[danshan11]

I guess the message got confused
what i was saying is 99.99% of people think wins and losses matter. YES they do matter in huge sample sizes and over extended time. 1000 is not enough to even dream of saying you have skill. it is hard to hear because most people are like hey look I win so I am skilled. I have a good friend he knows nothing about line value and he wins alot, he takes his betting very serious and he is serious, he thinks that looking at the last couple of games and pretty much takes a team based on what he feels or reads a few things, if a pitcher jhad a couple good starts or whatever and he says all the time "line value dont mean shit if you dont win" and I think people like him and others respect winners and that is a huge mistake in the long term the juice will get him unless he actually has an edge. He bets say 2 or 3 games a day and has for say 300 or so days this way and he is up and he thinks he is skilled and I say he is not, not even close. This sim should be an eye opener to him and show him that he might just be one of the 4 of the 20 in the sim. ROI yes ROI is so important to so many things but not betting, ROI means nothing, if you are lucky and win you have a high ROI, if you are unlucky and lose you have a -ROI, how is that different than WL record?

[jbayko]

I think I dont know what ROI is let me explain and you folks tell me if I know what ROI is

W 100
L 100
P 0
wl% 50%
bet 100 per game at just for argument sake at +100
I bet 20,000 won 10,000 and lost 10,000 so I am even my ROI is 0%

now tell me how would I come up with an ROI that did not involve wins and losses?

Wow, that’s quite a condescending reply when I was trying to be diplomatic.

Of course I know what ROI is. I now see that your assumption is that everyone flat bets.

Yes, if both of these things are true, then W-L is directly related to ROI:

1) Flat betting (same amount on every wager).
2) Always betting the same odds (e.g. -105 for EVERY wager).

Of course, these assumotions are asinine to make. No one always bets the exact same odds and many people vary their stakes...in fact, I’d wager that the majority of regular posters here do so. Since at least one of those two factors vary for everybody, then it is not accurate to say that wins and losses are directly related to ROI. It’s just not true. If that were the case, then # of wind and losses would be in the formula for ROI, but they aren’t for good reason.

[HeeeHAWWWW]

nah what? what would I be trying to learn from a binomial calculator relevant to this thread?

The mean and distribution of expected outcomes.

[danshan11]

yeah deviations and what part of that am i missing?

[tsty]

Say the book closes with Pittsburgh -7 in the NFL.

Say the game ends with Pitt winning by 10 points. The book's line error would be 3, it's the difference from reality.

Now, say I had Pitt winning by 11 points. My line error would be 1.

I always say that the goal in creating a forecast or line is to be able to reflect reality better than the line the bookmaker can hang. If you can forecast reality better, you will win.

Average line error can tell you if you are closer to reality than the book. If you are beating the line error of the book, you will likely beat the sport.

This also counts for pricing, like a NFL -3 (-105) moving to -3 (-130) or whatever.

This is a very basic way to think about the line

The line being -7 doesn't mean the score of the game will be at -7

Even if they won the game by 10 then it doesn't mean the line was supposed to be -10 either

[KVB]

This is a very basic way to think about the line

The line being -7 doesn't mean the score of the game will be at -7

Even if they won the game by 10 then it doesn't mean the line was supposed to be -10 either

I couldn't agree more. That's why I added this post...

...A line is hung, it is not meant to predict a score, but if you can predict the score of game better than that line does, even though that's not the line's purpose, you will go a long way toward winning long term, period.

In most markets, you will then likely beat the closing line, since you seem to really like that metric.

When measuring average line error, it is one of the times where the final scores are taken into account, that's how it's measured.

Otherwise, often times, we shouldn't read too much into the final score, just to beat the market at the error.

I, in no way, am trying to imply that the line offered by the book is meant to be a prediction of the score.

I am only trying to explain to this guy that making a line that can better reflect what actually happens than the line bookmaker will be forced to hang because of market pressures will lead to long term profit.

Anymore, these differences are small, but I just can't simplify ROI and Average Line Error, as far as tests of a handicapper, any more than that.

Guy thinks I made up line error.

[danshan11]

This is a very basic way to think about the line

The line being -7 doesn't mean the score of the game will be at -7

Even if they won the game by 10 then it doesn't mean the line was supposed to be -10 either

absolutely correct! I know people hate me cause I keep telling them that all the methods they believe in are not signs of long term success and they just cant take it. I was them once, I thought I knew more about the NFL than the entire betting world, funny now as I think back of how I felt!

[KVB]

Nah. Go play with a binomial calculator, that's more useful than small-run sims.

The only reason I chimed in here was to point out that there are tools better than those sim results to judge a handicapper.

Good suggestion.

[danshan11]

I couldn't agree more. That's why I added this post...



I, in no way, am trying to imply that the line offered by the book is meant to be a prediction of the score.

I am only trying to explain to this guy that making a line that can better reflect what actually happens than the line bookmaker will be forced to hang because of market pressures will lead to long term profit.

Anymore, these differences are small, but I just can't simplify ROI and Average Line Error, as far as tests of a handicapper, any more than that.

Guy thinks I made up line error.

the line offered by the book is kinda a prediction of the score so you are wrong here but not completely
you call it making a line but in reality its implied probability and if you are seeing a huge difference from your model implied probability and the actual line, you would be smart to not trust your line. IF A and B are close its a good thing, if A and B are not close its a bad thing. if I make a line +7 and it comes out +3, I dont run and bet, I try and figure out what the heck I did wrong.

ROI is a win loss metric and does not show skill unless you have a huge huge sample size
Average line error is a very cool thing you should trademark it and bottle it. I think avg line error is useless but if you have something more than chit chat to show me its value, you know I am ready willing and able to look at it and give you my 2 cents! And quit trying to be mean by calling me this guy, you know my username or call me NERF please.

I respect the effort you put in KVB but can you be wrong and if you cant is that a good thing? PLease do me a favor dont discount my opinion by just saying you are wrong, show me show me data, give me links, Missouri baby "SHOW ME"

[danshan11]

The only reason I chimed in here was to point out that there are tools better than those sim results to judge a handicapper.

Good suggestion.

I think KVB and heehaw you are under the impression I dont know what variance is or deviations but I do folks! I am just using this thread to try and show people that wins and losses are NOT a good metric, nor is ROI (essentially wins and losses) or Line error( i do not even know what you would look for in this metric and consider that successful, what you would be closer to the final score than the closing line, what the heck value would that have? isnt that the same thing as saying I am smarter than the market, my model is better than the entire betting consensus, dont you see those sentences are funny and crazy.

Please dont stop chatting about it , I love this shit so much it really irritates people.

[danshan11]

The only reason I chimed in here was to point out that there are tools better than those sim results to judge a handicapper.

Good suggestion.

you got confused those sims show that wins and losses are not a good indicator of skill UNLESS you have a zillion of them and I still want to know why you win with line value as justification for the success

[bhoor]

danshan, your threads are interesting ones.

[danshan11]

thanks for the comment, I am the true definition of a NERF, I know enough to make big mistakes and not enough to know not to do some things. Everyday I am one step closer to being a sharp and losing my nerf avatar but for now I am still a soft round NERF!

[Jayvegas420]

the line offered by the book is kinda a prediction of the score so you are wrong here but not completely
you call it making a line but in reality its implied probability and if you are seeing a huge difference from your model implied probability and the actual line, you would be smart to not trust your line. IF A and B are close its a good thing, if A and B are not close its a bad thing. if I make a line +7 and it comes out +3, I dont run and bet, I try and figure out what the heck I did wrong.

ROI is a win loss metric and does not show skill unless you have a huge huge sample size
Average line error is a very cool thing you should trademark it and bottle it. I think avg line error is useless but if you have something more than chit chat to show me its value, you know I am ready willing and able to look at it and give you my 2 cents! And quit trying to be mean by calling me this guy, you know my username or call me NERF please.

I respect the effort you put in KVB but can you be wrong and if you cant is that a good thing? PLease do me a favor dont discount my opinion by just saying you are wrong, show me show me data, give me links, Missouri baby "SHOW ME"

The line offered by the books does not reflect a prediction of the game"s outcome rather, it's trying to predict the Public's reaction to that line.
In one instance it appears that you are trying to argue that the inevitable outcome has nothing to do with the line, or it's margin of error. But then you are also saying that the opening line, is implied probability. Aren't those mutually exclusive to each other?

Return on investment has very little to do with wins and losses, someone tried to explain this to you but it didn't seem to sink in. You don't seem dumb, so I assume you understand that somebody can go 1 and 9 and have a return of investment of say.... plus two units.
While another player can go 9 and 1 after betting - 1200 lines, he is down 3 units.

Please explain to me how in that example, wins & losses are a reflection of your ROI

[KVB]

the line offered by the book is kinda a prediction of the score so you are wrong here but not completely
you call it making a line but in reality its implied probability and if you are seeing a huge difference from your model implied probability and the actual line, you would be smart to not trust your line. IF A and B are close its a good thing, if A and B are not close its a bad thing. if I make a line +7 and it comes out +3, I dont run and bet, I try and figure out what the heck I did wrong.

ROI is a win loss metric and does not show skill unless you have a huge huge sample size
Average line error is a very cool thing you should trademark it and bottle it. I think avg line error is useless but if you have something more than chit chat to show me its value, you know I am ready willing and able to look at it and give you my 2 cents! And quit trying to be mean by calling me this guy, you know my username or call me NERF please.

I respect the effort you put in KVB but can you be wrong and if you cant is that a good thing? PLease do me a favor dont discount my opinion by just saying you are wrong, show me show me data, give me links, Missouri baby "SHOW ME"

What you said in bold is a good idea.

The rest of this post is garbage.

Things like ROI and average line error are very basic, if you are as new as you say you are, then learn something.

You are "this guy" because you seem to be trolling.

You are not listening to repeated attempts to explain the concept of even ROI, you are just reverting to the same win/loss ignorance.

I can not help you any further if you are going to argue basics, basic definitions like ROI, without at least offering a definition of your own.

[tsty]

The line offered by the books does not reflect a prediction of the game"s outcome rather, it's trying to predict the Public's reaction to that line.

this is wrong

definitely wrong now

maybe it was correct 20 years ago

[danshan11]

The line offered by the books does not reflect a prediction of the game"s outcome rather, it's trying to predict the Public's reaction to that line.
In one instance it appears that you are trying to argue that the inevitable outcome has nothing to do with the line, or it's margin of error. But then you are also saying that the opening line, is implied probability. Aren't those mutually exclusive to each other?

Return on investment has very little to do with wins and losses, someone tried to explain this to you but it didn't seem to sink in. You don't seem dumb, so I assume you understand that somebody can go 1 and 9 and have a return of investment of say.... plus two units.
While another player can go 9 and 1 after betting - 1200 lines, he is down 3 units.

Please explain to me how in that example, wins & losses are a reflection of your ROI

I think you missed the part where I said I understand that wins losses and ROI are not connected in the sense of if you win you get 1 ROI point or something like that.

ROI is the return on the total cash invested and returned and has NOTHING to do with wins and losses in that sense. Example you bet 1 dollar 1000 times and lose them all and you bet 1001 on one bet and you have +ROI with a record of 1-1000. That is extreme and dumb way but that is the argument here you are making when you say ROI has nothing to do with wins and losses, now reality
100 games bet at even money
Guy A wins 51 games he has +ROI
Guy B wins 49 games he has -ROI
Guy C wins 50 games he has 0 ROI
now please explain how ROI is not directly related to wins and losses.if you are lucky and win 51 out of 100 games you have +ROI, is this true or false?
if you are unlucky and win 49 games out of 100 you have -ROI is this true or false?
and KVB, that line error thing, LOL, you call me a troll and talk out the side of your neck but I think people dont go against what you are saying because A they dont know or B they like you
KVB its simple share some links give me some number examples, dont do the famous you are a troll and I already tried explaining, give me some concrete evidence or examples of the following
Line error value LOL
ROI is not connected to wins and losses

[danshan11]

The line offered by the books does not reflect a prediction of the game"s outcome rather, it's trying to predict the Public's reaction to that line.
In one instance it appears that you are trying to argue that the inevitable outcome has nothing to do with the line, or it's margin of error. But then you are also saying that the opening line, is implied probability. Aren't those mutually exclusive to each other?

Return on investment has very little to do with wins and losses, someone tried to explain this to you but it didn't seem to sink in. You don't seem dumb, so I assume you understand that somebody can go 1 and 9 and have a return of investment of say.... plus two units.
While another player can go 9 and 1 after betting - 1200 lines, he is down 3 units.

Please explain to me how in that example, wins & losses are a reflection of your ROI

the outcome of a game is correlated to the line and total sure in a large sample size of the same game repeated like a 1000 times or so, individual game not so much. One obvious example of this is key player injury during the game or extreme weather during game not predicted, and excessive turnovers.
the opening line is the implied probability of that game scenario played out over a large sample size not this game specifically.

[KVB]

...I can not help you any further if you are going to argue basics, basic definitions like ROI, without at least offering a definition of your own.

Enough Dan, you haven't shown that you've read a word I've written.

You seem to know it all, good for you.

Track your bets and why you made them.

If you understand why the line opens where it opens and moves to where it closes then you will go a long way towards winning long term

Good Luck with whatever it is you are trying to do here.

I can not help you with links and numbers if you can't understand the words first. I'm not giving you a calculator if you don't demonstrate a willingness to work with fundamentals or create some yourself. It's the whole learn how to divide before using a calculator to do it.

The need to oversimplify these concepts is harming the conversation, not helping it.

Like I said, I feel this shouldn't be in the think tank.

[danshan11]

KVB its simple share some links give me some number examples, dont do the famous you are a troll and I already tried explaining, give me some concrete evidence or examples of the following
Line error value LOL
ROI is not connected to wins and losses


hmmm.... what does this tell us folks?

[danshan11]

I make bets for 1 reason, there is value in the line. If anyone bets for any other reason, they are making a juice costing experience for them.

[danshan11]

lets just be clear on what you believe KVB
1. ROi is an indicator of skill WRONG unless you have a huge huge sample size and it still could be misleading in regards to skill
2. Line error means something and actually has value

these are the 2 things you have said and you have 100% failed to justify or confirm that info with anything.

ROI as not skill example
bet 100 games at 10 bucks lose them all down 1000
say screw it and bet 2000 on the next game and win 2000
now I bet 3000 and now have 4000 and I have a sweet ROI and does this alone say I am skilled? NO

win loss why it dont work
bet 100 games at 10 bucks win them my record is 100-0 does this alone say I am skilled? NO

line error tells me these games are off by 7 points low so i bet 100 games over they all go over now does this alone show I am skilled? HELL NO

bet 100 games at -120 and they close at an average of -145 does this alone say I am skilled? YES

[KVB]

Good job Dan, you've made the think tank a better place.

Keep it up.

You still haven't read my posts. I address different bet amounts in the ROI description, yet you ignore it and post another irrelevant example.

You started this thread with a question, you are no longer asking any questions, I can not help you anymore, just stop.

You know it all, good job.

[danshan11]

I wish you would quit writing paragraphs and still not explaining your examples of line error and ROI

[danshan11]

KVB what is ROI and how does it have anything to do with skill, please explain

[KVB]

Already explained, you are not reading. I can't simplify it anymore.

Calculate it correctly, don't adulterate it with your bullshit examples of changing bet size then say how it is wrong.

And quit saying things are only good if you have 1000 samples, because you can have 1000 samples. It's not some impossible feat.

I posted 215 MLB best in less than 6 weeks, there are plenty of large samples out here.

You want a link that shows that a positive return is a good sign?

You want a link that shows that if you can make a line that better represents reality than what the book can hang that you will win?

You want a link?

Seriously dude, between fukking up ROI to needing those links, you must be trolling.

You want a link, go to the post where I wrote it. That's your internet reference.

This discussion is over, what a fukkin waste of time.

[danshan11]

a +roi is a good sign but you understand that is cannot be used as a measuring stick of skill. I gave you examples of why it cant be used can you give me an example of why it should ? I am not just arguing to argue, I really want to know what you are talking about. but you just keep saying I already said, dont say that dude just give me a fictitious example of ROI being used to show skill that does not involve a huge sample size. Please I am actually asking for your time in sharing this info with me, I am not trying to prove you are stupid, I am trying to gather knowledge on what you see that I am missing.

[danshan11]

if you already explained how it can give me the post number cause I cannot find where you explained it to me, please I am asking not trolling or joking

[jbayko]

I think you missed the part where I said I understand that wins losses and ROI are not connected in the sense of if you win you get 1 ROI point or something like that.

ROI is the return on the total cash invested and returned and has NOTHING to do with wins and losses in that sense. Example you bet 1 dollar 1000 times and lose them all and you bet 1001 on one bet and you have +ROI with a record of 1-1000. That is extreme and dumb way but that is the argument here you are making when you say ROI has nothing to do with wins and losses, now reality
100 games bet at even money
Guy A wins 51 games he has +ROI
Guy B wins 49 games he has -ROI
Guy C wins 50 games he has 0 ROI
now please explain how ROI is not directly related to wins and losses.if you are lucky and win 51 out of 100 games you have +ROI, is this true or false?
if you are unlucky and win 49 games out of 100 you have -ROI is this true or false?
and KVB, that line error thing, LOL, you call me a troll and talk out the side of your neck but I think people dont go against what you are saying because A they dont know or B they like you
KVB its simple share some links give me some number examples, dont do the famous you are a troll and I already tried explaining, give me some concrete evidence or examples of the following
Line error value LOL
ROI is not connected to wins and losses

Thank you! This is what we've been trying to tell you. But you insist on saying wins and losses are "directly related" to ROI. Therefore, I think I was half right in my first reply to you. I was correct that we're speaking a different language. But I was incorrect in what terms were being misused here. Please Google "direct relationship". It means that two variables are related via a constant ratio. In other words, there is a linear relationship between the two. If you want to say that wins and losses are loosely related to ROI, I can buy into that. But all of your examples seem o be assuming fixed stakes and fixed odds for every wager. That's not reality. And the more variance there is in an individual's odds and stakes, the looser the relationship between wins/losses and ROI. There is no direct relationship between W/L and ROI. Period.

[Lenny Zefflin]

Danshan11 please stop. I can't bear reading this anymore.

[danshan11]

what part can you not handle?

[danshan11]

Thank you! This is what we've been trying to tell you. But you insist on saying wins and losses are "directly related" to ROI. Therefore, I think I was half right in my first reply to you. I was correct that we're speaking a different language. But I was incorrect in what terms were being misused here. Please Google "direct relationship". It means that two variables are related via a constant ratio. In other words, there is a linear relationship between the two. If you want to say that wins and losses are loosely related to ROI, I can buy into that. But all of your examples seem o be assuming fixed stakes and fixed odds for every wager. That's not reality. And the more variance there is in an individual's odds and stakes, the looser the relationship between wins/losses and ROI. There is no direct relationship between W/L and ROI. Period.

My cousin bet 30 -40 games and killed it his ROI is 45% or higher is he skilled with that info?

[jbayko]

My cousin bet 30 -40 games and killed it his ROI is 45% or higher is he skilled with that info?

Please explain what that has to do with my post.

[danshan11]

I am sorry maybe nothing. do you believe ROI is a measure of skill in a small less than 1000 sample size?

[jbayko]

I am sorry maybe nothing. do you believe ROI is a measure of skill in a small less than 1000 sample size?

I could answer that question but I'm not going to because it's irrelevant to my point and I don't want to distract from it. To stay on subject;: There is no direct relationship between W/L and ROI.