Discussion in 'NFL Football Forum' started by PatriotSeven, Oct 1, 2013.
In regards to SB39 and PatriotSeven
Ok, first of all, good so were on the same page there and I'm not sure what you're ridiculous point is. You're telling me something I already know, and we clearly agree on.
But to to answer your question, with the probability being ONLY 1.5%, of hitting that string of 6 consecutive unique numbers, and with that being our "winning condition" I would say if you only rolled the dice 15 times, and you came out with that outcome, then you are incredibly lucky.
The only way that happens is if each of the 6 numbers have a 16.67% chance of being generated.
However, that is not the reality, and the question at hand, that you keep ignoring is where are you getting your 8% when the reality is the seeds don't even have an equal chance of advancement? My argument to you is, you're going to be lower than 1.5%, let alone 8%.
Does that make it more clear?
That isn't what I asked, so I am going to try again:
If you rolled a 6-sided die 6 times, and got exactly one 1, one 2 (etc) would you consider that evidence that the die is biased?
Because that's what you're doing here. Only you don't realize it.
Of course not, and it's not only something I do realize, I have already stated pages ago, that this alone is not evidence of a fix, it's just evidence that supports it. Call it circumstantial evidence if you want. I never stated impossible, just highly improbable.
There's plenty of examples(or I should say hearsay) where people play the lottery 1 time in their life and win it all, and no that doesn't mean the lottery is rigged, but that doesn't change the fact he was that 1 in a 177 million chance. As I have said, it could simply be evidence the NFL is getting that 1 in a million chance of lucky parity.
Having said that, you do still realize and at least acknowledge, that the chances of that happening are very freaking slim, right? And that this string of number, isn't just some random string, but it does in fact represent a perfect mix of winning seeds over the shortest possible time span. A Jackpot in terms of parity. This would be considered a rare occurrence.
Now I have answered your question, you still failed to answer mine and I'm betting you wont. Because once again, the number isn't 8% and it's not 1.5%.
I thoroughly enjoyed schooling you last night during the interminable UCLA game, my ignorant friend, but now it's time to move on and allow you to wallow in your own obscene ignorance. This will be my last post responding to you.
Wrong. Getting the most expected outcome, no matter how improbable it was, is not evidence of a fix.
This analogy is absolutely bankrupt for far too many reasons to go into here.
I absolutely acknowledge it is a rare occurrence but here's the most important thing in this discussion which you are obviously 100% clueless about. This is the #2 rule of statistical analysis:
Just because something improbable happened, that does not demonstrate anything conclusively. No matter what happens, it will be something improbable.
You're comparing 2 entirely different numbers. A p-value is not the probability of getting the results we observed. A p-value speaks to the statistical likelihood of there being conclusive evidence off anything occurring outside that which we would normally expect. 0.08 is low, but not low enough to reach the statistical threshold used in the grown up world.
The hilarious thing is that you're not even intelligent enough to realize your biggest piece of supporting evidence. Your "Exhibit A" should not be the 6-seeds-in-6-years thing, which is just a blip. Your "Exhibit A" is the fact that 3 Wild Card seeds (5 or 6) won in a 7 year timespan. THAT fact is the best evidence that something is afoot but even then, those of us that understand football realize that it has logical reasons behind it, such as the fact that the Giants front 4 d-lineman were able to pressure Brady, allowing 7 men to drop into coverage, etc.
The NFL does not, never has, and never will fix games and there is no tangible reason to think otherwise.
I was actually hoping I would wake up this morning, and find something other than you tooting your own horn. You're bolding something that has been clearly established as a common ground eons ago.
So let me get this wrong Mr. PHD, you're using deviation to determine whether a number that's obviously going to be within a set of data, actually belongs in the set of data and then you tried to pass that off as an actual argument against mine which is the probability, or improbability, of that occurring. Which is what we have been arguing in this thread.
And did it for a different time period, if I understood you correctly(7 years?). And still didn't actually take the time to look at the actual probability of this event taking place. And you continue to make ignorant statements, like the fact I would have never thought of the wildcards. Jut like you wrongly assume I have no concept of deviation.
You're ignoring the fact I have repeated numerous times in this thread just how many OTHER questionable oddities are also there. Including your recent revelation(why do you think I colored it brown and started there? For the same reason you didn't consider that I took the time to highlight the #3 seed all over the place and the #5 higher above...). And the reason why that is important, and easy to explain is because that actually represents something. The perfect mix of winning seeds in the shortest possible time span.
If you actually took the time to address this, actually provided some proof instead of bragging; instead of tip toeing around it or pushing your diploma around as some sort of excuse for you to be completely illogical or a pompous ass and look up the actual probability, this could have actually been constructive. But I'm glad you're finally taking notice of something.
I'll ask you once again. If you actually have this program, you could easily give the answer. What's the probability of a perfect parity of seeds winning the Superbowl in 6 consecutive years, using the given probabilities for each seeds?
I've asked at least 4 times, and you continue to ignore it. But then again you might have to compare it to the probability of 3 wild cards winning in 7 years, and something tells me you didn't actually do that either(considering your complete lack of proof) and are risking now shoving your other foot in your mouth. And God forbid, you might have to admit you are wrong. I don't know which one would short circuit first, your head or your ego.
The fact of the matter is that the chance of any particular seed ordering over 6 years advancing is freaking lottery odds.
And if it was a different ordering (say all 2 and 3 seeds over the 6 years), you'd be saying the same thing and making the same accusations using this slightly different scenario. That's what conspiracy theorists do. (I know, my brother is one and I've had ample opportunities to listen and think through the claims.)
The historical facts really aren't all that relevant to a conspiracy theorist. No matter what the facts are, they are always trying to convince someone (themselves at least) that "it is wildly improbable that this could happen by chance. Someone must be rigging the result!"
You're particular theory that has six different seeds advance in six straight years is novel in the sense that the ordinary person off the street would be saying, "who cares? Why would anyone be trying to rig that result?" Ah, but the conspiracy theorist is convinced that someone cares because that's what happened and there's a conspiracy. It's just a matter of statistically proving it (to others).
Hmm. Sounds like you're rewriting the probability theory I learned in high school. I won't ask you for the calculations you used to arrive at the statement above.
A wise man once said, "'tis better to be quiet and thought a fool than open one's mouth and remove all doubt".
No it's not. And that's actually the point. The chances of seeing certain patterns, even specific patterns, are much higher than others. Some are expected, others are extremely rare. For example having all 2 and 3 seeds winning over a period of 6 years as in your example, is still of an order of magnitute higher than all 6 seeds winning over a time span of 6 years.
The pattern should loosely follow the probability of each seed the tournament. Lots of 1s and 2,s a few 3s and 4s and very rarely 5s and 6s. And throughout history it did. Because this what the tournament format is designed to produce.
Exactly. No matter what outcome is observed, the probability of that specific outcome is going to be extremely low. The ignorant observer can grab any 6 year stretch, look at the results, and say "holy smokes! The odds of those exact results coming up are ridiculously low, so it must be rigged!" That is why statistical techniques exist to try and determine how far away the observed events are relative to the expected outcome.
In this case, it is worth noting that 3 WC's winning in a 7 year span is unusual. But it hardly rises to the level of statistical proof of anything. (The all-6-seeds-in-6-years thing is virtually meaningless)
The #1 thing working against our ignorant friend is the low sample size. When n=7, it's pretty tough to prove anything. 3 WC's winning in 7 years means nothing. But 30 WC's winning in 70 years would definitely get my attention.
No it's not exactly that. Because ordering isn't even an issue in this case, but the mix itself. It wouldn't matter if its 123456, or 653421.
And the wild cards is not only unusual, its improbable and yet still more probable than all 6 seeds winning in the shortest possible time span, which again, actually has meaning.
For example the chances of a group of 1 and 2 following each other, in any order, is roughly 21%, depending on what exactly you want to use for homefield advantage. Or roughly once every 5 years. And sure enough, despite the small sample, it's happened at roughly that rate.
The chances of 1,2,3 following each other, in any order, in a 3 year span is roughly 3.4 percent, or once every 30 years. And sure enough it happened once in 30 years.
Even the chances of 6 #1 seeds in a row, is actually not that improbable, since one would expect it to happen roughly once every 76 years, and in fact it already happened from 1982-1987. In fact the chances of 6 #1 seeds winning in a row is even more probable than seeing 5 #1 seeds and specifically 1 #2 seed, if you wanted to search for that, included in that same timespan. But less probable than 5 #1 seeds and any other seed.
But after that, when you start mixing in lower seeds, and specifically all 6 seeds, things start dropping off dramatically. Again, if you look at the history, it should be pretty obvious where things stayed consistent with expectations and where things went haywire, without even doing any math.
And once again sample size, has nothing to do with the specifically rare probability of ever seen all 6 seeds winning the tournament in a 6 year time span. Which once again, has meaning, and that meaning is basically perfect seed parity among Superbowl winners. If it happened twice, or three times or 4 times, and you wanted to see if it was in line with expectations you would need a very large sample. But that's not the argument here. The argument is that its very presence over such a small sample size is incredibly, incredibly rare. It would be something someone would expect to see happen at a time interval of hundreds if not thousands of years. And yet it's already happened pretty soon after division reshuffles. This would be a small sample size, if we were trying to look for it, and DIDN'T find it. But we're not searching for it. It's there. The question was what are the freaking odds?
Let alone, all the other unusual circumstances.
This is something the tournament is basically designed to PREVENT. And I didn't even include this year's results, #4 vs #3 which once again outcasted the #1 seed out of even appearing in a Superbowl, so it's continuing this bizzare deviation as well as the incredible run of underdog seeds which is now close to 90% in Superbowls over the past 13 years. This "perfect mix" of seeds that are taking place, is something someone would expect to notice if there was no bye week, no team strength and no homefield advantage. Or even something that came out of a round robin tournament, where seeding has little meaning in comparison to team strength.
But in this tournament, when it comes to making and winning the Superbowl, seeding is incredibly important. It's extremely important. Or....it should be. Until something inexplicably changed and the lower seeds have dominated 86% of the matches over the past 13 years along with all of these other highly suspect patterns. Pretty crappy odds to keep fighting for that #1 and #2 seed.
PS: I also enjoyed watching a statistician pump his chest out and blow out nothing but ignorance while waving a diploma(it's kind of like strippers with an MBA) by assuming he's the smartest guy in the room or that he would intimidate a physics and math major. It's always cute. It's like a lightweight pumping his muscles. Then you have to do what every statistician does when comes time to show proof. Ask a mathematician. Or ignore, tuck your tail between your legs and run off. In your case, both.
I just love Americans who brag about math and high school. Yeah I went there and you know what happened? All the hippies chicks took statistics in high school cause it was the easiest math class. Those of us with math and physics majors were waiting to blow through Calculus BC so we can move on to college where it actually picked off where I left off in math in middle school in Europe.
And yeah, if you don't understand 6/100 is 6% then I wouldn't ask for math proof either. We'd both be wasting our time.
Separate names with a comma.