / 0.69 = 0.434. So, if Allen just completed a pass (given the data set - and my data set was much larger), there is about a 43.4% chance the Bills are currently leading in the gwmeZ. I think. This is rich, coming from a poster that likes posts full of factual inaccuracies (literal statistical wrongs) because they have a (wrong) conclusion that agree with his own. You don’t care about facts - you want your feelings placated.

Yep. My analysis of the data is strikingly accurate with all of the other (similar studies) done on this topic. I understand @GunnerBill assertion that it doesn’t prove causation, and he is right that there is no absolute, but the evidence is so darn consistent that you would really have to squint to pretend that it’s not alarming.