Thought experiment for you "do I have a winning model" types

**mathdotcom** · 08-31-12, 01:33 PM

The fact that you want to hear what he has to say means you are doomed

**hutennis** · 08-31-12, 02:18 PM

Originally posted by cyberbabble

Now that we know what doesn't work, what do you do that does work? I'm not asking for any valuable, secret info. Just a general statement that arbs or whatever are the only thing that you have found that does work. Perhaps bonus whoring? Since I live in the U.S. neither one is practical for me, so I won't be stealing all of your +EV.

Here is a very general statement for you.

First, you need to chose a right approach.
Ability to do that directly correlates to your understanding of how the world works, i.e.
how/why money change hands (won/lost).

For eg.

I have two choices:

#1. To make money I have to outsmart, with a substantial margin, some of the most qualified and well prepared professionals in a world.
#2. To make make money I have to take advantage of a constant mistakes made by poor deluded soul who is chasing his tale in a never-ending and always confusing stream of random events with a very little clue of what he is actually doing.

For me there is no choice, really. I just can't do #1. If you can, great. Good for you.
But I, personally, have no problem admitting - I'm not smart enough.

#2. on another hand, I can try.

For that i need a few things though.
I need to know what to look for and where to be looking for it.
I need a complete understanding of what can go wrong, to what extend and why, so I can manage properly variety of risks involved.
I have to have a very well thought of plan based on probability and risk/reward ratio (among other things) analysis witch I must execute
with no fear, greed or any other emotions.

I hope it is general enough. Sorry for my poor english.
I totally expect a few more "I love you man! You are paying my bills" statements and it is fine by me.

**hutennis** · 08-31-12, 02:20 PM

Originally posted by cyberbabble

No.
Just giving hutennis an opening to make a decent post instead of his usual stuff.

My "usual stuff" is pretty much direct derivative of what is above.

**cyberbabble** · 08-31-12, 02:31 PM

hutennis quote if you didn't recognize it.
This is how you repay for all my nice, suckup posts about you?

I live in US so arbs, bonus whoring, etc. aren't much good to me.
Modeling major sports is somewhere between very difficult and impossible.

Dabbling in NFL props, NFL totals, and NASCAR as soft markets. Not a degenerate gambler that likes to lose.

What's a poor boy to do?

**hutennis** · 08-31-12, 02:31 PM

Originally posted by cyberbabble

I won't be stealing all of your +EV.

There is no much to left to steal. Duff85 pretty much cleaned me out.

**jgilmartin** · 08-31-12, 03:20 PM

If you think of NFL totals as a soft market you are doing OK.

**hutennis** · 08-31-12, 03:25 PM

Originally posted by cyberbabble

hutennis quote if you didn't recognize it.
This is how you repay for all my nice, suckup posts about you?

I live in US so arbs, bonus whoring, etc. aren't much good to me.
Modeling major sports is somewhere between very difficult and impossible.

Dabbling in NFL props, NFL totals, and NASCAR as soft markets. Not a degenerate gambler that likes to lose.

What's a poor boy to do?

What do you expect, man?

If you think there are obstacles beyond your control that are impossible to overcome, then what I can do?
I can commend you though for realizing things you realize and not jumping into an empty pool simply because you are bored, love jumping or need money.
Maybe you should get on a learning curve?
Try to reverse engineer my generalities. (generalities are the only things you gonna get without actually working hard yourself)
Try to figure out what is actually behind those words.
When you finish (if you will) you''ll be armed an degerous.
And maybe by that time you'll stumble on a few ideas how you can overcome those obstacles so you can apply yourself in some other things
besides bonus whoring sport speculation industry has to offer.

**cyberbabble** · 08-31-12, 03:27 PM

NFL totals are major market that is softer than major market sides since limits are smaller.

**cyberbabble** · 08-31-12, 04:04 PM

hutennis - Is it true that English is not you first language? I too have problems understanding some of these posts.
I feel your pain. My first language was American hillbilly. I'm working on my English but it is tough.
Don't give up. Someday you will get the hang of it.

**mathdotcom** · 08-31-12, 06:04 PM

Originally posted by hutennis

#1. To make money I have to outsmart, with a substantial margin, some of the most qualified and well prepared professionals in a world.
#2. To make make money I have to take advantage of a constant mistakes made by poor deluded soul who is chasing his tale in a never-ending and always confusing stream of random events with a very little clue of what he is actually doing.

For me there is no choice, really. I just can't do #1. If you can, great. Good for you.
But I, personally, have no problem admitting - I'm not smart enough.

There. So why do you insist on arguing with people who do #1?

Also I have a problem with you calling mostly uneducated hillbilly bookmakers the 'most qualified and well prepared professionals' in the world. They are morons. Most bettors are also morons. Give one set of morons a 10c edge and they will come out on top. Professionals are smart enough to overcome the 10c deficit and it is incredibly easy to do. Some just do it on different levels.

**MonkeyF0cker** · 08-31-12, 06:57 PM

Originally posted by cyberbabble

NFL totals are major market that is softer than major market sides since limits are smaller.

I can get $50k down on game day NFL totals without so much as the bat of an eyelash at several books.

I don't think I'd call those small(er) limits.

Limits on totals are almost always smaller than (about half of the) limits on sides. Doesn't mean they are necessarily soft.

**wrongturn** · 08-31-12, 08:13 PM

Originally posted by mathdotcom

There. So why do you insist on arguing with people who do #1?

Because #1 has very small chance of success. It is so small, similar to the probability of 1 hitting lottery with only 100 people buying ticket, that if it happens it deserves investigation within 2 minutes. If you refuse to be peer reviewed (ie investigated for lottery fraud), then the assumption that you are really good (ie simply lucky in lottery case) can be safely dismissed.

**mathdotcom** · 08-31-12, 08:55 PM

Originally posted by wrongturn

Because #1 has very small chance of success. It is so small, similar to the probability of 1 hitting lottery with only 100 people buying ticket, that if it happens it deserves investigation within 2 minutes. If you refuse to be peer reviewed (ie investigated for lottery fraud), then the assumption that you are really good (ie simply lucky in lottery case) can be safely dismissed.

Thanks wrongturn

I am flattered

I am one in a million !

**cyberbabble** · 08-31-12, 09:55 PM

Originally posted by Wrecktangle

MDC = MF = Hutinnis

Nicely done. Summed up whole thread in three words or less.

**scottybweyy** · 09-09-12, 08:58 AM

Ignoring the blatant no answers request this seems like a fun thread!

Anyone who can flip a coin at almost 4σ outside the mean is unlikely but within reality. You would expect it to happen roughly about 1/15000 for each flipper or about 1/150 for each 100 person experiment.

It is inconclusive what Mr. Ed thinks. He'll likely think he has a fair coin and is just good at flipping coins.

It is inconclusive what MDC thinks. He'll likely want to investigate the coin further if no other coins exhibited this type of behavior. I know this is absolute thinking but I digress...Perhaps MDC will want to perform the test with the same 100 people but have them using the same coin running sequential tests.

**mathdotcom** · 09-09-12, 09:51 AM

The only point of the exercise was put an inequality between Probability MDC Thinks Coin is Fair and Probability MrED Thinks Coin is Fair, and understand why.

Nothing more.

Unfortunately this was too much for people to grasp

**sayhey69** · 09-09-12, 10:27 AM

p mr ed thinks coin is fair < p mdc thinks coin is fair. because mdc expects 5 coin flippers to come back with p-values < .05, 1 coin flipper to come back with p-value < .01, etc. AM I DOING IT RIGHT?

**Optional** · 09-09-12, 01:26 PM

Originally posted by sayhey69

p mr ed thinks coin is fair < p mdc thinks coin is fair. because mdc expects 5 coin flippers to come back with p-values < .05, 1 coin flipper to come back with p-value < .01, etc. AM I DOING IT RIGHT?

Mr Ed thinks he has a bias coin (or thinks he is a good flipper as Scotty suggested)

MDC knows MrEd is wrong. (because he knows all coins are identical)

Basically, the lesson is if you come into the Think Tank with any new model, system or theory that You think gives an advantage, take care as the 'MDCs' around here will already know you are wrong.

**sayhey69** · 09-09-12, 02:46 PM

ya thats pretty much what i said braj

**MonkeyF0cker** · 09-09-12, 09:06 PM

Originally posted by Optional

Mr Ed thinks he has a bias coin (or thinks he is a good flipper as Scotty suggested)

MDC knows MrEd is wrong. (because he knows all coins are identical)

Basically, the lesson is if you come into the Think Tank with any new model, system or theory that You think gives an advantage, take care as the 'MDCs' around here will already know you are wrong.

If you have to ask a forum if your "model" has statistical significance, what does your (obviously vast) experience say about that person's mathematical/statistical erudition? Now, let's contrast that with the mathematical/statistical knowledge required to construct a long-term, profitable model. Let us know when you figure out the disconnect.

**Optional** · 09-09-12, 10:13 PM

Originally posted by MonkeyF0cker

If you have to ask a forum if your "model" has statistical significance, what does your (obviously vast) experience say about that person's mathematical/statistical erudition? Now, let's contrast that with the mathematical/statistical knowledge required to construct a long-term, profitable model. Let us know when you figure out the disconnect.

I get the problem Monkey. The only thing I don't agree about is that the best solution is to ridicule those trying to learn.

Although this thread does does show why you might want to at times I guess.