• The Cycling News forum is looking to add some volunteer moderators with Red Rick's recent retirement. If you're interested in helping keep our discussions on track, send a direct message to @SHaines here on the forum, or use the Contact Us form to message the Community Team.

    In the meanwhile, please use the Report option if you see a post that doesn't fit within the forum rules.

    Thanks!

Teams & Riders Froome Talk Only

Page 88 - Get up to date with the latest news, scores & standings from the Cycling News Community.
Franklin said:
Oh my how hilarious.

And yet the correct answer is once again: HEADS. The coin is not pure. This is why from a sufficiently big sample you should see a certain spread. If you see that spread there can be drawn conclusions.

You are really confused here because you want me to answer: The outcome of a coin toss is not predestined by an earlier coin toss. This is very true... but if we have a coin, toss it 100 times we certainly can guess what's most likely to happen unless it's really a pure coin. In your truely moronic example there can only be one answer. The fact that you do not understand this explains exactly why you can't analyze your way out of this.

Again, some people are just hilarious :D

You are still not getting it. You made bad example - dropping stones - what has good scientific answer (gravity), you do not need inject statistics and probability into it. I would even say that your example is disingenuous because with your (bad) example you create impression that probability estimations are simple and easy. But they are not, especially in real world. The way you chose to answer to my own example about coin toss perfectly demonstrates it. Coin toss is simple example, but you are not able to give simple straight answer. You made your answer conditional (pure or not pure coin) and changing this one little condition you created two totally different answers.
Imagine how many conditions, facts, known and unknown, uncertainties exists judging real world (pro cycling) situations. Dont pretend that you can make confident judgements, confident probability estimations.



Franklin said:
What a fantastic strawman. Because what Franklin did... was mention several indisputable facts which have everything to do with doping. Franklin did not use pictures or the position of the sun. He used facts.

Strawmen, I despise them. But this was to be expected.

I havent seen facts.
 
May 26, 2009
3,687
2
0
Visit site
LOLz
A coin that isn't pure?
Try constructing a non-pure coin anywhich way you want and flip it with a large no. of rotations before it lands. You'll never get it loaded to the extent where you get anything other than extremely close to 50/50.
To quote: Putting extra weight on one side of the coin does not alter [significantly] the bias of a coin if it is flipped in the air and undergoes a large number of rotations along its axis.

Physics for Dummies!

This is incredibly sad. I don't think anyone of you has any experience in statistics. The odds of a coin landing heads 100 times is infinitely small and thus is most likely not fair.

This is an odds game. If a coin toss is so biased... the coin is most likely not fair. Thus the next toss will be heads.

To put this one into perspective: if we do a weighted sample of 101 the 100 tosses are enough to give a near 100% margin.

I laugh at everyone who simply can't grasp the theory behind it. Yes, it is possible that this coin is fair. But anyone who knows anything about probability realizes that the chance that this coin is indeed fair is infinitely small.

If this is what the supporters of Froome argue I can only shake my head.


King Boonen said:
I'm sorry but this is horribly, horribly wrong. First, he is talking about a fair coin, you can't just make stuff up.

No he is not. Point out where he said it or stop pushing ridiculous nonsense.

If you know anything of probability, tossing a coin 100 times and getting 100 times head is a rather stark indicator of an unfair coin.


Next, he asked the likelihood, not what it would be, so I'm assuming you mean that it is 100% likely to be heads. This is wrong, a fair coin will have a 50/50 chance of being a heads or a tails, it doesn't matter what goes before it. You may need to flip it 10,000 times to get an even spread, or 1,000,000 times or even more.

Way over your head, the coin is more tham likely not fair. The sample of 100 tosses is incredibly big.

Empirical evidence can never prove anything completely.

I never claimed it did. You are absolutely disingenious if I ever used the word proof. STRAWMAN.

Retract please.
 
May 26, 2009
3,687
2
0
Visit site
Von Mises said:
You are still not getting it. You made bad example - dropping stones - what has good scientific answer (gravity), you do not need inject statistics and probability into it.

the probability remains rather untouched wouldn't you agree ;)

I would even say that your example is disingenuous because with your (bad) example you create impression that probability estimations are simple and easy.

You are not nearly as silly as you act out to be with your rant against probabilities. Thus I realize you fully know the chances of Frroome being clean are indeed remote. You don't like to admit it, but considering you are the only one catching it says enough. Yes it was a trap. The probability is still healthy, but the result was indeed preordained.

But they are not, especially in real world. The way you chose to answer to my own example about coin toss perfectly demonstrates it. Coin toss is simple example, but you are not able to give simple straight answer. You made your answer conditional (pure or not pure coin) and changing this one little condition you created two totally different answers.

Actually... no. The stone was disingenuous, but the answer about the coin wasn't. You know fully well that in a 101 spread 100 times heads is incredible. It could be a fair coin, but the probability of something being of is indeed much, much bigger. Had you said "te tosses" it would have been less obvious, but a 100 toss sample is statistically rather strong.

The error margin on this particular excercise (101 sample, 100 tosses) is less than 0.00 => As someone said, if this happens on the street, what is your bet?

Imagine how many conditions, facts, known and unknown, uncertainties exists judging real world (pro cycling) situations. Dont pretend that you can make confident judgements, confident probability estimations.

Again, you should know that sample size counts. In this case we have a large sample and extremely heavy bias. And that's just GT winner and dope. Then we look at dirty doctors, lieing managers, prior palmares. Every spread we have is heavily against Froome.

I havent seen facts.

That you deny prior palmares and prior winners as facts is simply disqualifying your opinion. But Von Mises dealt in ideology, not facts, so it's not a big surprise ;)
 
May 26, 2009
3,687
2
0
Visit site
slowup_fastdown said:
Seriously, if some street hustler just flipped a coin 100 times and got all heads, you would put your money on tails for flip 101? I would be wanting to inspect the double headed coin.

That's because you understand probability and use logic.

The spread of a 100 time toss is extremely accurate. if there's a 100 time heads result we could have an extremely rare occurence. Or the coin is unpure.

Does anyone else now see the similarity of the Froome case and why I find the Von Mises example so hilarious? ;)

This is quite simple a combination of both statitistics and probability.
 
May 26, 2009
3,687
2
0
Visit site
To put this one into perspective:

0.5^50=0.000000000000001

So that happens once in 100.000.000.000.000 times we try this excecise.

Anyone able to calculate 0.50^100?

Is that how the defenders of Froome are stating their case? :D
 
if only.........

Franklin said:
Is that how the defenders of Froome are stating their case? :D

if only life was so simple..............line me up v u bolt one thoosand times
.............when i never won no-one would say the race was rigged

one must look at the reality of the occurance it can't be compared with an abstract idea

Mark L
 
King Boonen said:
Chance is indeterminably assigned, probability is calculated. But in essence the terms are interchangeable, particularly when discussing something on a forum. Go troll elsewhere :rolleyes:

The probibilty that you're right is zero.

The chance that you'll pull the wool over the more simple types on this forum is high.

I prefer the probability that I'm right than the chance took that nobody would notice your error :rolleyes:

Forum or no forum it's important to be accurate.
 

martinvickers

BANNED
Oct 15, 2012
4,903
0
0
Visit site
Franklin said:
No, you can use statistics to get to set a probability. If I drop a stone a 100 time, what's the probability it also drops the 101 time? Observation. A scientific tool.

Sorry, nonsense. And actually, to be pedantic, a good example of the bias going wrong.

This is not prediction based on observation of probability, its prediction based on PROVEN science. Galileo's tower and all that.



Nonsense. Absolute and utter nonsense. I am not prdicting the future (nice strawman, care to take it down?), I am actually comparing past events. Froome's ascent is not something in the future, the whole problem here is that it is already happening.

Sorry Martin, I react very badly on strawman and this one deserves deep scorn. I suspect you are smart enough to realize I try to compare things which are undisputed facts.

React away, Franklin. I think you know me, and know I like you, well enough to know I'm not backing down on that sort of stuff. I'm not talking about an analysis of Froome's previous rides, i'm talking about the general idea of using previous rides to predict what future rides might mean.

Enough for what? I destroyed this strawman, thus what is there left to cling to? Are you saying you are deluding yourself as you don't dare to draw the logical conclusions.

Sorry, don't understand a word; can you rephrase?

I think that you fall into the trap of many others: This is not a court of law. If this was a court of law I would agree, we haven't got anything on Sky (Brailsford deserves a punishment, but both UCI as BC are spineless cowards).

It's not about rules of law (or actually, strictly speaking the rules of evidence - i recommend Keane on the law of evidence, genuinelly - easy read), it's about rules of logic. The former is simply an attempt to socialise the latter.

p.s. UCI, spineless cowards? You're actually too kind. BC? jury out, don't know enough of their internal carry on. But Anto Moran apart, Irish cycling certainly deserve it.



I knew someone would bring this up :rolleyes:

But there is of course a way and Wiggins once wanted to go that way: Post the neccesary medical records, be transparent in your actions.

It's sad you refuse to acknowledge there isn't away to go foprward and instead defend the status quo.



And full marks for you. We can not and should not prosecute just because of a statitical likelihood or suspicion. But this is not a court of law.

The problem is a doping violation system IS, in going to happen in a manner of speaking in a court of law, or certainly a mutually agreed tribunal of fact (that's how CAS operates as well, come to think of it).
 
May 26, 2009
3,687
2
0
Visit site
ebandit said:
if only life was so simple..............line me up v u bolt one thoosand times
.............when i never won no-one would say the race was rigged

one must look at the reality of the occurance it can't be compared with an abstract idea

Mark L

First of, I wasn't the one introducing the silly 100 coin toss.
Secondly, I pointed out several spreads.

I hope Sky is clean, that says enough about how I estimate the probability. Very small, but nowhere near as bizarre as the 100 coin toss.

But what the defenders really, really, really should understand is that the position of the critics is by far the logical position. The other position is religous like faith.

Does this mean absolute proof? No and I never claimed it was. Did I say it's enough for prosecution? HELL NO.

But the probability is quite simply extremely heavily against Froome in this case.
 
Franklin said:
This is incredibly sad. I don't think anyone of you has any experience in statistics. The odds of a coin landing heads 100 times is infinitely small and thus is most likely not fair.
Three degrees from three different institutions and several peer-reviewed papers, many involving multivariate analysis. My "experience" in statistics is fairly high.

This is an odds game. If a coin toss is so biased... the coin is most likely not fair. Thus the next toss will be heads.

To put this one into perspective: if we do a weighted sample of 101 the 100 tosses are enough to give a near 100% margin.

I laugh at everyone who simply can't grasp the theory behind it. Yes, it is possible that this coin is fair. But anyone who knows anything about probability realizes that the chance that this coin is indeed fair is infinitely small.

I laugh at someone who can't grasp the fact that any coin you find is unlikely to result in a 50/50 split, but that's not even the point, this is a discussion about the maths NOT AN ACTUAL COIN.

No he is not. Point out where he said it or stop pushing ridiculous nonsense.

Point out where he says it is biased? Again, this is not about a real coin, it is an example discussing the maths.

If you know anything of probability, tossing a coin 100 times and getting 100 times head is a rather stark indicator of an unfair coin.
What happens if the next 100 tosses are tails? Does the coin suddenly become fair? :rolleyes:


Way over your head, the coin is more tham likely not fair. The sample of 100 tosses is incredibly big.

It really isn't a big sample set, but that just shows your own ignorance on the subject.

I never claimed it did. You are absolutely disingenious if I ever used the word proof. STRAWMAN.

Retract please.

You claimed the answer was obviously heads, that is purely based on the empirical evidence of 100 tosses resulting in 100 heads. You don't need to use the word.

So no, no retraction, I can recommend some good books on statistics though if you need them.
 
thehog said:
The probibilty that you're right is zero.

The chance that you'll pull the wool over the more simple types on this forum is high.

I prefer the probability that I'm right than the chance took that nobody would notice your error :rolleyes:

Forum or no forum it's important to be accurate.

To be honest, this just sounds like you're talking to yourself.
 
May 26, 2009
3,687
2
0
Visit site
martinvickers said:
Sorry, nonsense. And actually, to be pedantic, a good example of the bias going wrong.

This is not prediction based on observation of probability, its prediction based on PROVEN science. Galileo's tower and all that.

No it's actually both. Gallileo used empirics to proof his theory. This being bound by the law of gravity does not invalidate the probability. Besides... Quantum theory! ;)

React away, Franklin. I think you know me, and know I like you, well enough to know I'm not backing down on that sort of stuff. I'm not talking about an analysis of Froome's previous rides, i'm talking about the general idea of using previous rides to predict what future rides might mean.

I did tone down the reply quite a bit before I pushed post. ISober me in the morning thought I was to harsh ;)

but the complaint you had about predicting the future was wrongly adressed. I did not make that case. But considering I make bad posts too (^^) I certainly take this in stride.

Sorry, don't understand a word; can you rephrase?

I think I missread... I thought you wanted more indicators to calculate probability, but what you most likely meant it would not be enough for prosecution. We agree on the latter (and I think we do on the former).

It's not about rules of law (or actually, strictly speaking the rules of evidence - i recommend Keane on the law of evidence, genuinelly - easy read), it's about rules of logic. The former is simply an attempt to socialise the latter.

You are losing me here, but that could be because I had a work party and never drink (I did now^^)

p.s. UCI, spineless cowards? You're actually too kind. BC? jury out, don't know enough of their internal carry on. But Anto Moran apart, Irish cycling certainly deserve it.

Meh, it was my pet peeve about DB which is heavily coloured by my dislike. If I ruled the world (and I'm very happy I don't) I would toss out every manager who broke multiple doping policies. In this case DB made quite a few stinkers.

but that's my personal sense of justice which, thank god, is not how the world works.

The problem is a doping violation system IS, in going to happen in a manner of speaking in a court of law, or certainly a mutually agreed tribunal of fact (that's how CAS operates as well, come to think of it).

It's why Cas rulings are sometimes so edgy.

For instance Contadors case about tainted meat is actually rather plausible considering the meat scandals. Of course, there were other indicators (plasticizers to start) which makes the probability of AC being clean rather small as well (^^). All things considered I do think he was punished just because they finally got something. The clen ruling in itself was (in my opinion, IANAL)not very convincing.
 
May 26, 2009
3,687
2
0
Visit site
King Boonen said:
Three degrees from three different institutions and several peer-reviewed papers, many involving multivariate analysis. My "experience" in statistics is fairly high.

If so I am utterly shocked you fail to understand the difference between odds and probability. If the coin is pure, the odds will be 50%. This is what von Mises tried to infer.

But a spread of 50% which gives a 100% score in 100% of the tries... well, that's either once in a googolplex or something is wrong with the test.

I laugh at someone who can't grasp the fact that any coin you find is unlikely to result in a 50/50 split, but that's not even the point, this is a discussion about the maths NOT AN ACTUAL COIN.

No. Nonsense.

Von Mises forgot to set the conditions. He had a coin and he tossed it 100 times and got heads. There is no further info. I simply dare you to point out the extra info you conjure out of thin air.

Ok, so here we have an experiment with two outcomes. He does it 100 times and always get result A. We enter attempt 101. Now we can either think this is indeed once in a googolplex... or we can deduce that the test is rigged.

Point out where he says it is biased? Again, this is not about a real coin, it is an example discussing the maths.

No. Nonsense. I do not have to point out stuff he did not say! Instead, point out the conditions you infer to or admit that indeed no conditions were given.

What happens if the next 100 tosses are tails? Does the coin suddenly become fair? :rolleyes:

What do you think ;)

It really isn't a big sample set, but that just shows your own ignorance on the subject.

Either you are taking the p. out of me or you are lieing about your experience. 100 out of 101 is a big sample indeed.

You claimed the answer was obviously heads, that is purely based on the empirical evidence of 100 tosses resulting in 100 heads. You don't need to use the word. So no, no retraction, I can recommend some good books on statistics though if you need them.
The amusing part is that you know that I motivated why the obvious answer is "heads". And yes, that's based on empirical evidence of a 50% test done a 100 times. But somehow you make a leap and turn my answer of the obvious probability as if I claim it is proof. Of course this is quite disingenuous, but lucky for me everyone can check I never said this was proof. Sorry King Boonen... a strawman is a strawman and yours has been torched right where you put it.

I do find it bad form to erect a strawman and not retract iot, but that's my sense of manners. I guess some people find it fair to put words in people's mouth.

And about your books. If you are as good as you say you are this discussion is over.. and we both know why.
 
Jun 15, 2009
835
0
0
Visit site
slowup_fastdown said:
Seriously, if some street hustler just flipped a coin 100 times and got all heads, you would put your money on tails for flip 101? I would be wanting to inspect the double headed coin.

Read what I just wrote. Construct any coin you like, to "load it" as it's termed. Won't matter one iota. The probability of getting either side will still be close to 50/50 if the coin is flipped, rotating around its axis. This is physics.

Now, spin it on a table and things will be different, with a propensity for falling with the heavy side down. But flipping? No way.

I'd be very, very surprised if some street hustler would flip a coin a 100 times and got all heads, and I'd probably hold on to my wallet for dear life, but as a thought-experiment, the probability would still be 50/50 at flip nr. 101.
 
May 26, 2009
3,687
2
0
Visit site
hektoren said:
I'd be very, very surprised if some street hustler would flip a coin a 100 times and got all heads, and I'd probably hold on to my wallet for dear life,

:rolleyes:

but as a thought-experiment, the probability would still be 50/50 at flip nr. 101.

Then why would you hold on to your wallet? ;)
 
thehog said:
My opinion is that those who accuse others of trolling are using deflection.

Just an opinion.

If your case is strong then you'd have no problems in arguing the point.

Again an opinion and an observation.

Please continue. As you were.

How exactly can I argue a point when you say something is wrong but don't say why? That's just blatant trolling.
 

TRENDING THREADS