[rake922]

Hypothetical situation,Would you tail or fade a capper that's 33-5 over last 30 days?

thanks

[donjuan]

Neither.

[20Four7]

Originally Posted by donjuan

Neither.

I saw that answer coming from a mile away

[dogman]

I would fade him unless he has proven in the past that he is a winning capper. If he has been a loser in the past I would definitely fade him as he has probably just gotten lucky and hit a winning streak.

[MrX]

Originally Posted by dogman

I would fade him unless he has proven in the past that he is a winning capper. If he has been a loser in the past I would definitely fade him as he has probably just gotten lucky and hit a winning streak.

Care to explain why this would justify fading him?

[dogman]

Mr X, I dont follow any handicappers but just because he has a great record over 38 games means nothing UNLESS I knew he is a longtime winner then fading him wouldn't make any sense. I would never fade someone who is winning and has won in the past.
But you could take the world's worst handicapper and he could hit a hot streak but that doesn't mean you should start following him because he has a great record over 38 games.
I would only tail a winner and fade a loser. You have to have alot of past information on this person before you would know what the correct strategy would be. Just my 2 cents.

[DrunkenLullaby]

Fading is an impossible decision. Tailing is possible, but only if I could justify to myself that 38 games represents a sufficiently valid sample space (not likely).

[Arnold]

Just because someone went 33-5, doesn't mean he'll continue winning. You can't predict how the guy gonna do next.

[Dark Horse]

If I had to make a choice, other than pass, I would fade him.

The problem is that the 87% will definitely come down, but that he could still be producing more winners than losers.

[donjuan]

If I had to make a choice, other than pass, I would fade him.

The problem is that the 87% will definitely come down, but that he could still be producing more winners than losers.

Of all the 3 possible answers, fading him is definitely the worst. There is no such thing as "being due" and at 33-5 he is more likely to be breakeven or better than he is bad enough to profitably fade.

The question is similar to this situation: you flip a coin 10 times and it comes up heads all 10 times. For the 11th flip you are given the choice of betting heads or tails at -105 or not betting either. Not betting either is the best choice but if forced to choose one, you should almost certainly choose heads as there is a chance the coin is not a fair coin and is more likely to have heads come up.

[Dark Horse]

Oh please. Gone over that too many times. The answer is that your experiment occurs outside of TIME. And because you don't have the first clue about time, you think it has no influence. But it does.

The 33-5 occurred on an upswing. It is a streak. An upswing is balanced by a downswing.

Keep flipping that coin. (I always thought they'd get weightless in a vacuum).

[donjuan]

Can't wait to see your peer reviewed article on time and streaks in Pseudoscience Weekly.

[Dark Horse]

Let's see. Proving the improbable, something that breaks with traditional beliefs, what does that go for these days in the skeptical world of modern science? Hundred million bucks? At least. Surely respectable academic institutions would line up to sponsor such an unlikely experiment as proving the unseen influence of Time.

Let me know when the professors are ready to put up the cash.

[donjuan]

Let's see. Proving the improbable, something that breaks with traditional beliefs, what does that go for these days in the skeptical world of modern science? Hundred million bucks? At least. Surely respectable academic institutions would line up to sponsor such an unlikely experiment as proving the unseen influence of Time.

Let me know when the professors are ready to put up the cash.

If you have some type of study done, go ahead and publish it so we can all have a good laugh. I'm pretty sure it won't be worth the pixels/paper it appears on in terms of scientific or anything other than comedic value.

[Dark Horse]

Of course you think that. You couldn't find outside of the box if there was no box. My proposal stands. For real.

[donjuan]

Look, I'm all for new knowledge and progress. It's just that you make an outrageous claim, which is refuted by logic, reason and science, and then refuse to provide any proof for your claim. Excuse me for remaining a tad skeptical.

[Dark Horse]

Skeptical is good. I can offer the proof in the form of a demonstration.

I'm sorry. I must have missed that part. You want me to prove an 'outrageous claim' for free? Why would I do that? Plenty of money in the academic world, last time I checked.

[donjuan]

Then why have you not taken advantage of that?

[Dark Horse]

Explained in pm.

DH out (of this thread).

[turnip]

If this person was 33-5 on +155 or better odds, I'd consider tailing him then.

If one pig flies, I will try to catch it.

If fading this person resulted in the fader being 5-33 at +53000 or better odds, I would consider fading him.

[Justin7]

I'd look at this plays and analyze his methodologies. If I were convinced that he'd win going forward, I'd work with him or hire him.

[GOJONNY]

Depends how often he bet - if he was going 2 plays a day or more, and was betting on over/unders, then there's no way he can keep up such a high winning rate on 50% bets over the long term - it should revert from 87% to closer to 55% sooner or later.

If they were playing moneylines above 1.60 odds though I'd consider it too small a sample to start tailing him, but wouldn't fade him.

[BuddyBear]

It's a little bit tricky. I guess neither is probably the best answer here.

[ketut]

OK, One can do this scientifically and with the numbers. This will ALWAYS even out or come very close at some time. Maybe this person will hit another 50 games before the "even-out" probably statistics kicks in, or maybe he's do for some loses now. No one knows for sure. If I had to bet his picks, I would probably?? ride him out until I start to see a downward loss, and at this point I would start to fade him. I only agree to this as long as it is not a "doulbe-up,triple-up" type of betting system he's using. Just my input and gl to everyone.

[Ganchrow]

Originally Posted by ketut

OK, One can do this scientifically and with the numbers. This will ALWAYS even out or come very close at some time. Maybe this person will hit another 50 games before the "even-out" probably statistics kicks in, or maybe he's do for some loses now. No one knows for sure. If I had to bet his picks, I would probably?? ride him out until I start to see a downward loss, and at this point I would start to fade him. I only agree to this as long as it is not a "doulbe-up,triple-up" type of betting system he's using. Just my input and gl to everyone.

This mentality is commonly referred to as the Gambler's Fallacy.

[DrunkenLullaby]

Originally Posted by ketut

OK, One can do this scientifically and with the numbers. This will ALWAYS even out or come very close at some time. Maybe this person will hit another 50 games before the "even-out" probably statistics kicks in , or maybe he's do for some loses now .

[GOJONNY]

Originally Posted by Ganchrow

This mentality is commonly referred to as the Gambler's Fallacy.

Yep, the Gambler's fallacy is really the main thinking behind this thread. But I keep thinking about how the "coin flip" evens out over time. It could take years to "even out" closer to 50%, but then again this tipster might only be betting selectively when he sees the true prob. at 70% (for example), which would decrease the chances of it "evening out".