Edge against an effecient market?

[danshan11]

Please can anyone answer this question for me

a guy bets 1000 times on an NBA game spread lets assume -110 is the standard margin so minus margin that is a +100 bet
Blazers -5.5 -110 is the bet 1000 times over 1000 games and the game closes at -5 -110 (+100 no margin)

how many games will the guy win?

[Derailer]

LOL. What is that? You are already have an answer for this.

[Derailer]

MonkeyFOcker bets Old Dominion -4 because he has -7. He makes a larger than average bet. Line closes 4 when Old Dominion starter is found to have taken $100,000 from a booster. Why should he not make outsized bets on these games?

[danshan11]

please anyone start your answer with how many games the guy will win, please
my answer is 500 assuming no variance or luck does not matter how skilled he is? what is your answer?
guy A noob picked them game drawing out of a hat never seen a line does not know what college football is wins 500
pro the best in the business 2000 time WSOp winner 10 times handicap sports champion wins 500
what is your answer guys?

[danshan11]

How many will Monkey win? rememeber the game closed at -4 -110 +100 no margin
Monkey wins 500 too

[Derailer]

I think everyone will agree that he will win 500 more times than not. But again, we are assuming we are dealing with professional level models. So the guy's bet did not work out as is the case almost half the time but not quite.

[danshan11]

yes this is not an edge for anyone I dont care what anyone's model says they will win 500 times
now take Monkey say he has -7 and bets it -4 and it closes -5.5 now we are cooking and have an edge
his edge is -4 was +100 but now -4 is -130 now that is an edge, so the hypo is -4 at -110 closes at -4 -110 is a coin flip for everyone and the guy who paid -4 and it closed at -5.5 has a coin flip plus a 1.5 hence -135 to get -4 and he will win 53-54% or whatever the math is on -130 minus juice. so back to the kelly using win percentage to calculate edge is wrong, if you are so set on using kelly use edge not win%. win% is useless. You use win % you will over bet or underbet accordingly

[Derailer]

What are you talking about? Professional models are not based on winning percentage. They are based on CLV. People are going to bet on their expected winning percentage based on closing line value.

Again, you are dealing with professional level gambling here. Not people that just turned on the computer.

[danshan11]

I think everyone will agree that he will win 500 more times than not. But again, we are assuming we are dealing with professional level models. So the guy's bet did not work out as is the case almost half the time but not quite.

Dude it does not matter how pro how good a model if he bets -4 -110 and it closes -4 -110 he has no edge, if you consider margin he has a negative edge. everyone not just scrubs not just nerfs not just sharps everyone, -4 -110 bet and closes -4 -110 is a loser because you will win 500 and win 450 EVERYONE

[danshan11]

Monkey is upset about the bad bet but that is not going to stop him from betting Dayton +7 when he has +4 and making an outsized bet. So why does he want to avoid betting it larger because of one bad break?

if his model is getting him in a situation where he is paying -110 for a closed bet at -130 he should bet the ranch! as long as he has a lot ranches to ride the variance and he will win, but if he bets -110 avg line and they close avg -110 he should move on to another sport EVEN IF HE IS WINNING LONG TERM!

[danshan11]

He might be a great guy and if his line value shows he is a good capper, he is probably winning but I highly doubt there are very many long term cappers not beating the line regularly at a 6.5 or more out of 10 rate are winning long term.

[Derailer]

You are making false arguments. You have no idea how professional models work.

[danshan11]

thanks derailer I appreciate the conversation can you please tell me specifically where I am wrong with what I am saying, this is my opinion and if you think it is flawed tell me which part please.

[Derailer]

People are telling you they don't bet based on the closing line. You are assuming they are saying they pay no attention to CLV. That is ALL they pay attention to. But they only pay attention to what their model makes the line. They pay no attention to the actual closing line in determining any particular bet. They have no reason to.

So again, you are engaging people with very advanced models with an assumption that they do not even use CLV which is so ridiculous it is beyond belief.

[danshan11]

when they bet they dont use clv because the game has not closed. Pros do not model to win games they model to beat lines, they look for deficencies in the market by comparing their probabilty to the line and most times the line will go in their direction as the market gets more effecient as it gets closer to closing. Pro's dont think I got -7 so if the line is -4 I will take it because they are going to win by 7 or more they say the market is soft here and if I can get them -4 at -110 that is a good buy right now because based on all the info available right now I think -4 is worth -145 or whatever, everyone would take a -110 bet everytime if they think it will close -145, same thing buy cheap. If i set up a coin flipping but you got to pick tails contest outside my home now and offered 50-50 odds but you get three flips to win once at even money people would line up around the door and bet EVEN if they had a feeling it was gonna be heads they would bet tails with a mulligan flip!

[Derailer]

That's right. And they base the amount of their bet on what they believe their winning percentage is on that particular line. They are constantly evaluating their CLV to determine if their model is as good as they think it is.

You have been telling them they should not do that. They need to look at closing lines on games to make their bets and not bet based on how good a bet they think they have because they cannot determine with any accuracy how good their bet is because nobody can beat closing lines or some such thing.

[KVB]

Bettors struggling to here because, in the end, the market is not that efficient.

[danshan11]

the market is super effecient if you are talking the big markets, NFL, EPL, MLB and like

[Derailer]

Right. The markets are very efficient but not so efficient that you don't try to beat them yourself. And people do likewise and you make assumptions about how they do those things that have no basis in reality and engage people with highly advanced models as if they do not know these things.

[danshan11]

That's right. And they base the amount of their bet on what they believe their winning percentage is on that particular line. They are constantly evaluating their CLV to determine if their model is as good as they think it is.

You have been telling them they should not do that. They need to look at closing lines on games to make their bets and not bet based on how good a bet they think they have because they cannot determine with any accuracy how good their bet is because nobody can beat closing lines or some such thing.

sorry this sounds like confusion
I am saying we dont know any of us what the line will close at so we dont know our edge before the game closes, so making some kelly formula with BS calculations is not sound. if you use your avg edge you are flat betting.
I think most pro bettors know in the end they will have +EV but they dont know how much and to base bet size on that may be flawed or inconsistent at best. I do the same thing my stake method of fixed profit betting is just as flawed if I bet the Yankees with my system and they are -150 I bet 1.5 units and if it closes at -180 I should have bet 1.8 on it, I make the same mistake BUT I was saying you should NEVER EVER NEVER base your bet size on any type of formula that involves past win% because as we just agreed win% is 50/50, the line is effecient

[Derailer]

sorry this sounds like confusion
I am saying we dont know any of us what the line will close at so we dont know our edge before the game closes, so making some kelly formula with BS calculations is not sound. if you use your avg edge you are flat betting.
I think most pro bettors know in the end they will have +EV but they dont know how much and to base bet size on that may be flawed or inconsistent at best. I do the same thing my stake method of fixed profit betting is just as flawed if I bet the Yankees with my system and they are -150 I bet 1.5 units and if it closes at -180 I should have bet 1.8 on it, I make the same mistake BUT I was saying you should NEVER EVER NEVER base your bet size on any type of formula that involves past win% because as we just agreed win% is 50/50, the line is effecient

Yes, well. Some people have told you that they do exactly that and they have reason to do so. Maybe they do or maybe they don't. But if they are doing it, they probably have reason to do so that they have evaluated. Because some models are better than others and it may be that their model actually does predict better than the closing line.

So if a person knows for a fact that they have outperformed a certain expectation that you would have for the model based on your assumptions they have reason to continue doing so and you will just have to advance your model to that level if that level does exist.

[KVB]

Right. The markets are very efficient but not so efficient that you don't try to beat them yourself. And people do likewise and you make assumptions about how they do those things that have no basis in reality and engage people with highly advanced models as if they do not know these things.

As long as your highly advanced model is taking and weighting the relevant factors that help you make a line that bettor reflects reality than the line the bookmaker can hang. You need to beat that average marketplace when in comes to reality or line error.

Just make sure you are not going around the block to get next door.

Regardless, if you can make that line, you can exploit the inefficiencies of the market place.

There are as many ways to handicap as there are handicappers.

[danshan11]

Yes, well. Some people have told you that they do exactly that and they have reason to do so. Maybe they do or maybe they don't. But if they are doing it, they probably have reason to do so that they have evaluated. Because some models are better than others and it may be that their model actually does predict better than the closing line.

So if a person knows for a fact that they have outperformed a certain expectation that you would have for the model based on your assumptions they have reason to continue doing so and you will just have to advance your model to that level if that level does exist.

I dont think such model exists if you are saying someone has a model that is more effecient than the entire markets consensus. The market has proven it is effecient so that leaves all of us books and bettors with one way to win MARGIN. we are all looking for margin and nothing else. if Joe has the best model ever and Joe's model says the Ducks -7 and the market closes Ducks -5, I would stick with the market on that one and I hope you would to or your pocket is going to get lighter!
we know the market is effecient and we know most cannot beat the market, and the few that do beat the market that is more a symptom of a good market that over time is beating the line.
Maybe I should define beating the line to me
I pay -4 -110 and it closes at -4 -135 that to me is beating the market, that to me is the definition of winning.
Winning 10 in a row losing 10 in a row does not mean anything all I care about is getting a good price (beating the line)

[danshan11]

As long as your highly advanced model is taking and weighting the relevant factors that help you make a line that bettor reflects reality than the line the bookmaker can hang. You need to beat that average marketplace when in comes to reality or line error.

Just make sure you are not going around the block to get next door.

Regardless, if you can make that line, you can exploit the inefficiencies of the market place.

There are as many ways to handicap as there are handicappers.

methods yes tons, but the thing is DONT bet with real money if you dont have an edge

[Derailer]

I dont think such model exists if you are saying someone has a model that is more effecient than the entire markets consensus. The market has proven it is effecient so that leaves all of us books and bettors with one way to win MARGIN. we are all looking for margin and nothing else. if Joe has the best model ever and Joe's model says the Ducks -7 and the market closes Ducks -5, I would stick with the market on that one and I hope you would to or your pocket is going to get lighter!
we know the market is effecient and we know most cannot beat the market, and the few that do beat the market that is more a symptom of a good market that over time is beating the line.
Maybe I should define beating the line to me
I pay -4 -110 and it closes at -4 -135 that to me is beating the market, that to me is the definition of winning.
Winning 10 in a row losing 10 in a row does not mean anything all I care about is getting a good price (beating the line)

That may well be. But it seems some here think they have one. So the thing to do would be to say so directly and determine why they think so.

Instead, you went on a tangent that got everyone confused.

So the thing here is to consider that you are dealing with people that are as advanced as you are and perhaps acknowledge their position and try to argue that point specifically.

[danshan11]

would you agree with this statement in general
if you bet -5 and it moves to -7 that you made a good bet and in general the market agrees with you?

[Barrakuda]

People incorrectly assume if their model starts to fail, it will show up in shrinking CLV. That's not what happens, unfortunately. You'll still get your CLV, but you will -- often quite suddenly -- have little to no edge bc everyone is now doing something similar to what you are -- they are just a little bigger/slower. E.g., Dr. H has been underwater for over 12 months, but still gets a ton of CLV. Same thing happened to him in CBB a few years ago.

If you're going to ride a model until its CLV dries up, you'll be broke in a hurry.

[KVB]

People incorrectly assume if their model starts to fail, it will show up in shrinking CLV. That's not what happens, unfortunately. You'll still get your CLV, but you will -- often quite suddenly -- have little to no edge bc everyone is now doing something similar to what you are -- they are just a little bigger/slower. E.g., Dr. H has been underwater for over 12 months, but still gets a ton of CLV. Same thing happened to him in CBB a few years ago.

If you're going to ride a model until its CLV dries up, you'll be broke in a hurry.

Take note, this happens over and over again.

The struggle is real.

[Derailer]

would you agree with this statement in general
if you bet -5 and it moves to -7 that you made a good bet and in general the market agrees with you?

Yes, and we are all looking for that bet. And if the line moves back down, we will need to re-evaluate to perhaps bet more or we may just sweat it out. But anyone doing either will base on their expectation of winning which will be based on a variety of factors and all of them will be a little bit different. And to the victors will go the spoils.

[danshan11]

derailer I disclosed upfront I am as dumb as a doorknob and that my writing skills are that of a 8 year old who got held back twice, so it can get confusing, full disclosure I am dumb!
I hope that my point still sticks and people learn to not trust some wins, even 100s of wins can be deviation. and NEVER EVER use wins to adjust your stake size.
most people are writing how to cook a gourmet steak with commercial grade equipment when most bettors got the internet and their buddy Frank to help them pick winners, these are the people who should not be messing with Kelly, modified staking, adjusting for wins losses, calculating EV and other stuff.

Understand this
for every winner there are 1000s of losers and the losers go away and the winners stay until they lose
so if you see 10 guys on here saying they win long term more than likely most of them are deviations not winners UNLESS they have documented win records showing them beating the line consistently

[danshan11]

requirement to bet more than a few bucks a game
requirement 1
BEFORE ANY lines come out make lines for 10 major market games overs, unders, spreads whatever do 10 lines and wait for those to close dont change them if you are far off on most of them go back to the drawing boards DO NOT BET
when you get to where 8 out of 10 are really close now you can start looking at small bets

[danshan11]

close is say you got -115 and it closes -124 that is close
close is not -115 closes -138

[danshan11]

People incorrectly assume if their model starts to fail, it will show up in shrinking CLV. That's not what happens, unfortunately. You'll still get your CLV, but you will -- often quite suddenly -- have little to no edge bc everyone is now doing something similar to what you are -- they are just a little bigger/slower. E.g., Dr. H has been underwater for over 12 months, but still gets a ton of CLV. Same thing happened to him in CBB a few years ago.

If you're going to ride a model until its CLV dries up, you'll be broke in a hurry.

Dr H was charging 5 grand a signup, he had serious money behind his plays, he was moving the line
if you look closely you can see at the end of the season when he was down 15 or more units people started getting scared and his CLV dropped as well. his followers were moving lines

[Derailer]

derailer I disclosed upfront I am as dumb as a doorknob and that my writing skills are that of a 8 year old who got held back twice, so it can get confusing, full disclosure I am dumb!
I hope that my point still sticks and people learn to not trust some wins, even 100s of wins can be deviation. and NEVER EVER use wins to adjust your stake size.
most people are writing how to cook a gourmet steak with commercial grade equipment when most bettors got the internet and their buddy Frank to help them pick winners, these are the people who should not be messing with Kelly, modified staking, adjusting for wins losses, calculating EV and other stuff.

Understand this
for every winner there are 1000s of losers and the losers go away and the winners stay until they lose
so if you see 10 guys on here saying they win long term more than likely most of them are deviations not winners UNLESS they have documented win records showing them beating the line consistently

Yes, I understand your intention was to address to people that they need to make certain they have positive CLV.

The problem was and continues to be that you are attempting to address this issue with highly advanced gamblers and arguing accepted points of view.

You have an underestimation of the average reader of a betting forum and you have a tendency to unknowingly preach to the choir.

But if you are going to continue to engage with advanced bettors on such thing as models, you need to take the position that you are engaging with professionals even if they might not be to that level because they are at least advanced enough to engage on that level. To not do so is to waste everyone's time. We all have lots of time to waste but would prefer not to do it on things that it turns out we agree on.

[danshan11]

Take note, this happens over and over again.

The struggle is real.

IMHO DrH is a jerk and rude dude BUT I think he is a very talented baseball capper who got zapped by the variance bug! but his CLV was slightly pushed by the strong following that helped move lines his way!