calculating soccer totals

[luegofuego]

this puzzles me and i can't seem to grasp my head around how the markets calculate it.

there's some games this afternoon for example, where the market says the avg totals is at ~2.4. but when i check the stats of both teams, the home team concedes 1.3 goals on average and score 1.3 goals on average for an avg total of 2.6, while the away team scores 1.9 on average and concedes 1.6 on average for a total of 3.5.

the league average totals after 80 games played so far is 2.65, so even reverted to the mean, it would end up higher than 2.5.

50% of the home teams games have ended up under and 50% have ended up over. 70% of the away teams games have ended up over and only 30% under.

so whichever way you jerk around with the stats, i get a significantly higher total than the market is telling me. if this was just on this one game, that would be one thing, but my way of anticipating totals consistently end up much higher than what the market ends up at. what am i doing wrong?

[Pokerjoe]

Soccer totals require the poisson function. But in any particular case, it might be a particular situation. Leading scorer injured; one team needing only a tie; weather, etc. Also, the more evenly matched the sides, the lower the total.

[RickySteve]

Soccer totals require the poisson function.

As a rough first guess maybe. It takes a lot more work to get accurate numbers.

[Justin7]

Four thoughts. The mean is not nearly as important as the median (as with any spread or total). Second, the scoring distribution often changes after the first goal. A good rule of thumb for elite tournaments where both teams want to win is that the "pace" of a game goes up 20-25% after the first goal. Third, you need to normalize your distributions relative to the league average. If a league averages 3 goals a game, and both teams score and allow 1.25 goals per game, that game's total will be much lower than 2.5. Fourth, you need to be aware of injuries. Roster changes can have a major impact on spreads, and often more so on totals.

[Ultralord]

It is impossible to calculate totals in soccer. You need to know the teams and their playing style. For example Germany usually uses an offensive style and has an average defense. Same for Argentina and Brazil. Other teams like Serbia, Algeria, Italy etc use an defensive style. You have to compare the teams, their coaches, their playing style etc. You simply can not calculate totals in soccer without any knowledge.

[luegofuego]

Four thoughts. The mean is not nearly as important as the median (as with any spread or total). Second, the scoring distribution often changes after the first goal. A good rule of thumb for elite tournaments where both teams want to win is that the "pace" of a game goes up 20-25% after the first goal. Third, you need to normalize your distributions relative to the league average. If a league averages 3 goals a game, and both teams score and allow 1.25 goals per game, that game's total will be much lower than 2.5. Fourth, you need to be aware of injuries. Roster changes can have a major impact on spreads, and often more so on totals.

this is good info, thanks!

[luegofuego]

It is impossible to calculate totals in soccer. You need to know the teams and their playing style. For example Germany usually uses an offensive style and has an average defense. Same for Argentina and Brazil. Other teams like Serbia, Algeria, Italy etc use an defensive style. You have to compare the teams, their coaches, their playing style etc. You simply can not calculate totals in soccer without any knowledge.

this is an absurd statement in a ton of different ways. if you can't figure out what a numeric representation of what a "defensive style" might look like, i don't know what to say really...

[Pokerjoe]

Originally Posted by Ultralord
It is impossible to calculate totals in soccer.

Then, how do books calculate totals in soccer every day?

[Pokerjoe]

As a rough first guess maybe. It takes a lot more work to get accurate numbers.

Duh.

That's why I included, in my post, such thoughts as "But in any particular case, it might be a particular situation. Leading scorer injured; one team needing only a tie; weather, etc. Also, the more evenly matched the sides, the lower the total."

[saintjames]

does anybody know the math for calculating soccer pointspreads?

[Pokerjoe]

Now that I think about it, I'm pretty sure there was a pinnacle pulse article on this subject. You can google it.

[necro]

http://www.bookmakers-code.net/report.html?id=rss

Although above site looks like typical 'scam', I managed to get their .pdf file which they are selling for muchos money. If someone is interested, I can send it for free. It can't hurt

[ZXCVBNM]

It is impossible to calculate totals in soccer. You need to know the teams and their playing style. For example Germany usually uses an offensive style and has an average defense. Same for Argentina and Brazil. Other teams like Serbia, Algeria, Italy etc use an defensive style. You have to compare the teams, their coaches, their playing style etc. You simply can not calculate totals in soccer without any knowledge.

Precisely. This is a very delicate total always. It's either 2 in low-scoring leagues (France), 2.5 in normal leagues (England), and 3 in big scoring leagues (Germany).

It's about chasing the match, and how the home team likes to play. You must look at the records. Some away teams have no draws and like to go for it all, this will always result in higher totals. Some teams may be happy with a draw. Some home teams like to play it tight and their first priority is not to let the other team score before scoring themselves.

This is why depending on matchups the same team can go from 2.5, to 2/2.5, to 3 playing at home against different opponents. I'm sure this is wrong, but I think the lines must be set on "feel" more than anything, because there are no numbers to back up the styles involved. Why is Fulham/Birmingham juiced under, and why is Fulham/Portsmouth set even on both sides?

[WDR]

http://www.bookmakers-code.net/report.html?id=rss

Although above site looks like typical 'scam', I managed to get their .pdf file which they are selling for muchos money. If someone is interested, I can send it for free. It can't hurt

please pm download link thanks

[emmerva]

http://www.bookmakers-code.net/report.html?id=rss

Although above site looks like typical 'scam', I managed to get their .pdf file which they are selling for muchos money. If someone is interested, I can send it for free. It can't hurt

Please, PM me the bookmarkers code ebook.
thanks

[infamousbacardi]

http://www.bookmakers-code.net/report.html?id=rss

Although above site looks like typical 'scam', I managed to get their .pdf file which they are selling for muchos money. If someone is interested, I can send it for free. It can't hurt

Please Pm me the .pdf. Thnaks!

[That Foreign Guy]

Is there anything useful in this book? I'm actually not averse to paying for things if they are good, but this site looks super scammy.

[basilis]

could i have a link of the pdf as well?
cheers

[v-zero]

Football (I'm from the UK, soccer if you like) modelling is my bread and butter, and I can tell you that there is no real statistical correlation between average totals for certain teams and their predicted totals in individual games. The totals line is a result of the way in which the game evolves, or rather the way in which it is expected to evolve. Close games between two equally matched teams are usually cagey, whereas games between mismatched sides are very open, as the underdog is usually looking to get a lucky goal, and the dominant side is looking to stamp their mark firmly. As such the expected total in a game is set as a function of the difference in strength between the two sides. If you can calculate a value for the expected goal difference in the game then within an individual league this should be proportional to the total.

The football totals markets are quite efficient, but not as efficient as the match odds markets, as most people betting tend to bend the lines, misunderstanding the way that totals occur.

[jetsjets1028]

nice win by usa women yesterday early morning!!!

[CHUBNUT]

goal averages have nothing to do with soccer totals except when high scoring teams take on very low scoring teams. realistically the best side will have to score all 3 goals giving the under a fair chance but for some reason the books get fearful and overvalue the over. I suggest people concentrate more on clean sheet and fail to score stats and compare those to the league average. its also essential to beat others to rouge numbers meaning work has to be completed as early as possible then you have a safety net of sorts if later info goes against you.

[whodat1976]

Changing trends in totals in leagues during the season

Has anyone noted in certain leagues substantial changes in the total number of goals in certain leagues that take place during the season and that the bookies dont take into account?

[cmaulsby]

Could you pm the pdf? thanks

[wtt0315]

you guys really the orginal posts were from 2 years ago. I am not sure he is even around