A question about variance

[Gaze73]

Imagine 3 teams that close at -200/1.5

Team A wins 5:0
Team B draws 0:0
Team C loses 0:5

The question is, did they all have a ~66% chance to win before kickoff? If not, why did they have the same odds?

[TommieGunshot]

I would believe they all had a less than 66% chance of winning. If it was really 66%, that means the book would only have 0.5% edge on any bets on the favorites. The sportsbooks almost never want something that small.

[Gaze73]

66 or 63, the juice doesn't matter here. What matters is whether it's possible to tell which team is more likely to fail or win at the same closing odds.

[KVB]

66 or 63, the juice doesn't matter here. What matters is whether it's possible to tell which team is more likely to fail or win at the same closing odds.

In my opinion you need context. While we want every game to be independant from each other they really aren't. I think it helps too know the direction the teams are going and whether or not the market is forced to adjust because of that type of momentum.

This way you can tell if they are holding a line that should move. You might even be able to pick up on whether or not the book is taking a position.

Not sure I'd trust line movement either, there's room for deception there.

[Gaze73]

Team A has form 12-0-0, today's odds are 1.5
Team B has form 5-3-4, today's odds are 1.5

Sure, the opponent matters, but which of these two would you rather bet on? If team A won 3:0 and team B lost 0:2 it would be no surprise at all, yet according to the efficient market they are almost equally likely to get those results.

[u21c3f6]

Based on the limited info provided I would rather wager on team B.

I have written in previous posts that the “efficient “ market theory isn’t as efficient as most everyone thinks. The way to “win” long-term is to make wagers that have a greater chance to win than their odds indicate.

For me the question always is: which teams line/odds are likely to be too low and which teams line/odds are likely to be too high?

You can find many threads here that question the odds as suspicious on a given event. Most will question why one team’s line/odds appear too low as they thought the line/odds should be higher and therefore pass wagering on the event because something is suspicious. Then when the team with the suspicious line/odds wins, everyone writes how they were correct to pass the event. It doesn’t occur to most that they should have wagered on this event because IMO they were trying to pick the “winner” which usually doesn’t produce a long-term profit for most.

This is really a very large discussion which I have attempted to discuss previously. I am not an advocate of “beat the closing line” and have written about this many times in previous threads. There was one thread in particular where a poster attempted to show that beating the closing line was the way to go. He in fact managed to beat the closing line in each of his 3 attempts. Unfortunately, he was losing approx 8-10% of his wagered money while beating the closing line! I do not find that unusual which I have posted about previously.

Joe.

[Gaze73]

Yeah, I used to follow people who beat the closing line and I also lost money. So I decided to fade overbet favorites, but sometimes they are too strong.

E.g. Kongsvinger in soccer - their recent results are 6:0 5:0 4:0 7:0 3:1 - in the last game their highest opening odds were 1.9, closing odds 1.4. Why did crazy bookies open it at 1.8-1.9 when their home form was 10-0? Was the 1.4 closing line efficient long-term? Who knows.

[Roger T. Bannon]

You don't determine which was more likely to win by the closing price after the fact. That is known. You just don't know why the outcome took place if it was unexpected.

For that, you have to drill down and get the stats or just make a judgment on such things as whether or not one team was flat or underperformed. That is where variance comes in.

[d2bets]

Good discussion here.

Still a believer generally in beating the closing line, but that sure isn't working this year so far in the NBA. The line movement has been spectacularly wrong. Now, is this variance? Or is this a super inefficient market?

[Roger T. Bannon]

It is just the market getting things wrong. Most of the market moves are based off of the same metrics and those are missing the mark. CLV has been having some problems in a lot of sports from what I have been reading.

[d2bets]

It is just the market getting things wrong. Most of the market moves are based off of the same metrics and those are missing the mark. CLV has been having some problems in a lot of sports from what I have been reading.

Any opinions on cause? Could it be related to the massive market expansion with legalization and so much more money pouring into the market? I would assume that it would sort itself out...but maybe not? This definitely bears watching and discussing.

[Roger T. Bannon]

No, in the case of the NBA it is just an inability to predict the future. You have a lot of changes that take place and stuff that cannot be measured and there will be certain years when there is a lot of stuff that happens that cannot be predicted.

I think MLB bettors have had a lot of problems with CLV in the past couple of season and particularly this year. It could be that there is just too much of the same stuff being used and it has lost all value. The more a market becomes dominated by a certain line of thought, the more other lines of thought will have more value.

[d2bets]

No, in the case of the NBA it is just an inability to predict the future. You have a lot of changes that take place and stuff that cannot be measured and there will be certain years when there is a lot of stuff that happens that cannot be predicted.

I think MLB bettors have had a lot of problems with CLV in the past couple of season and particularly this year. It could be that there is just too much of the same stuff being used and it has lost all value. The more a market becomes dominated by a certain line of thought, the more other lines of thought will have more value.

Well that's always the case, but it's something more. My impression is that opening lines seem more accurate than closing lines. This is not normal. Million dollar question is whether this persists.

[KVB]

Good discussion here.

Still a believer generally in beating the closing line, but that sure isn't working this year so far in the NBA. The line movement has been spectacularly wrong. Now, is this variance? Or is this a super inefficient market?

Beware of a recency bias in your thinking.

That's the first step. The next is to stop trying to seek efficiency there, efficiency is purposefully not achieved in many instances. It's simply not always as efficient as you might think.

As sportsbetting grows across the nation expect more and more "volatility" or what is better translated to more extreme variance.

It's the nature of the beast.

But don't let recency bias get to you, that's my main point.

[Roger T. Bannon]

Well that's always the case, but it's something more. My impression is that opening lines seem more accurate than closing lines. This is not normal. Million dollar question is whether this persists.

You should expect at the beginning of the season a lot of problems with CLV because there is a lot of changes that have to be accounted for and people are just not good at accounting for those changes. As the season goes they start to properly value things and then you get CLV.

In the case of baseball, there may be some overvaluation of something that is persistently creating negative value. As an example, if home field advantage is being systemically overvalued, you would get problems with CLV and it might take a while to figure out the problem.

[d2bets]

Beware of a recency bias in your thinking.

That's the first step. The next is to stop trying to seek efficiency there, efficiency is purposefully not achieved in many instances. It's simply not always as efficient as you might think.

As sportsbetting grows across the nation expect more and more "volatility" or what is better translated to more extreme variance.

It's the nature of the beast.

But don't let recency bias get to you, that's my main point.

Agree on the recency bias. It's still early in the season, so I don't want to draw too many conclusions. That said, last year was also pretty "poor" for line movement. I just wonder if there is some misdirection/fakeout being achieved in some of these line moves. It's very interesting.

[d2bets]

You should expect at the beginning of the season a lot of problems with CLV because there is a lot of changes that have to be accounted for and people are just not good at accounting for those changes. As the season goes they start to properly value things and then you get CLV.

In the case of baseball, there may be some overvaluation of something that is persistently creating negative value. As an example, if home field advantage is being systemically overvalued, you would get problems with CLV and it might take a while to figure out the problem.

Good stuff, thanks. Although you would think that early in the season there might also be a problem with OLV (opening line value). But it seems like the linesmakers are doing a damn fine job of setting lines.

[Roger T. Bannon]

Agree on the recency bias. It's still early in the season, so I don't want to draw too many conclusions. That said, last year was also pretty "poor" for line movement. I just wonder if there is some misdirection/fakeout being achieved in some of these line moves. It's very interesting.

The world is always changing. The future never gets any easier to predict. Some people get better, some people get worse.

[Roger T. Bannon]

Good stuff, thanks. Although you would think that early in the season there might also be a problem with OLV (opening line value). But it seems like the linesmakers are doing a damn fine job of setting lines.

That is not really the case. The openers are always going to be weak. The issue really is that the market is just having a problem valuing the different changes that have taken place. New players, new coaches, players being better than expected, players being worse than expected, etc, etc. The market is just getting it wrong.

[Roger T. Bannon]

One specific recent change in the NBA that I could see being especially problematic is the tendency now for teams to sit players out. How a player is going to perform as part of the game plan with practice and team strategy is potentially going to be far different than what how that player will perform coming off the bench. You could very easily have that player have quite a few points more in value or potentially that value could shift to another player in unexpected ways.

[Gaze73]

I also subscribe to the idea that the opening lines are more accurate than closing lines. When a book opens an NBA fav at -3.5 and the line moves to -6.5, the squarest thing you could do is follow the crowd on this "cert". There's a good chance the true line should be -2.5, and taking the dog +6.5 gives you a 4 point edge, because the books shaded a point in the opening lines. This doesn't apply to the cases where there's a significant lineup change.

[d2bets]

That is not really the case. The openers are always going to be weak. The issue really is that the market is just having a problem valuing the different changes that have taken place. New players, new coaches, players being better than expected, players being worse than expected, etc, etc. The market is just getting it wrong.

OK, but my question is which line is getting it closer to the true number (50/50 either way) -- the opener or the closer? Normally, I would say the closer for sure. But I'm not sure that's always the case. If not, I'm just not sure if it's because of bad info/modeling or some sort of intentional misdirection, or what. As gaze was saying, I'm almost wondering if these books are moving on air trying to get people to follow the tail of the line movement into a bad number. Pretty advanced stuff. But there's big stakes now with the legalized markets.

[d2bets]

I also subscribe to the idea that the opening lines are more accurate than closing lines. When a book opens an NBA fav at -3.5 and the line moves to -6.5, the squarest thing you could do is follow the crowd on this "cert". There's a good chance the true line should be -2.5, and taking the dog +6.5 gives you a 4 point edge, because the books shaded a point in the opening lines. This doesn't apply to the cases where there's a significant lineup change.

By and large, that has typically not been the case historically. Absolutely not. But in this new era of legalized betting, the world is turned upside down and maybe there is some of that happening.

[Roger T. Bannon]

I think you are ascribing a lot of talents that do not exist.

[Roger T. Bannon]

If we were to say that the openers were better than the closers or the entire gambling market, what you would find is very little line movement. They would just be too tough to beat. I think the math guys have a measurement called the mean squared error or some such thing which is basically how close the actual result came to the forecasted result.

If you went back and looked at the mean squared error for openers, you would likely find that they were no better than your average gambler's. Basically the openers have been no better than the closers because the results have been too unpredictable. All lines have been bad.

[d2bets]

If we were to say that the openers were better than the closers or the entire gambling market, what you would find is very little line movement. They would just be too tough to beat. I think the math guys have a measurement called the mean squared error or some such thing which is basically how close the actual result came to the forecasted result.

If you went back and looked at the mean squared error for openers, you would likely find that they were no better than your average gambler's. Basically the openers have been no better than the closers because the results have been too unpredictable. All lines have been bad.

In theory, yes. But in reality, not necessarily the case if there is manipulation/misdirection.

[Roger T. Bannon]

What kind of misdirection do you think is happening?

[d2bets]

What kind of misdirection do you think is happening?

I don't know, that's why I'm asking lol.

Or, could it be that public square money is moving the line without cause?

[Roger T. Bannon]

I don't know, that's why I'm asking lol.

Or, could it be that public square money is moving the line without cause?

LOL. Misdirection basically applies to moving a line to set up a bet. That is done by some pro bettors but not going to be done by sportsbooks.

There is a lot of imagination about grand schemes. But sports betting is pretty simple stuff. You are either right or wrong. There is not a whole of room for in betweens.

So what you have is just a lot of bad bets.

I've had the same problem this year with college football. As I evaluate my bets, I've had some really good ones, I've had some really bad ones so maybe +2 positive but down substantially with positive CLV. Most of it is just stuff I could not have forecasted. Maybe a little bad luck. That is basically what you have with the start of the NBA season and many have experienced the same in college football this season from what I have read.

You can get a lot of wild results when you have a lot of uncertainties. That is where the opportunity lies but not that easy to cash in.

[Gaze73]

I would say that the main reason the closers are more accurate than openers is simply because you know 100% confirmed lineups with the closers. When a superstar player won't be playing, the line can move by 20%. If that's not the case, are line makers worse than the public? Why would they make lines that sometimes move by 50% by closing time? They can't be that bad at their job.

[StackinGreen]

I would say that the main reason the closers are more accurate than openers is simply because you know 100% confirmed lineups with the closers. When a superstar player won't be playing, the line can move by 20%. If that's not the case, are line makers worse than the public? Why would they make lines that sometimes move by 50% by closing time? They can't be that bad at their job.

The line makers are between joe schmoe and the sharpies, but it is true that usually only comes out for 1-3 games a week, proving the point that they are better, but not very good. They use versions of models just like Sagarin, as you'll notice the line is rarely straying from the pure points type model.

If you put me against openers I would make the easiest living of the year, particularly with MLB and NFL - I could live a good American life just based on NFL if that was an opportunity. But books/linemakers know better than that, and limit early bets and sharps.

[StackinGreen]

I would say that the main reason the closers are more accurate than openers is simply because you know 100% confirmed lineups with the closers. When a superstar player won't be playing, the line can move by 20%. If that's not the case, are line makers worse than the public? Why would they make lines that sometimes move by 50% by closing time? They can't be that bad at their job.

This and pure volume/books knowing who is on what side (sharps vs volume of bets) - of course, they go hand in hand on some level.

[Roger T. Bannon]

I would say that the main reason the closers are more accurate than openers is simply because you know 100% confirmed lineups with the closers. When a superstar player won't be playing, the line can move by 20%. If that's not the case, are line makers worse than the public? Why would they make lines that sometimes move by 50% by closing time? They can't be that bad at their job.

That is part of it. That is why closers are better than openers, more information is available to bet on and more people have bet. Yes, linesmakers are not very good at their jobs. They do miss by wide margins. That is why they put limits on openers. The linesmaker is way better than the average bettor but nearly as good as winning bettors.

[Roger T. Bannon]

Ultimately, you can call all of this variance because if you get CLV you are likely to win so it means that the market is thinking along the same line and the results are just not working out. In the longest of terms, you will wind up winning even if in this brief period of time, you lose.

However, if you are looking at a situation like baseball where people are losing over an entire season or two. It might not be variance. It might be that the end has been reached by models for some time period. Probably what is happening is that a lot of people that used to win aren't able to anymore.

In my case, I can say that what has happened is beyond variance. Even though I have had the same CLV as in the past, I will not win going forward if this is the new world due to transfers and such. So I will need to reduce my number of bets to win in the future if next year is not as predictable. I deserved to lose this year.

[coolguy73739]

BTCL is crap for all games.. Some games are already pre-decided.. Pls don't spend your energy analysing line movement, CLV and all craps.. Bookies have already pre-decided on these games.. Run a MODEL and know where true number is going to be..
without a great model, you can never win more than 60%.. you will be illusioned and thats what bookies want you to be in..