The Daily bets, Banter & Accumulators thread IV

Burial. · 22-10-2017 07:36PM

Any of ye lads use stat based models to inform your bets lads? It's been something I have been meaning to get into with ages but I've always won just enough to keep myself from doing it....however for the most part I'm being destroyed this season. Might be just variance but regardless I think I'd enjoy getting a bit more professional about it as long as it doesn't fry my brain.

I have googled some articles and that but even though I'm very fond of poker and betting they seem extremely complicated. I do get the gist of finding the required data, scraping it and plugging it into something, and then making formulas to extract an implied probability from the data and then compare it to the bookies but the computer side to it has me frazzled a little as well as making it reliable. I tried to scrape some league tables into Excel to get me started and Excel nearly broke! I have been informed you can make handy ones in Excel that'll help to some regard but I haven't used Excel since I did the ECDL course in TY. A lad I used be in a poker skype group with used proper programming language and that for his which he said was a million times more straight forward than Excel once you learn the coding language....however I don't think that'll be easy for me. R I think it was. Unfortunately he's since disappeared.

Would be interested in what ye lads do.

mailburner · 22-10-2017 07:44PM

lazio could score 5/6 here if they can be bothered, the cagliari players will be well knackered late on, lazio took most of the first half off
after scoring that first goal so who knows?

Autochange · 22-10-2017 07:54PM

mailburner wrote: »

lazio could score 5/6 here if they can be bothered, the cagliari players will be well knackered late on, lazio took most of the first half off
after scoring that first goal so who knows?

Theres the third.So early in the second half too. Hopefully they keep the foot down .

mailburner · 22-10-2017 08:34PM

mailburner wrote: »

went a bit mad on the lazio cagliari game for goals, good fixture for goals but priced accordingly

big bet for me on lazio over 2.5 goals at 4/5 888
lazio over 1.5 1st half goals 9/5 bfsp
small stakes on
lazio over 2.5 1st half goals 7/1 bfsp
over 4.5 lazio goals 15/2 bfsp

They tend to be fast starters and having played on thurs I could see them
trying to put cagliari away early, just hoping that if they go 2/3 up they don't take their foot off the gas

pretty much over after lazio scored the third but two main ones in.

Copper_pipe · 22-10-2017 08:52PM

Burial. wrote: »

Any of ye lads use stat based models to inform your bets lads? It's been something I have been meaning to get into with ages but I've always won just enough to keep myself from doing it....however for the most part I'm being destroyed this season. Might be just variance but regardless I think I'd enjoy getting a bit more professional about it as long as it doesn't fry my brain.

I have googled some articles and that but even though I'm very fond of poker and betting they seem extremely complicated. I do get the gist of finding the required data, scraping it and plugging it into something, and then making formulas to extract an implied probability from the data and then compare it to the bookies but the computer side to it has me frazzled a little as well as making it reliable. I tried to scrape some league tables into Excel to get me started and Excel nearly broke! I have been informed you can make handy ones in Excel that'll help to some regard but I haven't used Excel since I did the ECDL course in TY. A lad I used be in a poker skype group with used proper programming language and that for his which he said was a million times more straight forward than Excel once you learn the coding language....however I don't think that'll be easy for me. R I think it was. Unfortunately he's since disappeared.

Would be interested in what ye lads do.

Poission Distribution using Excel and Soccerway would be a start. Should be plenty articles on that

Autochange · 22-10-2017 08:55PM

Autochange wrote: »

Iv gone for Immobile (Lazio) and Cavani (PSG) both anytime scorers in a double at 1.9

Both are the top scorers in their leagues.

Immobile with an early penalty and Cavani with a belter of a free which was the last kick off the game.

Winner

mailburner · 23-10-2017 05:28PM

oyarzabal anytime 3.3 888

scored in his last five, nice price imo...am thinking sean might have a little
on him again also :cool:

mailburner · 23-10-2017 05:38PM

like braga over 1.5 at 2.23 888 usually good for a couple of goals away from home

stuck it in a double with sociedad over 1.5 also at 3.66 888

both kick off at 7

mailburner · 23-10-2017 07:58PM

mailburner wrote: »

oyarzabal anytime 3.3 888

scored in his last five, nice price imo...am thinking sean might have a little
on him again also :cool:

no joy

mailburner wrote: »

like braga over 1.5 at 2.23 888 usually good for a couple of goals away from home

stuck it in a double with sociedad over 1.5 also at 3.66 888

89 mins to score a goal and should have got one in injury time, bunch of wasters.

inforfun · 23-10-2017 10:21PM

Dutch cup, at home against some 5th tier amateur team.
No matter who to put on the pitch, anything less than 5 goals is a failure.

Autochange · 23-10-2017 11:42PM

Oostende v Charleroi at 745 in the Belgian juliper league

Charleroi at 11/5 seem overpriced. 2nd in the league with the second best away form Vs Oostende who are bottom of the table with 5 points from 11 games. Zero of those points have come at home either.

I'll be on Charleroi at 7/10 double chance and win at 11/5

Division · 24-10-2017 10:00AM

Burial. wrote: »

Any of ye lads use stat based models to inform your bets lads? It's been something I have been meaning to get into with ages but I've always won just enough to keep myself from doing it....however for the most part I'm being destroyed this season. Might be just variance but regardless I think I'd enjoy getting a bit more professional about it as long as it doesn't fry my brain.

I have googled some articles and that but even though I'm very fond of poker and betting they seem extremely complicated. I do get the gist of finding the required data, scraping it and plugging it into something, and then making formulas to extract an implied probability from the data and then compare it to the bookies but the computer side to it has me frazzled a little as well as making it reliable. I tried to scrape some league tables into Excel to get me started and Excel nearly broke! I have been informed you can make handy ones in Excel that'll help to some regard but I haven't used Excel since I did the ECDL course in TY. A lad I used be in a poker skype group with used proper programming language and that for his which he said was a million times more straight forward than Excel once you learn the coding language....however I don't think that'll be easy for me. R I think it was. Unfortunately he's since disappeared.

Would be interested in what ye lads do.

This is a very interesting topic. Before you get to the stage of programming you have to decide what data you will use and how you will assign importance/weighting to it.

Most models have a program but the input data is simply goals scored and conceded which doesn't go into enough detail imo. At this stage of the season Everton have played a lot of the top teams while other teams have had a much easier run of fixtures so there has to be a way to put the results into context.

You don't need programming skills yet until you know what you want to do regarding that aspect (among many other factors).

Take possession for example - a team may average 55% and have less points per game than a team that has 40% average possession for a number of reasons such as counter attacking ability and having a striker with a high strike rate.

Liverpool have the second highest shots per game (18.7) but Watford have 11.7 shots per game but have more points so how will you interpret this?

Burnley win more aerial duels than any other team in the league - what do you make of this?

Southampton have the 4th highest possession stats in the league but can't convert this into goals so if you weight possession too highly you will give Southampton too high a chance of winning.

Backstreet Moyes · 24-10-2017 10:54AM

Personally myself I would use it to predict results to look into the market of over/under 2.5 goals.

Also would be interested in using it for card markets.

Backstreet Moyes · 24-10-2017 10:55AM

Copper_pipe wrote: »

Poission Distribution using Excel and Soccerway would be a start. Should be plenty articles on that

Can you scrape the data from Soccerway?

Burial. · 24-10-2017 12:46PM

Nuts102 wrote: »

Can you scrape the data from Soccerway?

I know there are Chrome extensions that will basically copy and paste tables and stat tables from anywhere but I'd love to be able to get something that could refresh after every game with a click of a button instead of having to copy/paste every time. It can be done as well but my tech skills are brutal.

I have decided to set up a proper bet tracker with Excel...found one online that'll calculate things like ROI and profit/loss per sport as well as average odds and that.

Of course I'd love to have a fully automated model that'll spit out picks but that's not realistic at all now or perhaps ever...I'd love to have a little system that'll give a small bit of guidance that'll be obviously better than just going through the tabs on Flashscore and making picks based on simple stats and intuition.

Stuff like ELO Ratings, Poisson, Dixon Coles is flying way over my head right now for the most part. That ELO rating is nice though as it tells a lot more of how a game went rather than just by the result.

Backstreet Moyes · 24-10-2017 01:09PM

Going to try download a free sql database and look into scrapping data after that

Copper_pipe · 24-10-2017 01:11PM

Nuts102 wrote: »

Going to try download a free sql database and look into scrapping data after that

Keep me posted on how you get on with that. 538 are models id pay attention to for NBA and NFL

Backstreet Moyes · 24-10-2017 01:29PM

Copper_pipe wrote: »

Keep me posted on how you get on with that. 538 are models id pay attention to for NBA and NFL

Work in I.T. so know how to use databases and write queries.

Downloading a free version now and I will see what that is like when it downloads.

https://www.microsoft.com/en-ie/sql-server/sql-server-downloads

Hollister11 · 24-10-2017 06:02PM

Untied to win @ 1.61

Leicester Over 1.5 Team Goals @ 1.57

zimmermania · 24-10-2017 06:19PM

Not much of a football punter (often thought about betting on it),do not know enough about it but may decide to bet on football if the bookmakers get too greedy with the horseracing odds.

Someone writing in the Racing Post today made a case for backing o a goal to be scored after the 73rd minute in the Utd- Swansea game so I took his advice and had a bet at 10/11.

jbt123 · 24-10-2017 06:35PM

Swansea - Man Utd

Utd to score 2 or more & 6+ corners 5/2

BTTS, Both have 3+ corners & Both have 2 cards each.

Both taken under Paddy specials.

Good luck all.

Backstreet Moyes · 24-10-2017 08:07PM

Copper_pipe wrote: »

Keep me posted on how you get on with that. 538 are models id pay attention to for NBA and NFL

Got SQL Server on to the laptop.

My next step is getting the data into SQL Server.

If I can get the data into Excel, I can load it into a table and then look at ways to analyse the data.

Then daily I could keep updating the table.

Anyone know a way to scrape the data from Soccerway into Excel?

Fitz* · 24-10-2017 08:11PM

Nuts102 wrote: »

Got SQL Server on to the laptop.

My next step is getting the data into SQL Server.

If I can get the data into Excel, I can load it into a table and then look at ways to analyse the data.

Then daily I could keep updating the table.

Anyone know a way to scrape the data from Soccerway into Excel?

Also interesting me this idea. Have wanted to get around it it for a while.

Qlikview, a front end SQL application can load in data from Excel, SQL server & websites. There's a handy built in wizard for it too.

Ideally, you would want all the data in one source so you could trust it fully. An automatic daily load of data into your system....I'm not sure of ways to do that at the minute

Backstreet Moyes · 24-10-2017 08:36PM

FitzShane wrote: »

Also interesting me this idea. Have wanted to get around it it for a while.

Qlikview, a front end SQL application can load in data from Excel, SQL server & websites. There's a handy built in wizard for it too.

Ideally, you would want all the data in one source so you could trust it fully. An automatic daily load of data into your system....I'm not sure of ways to do that at the minute

I have used Qlikview in college and it is good for graphical viewing.

Once I have the data in Excel I can load into a table in SQL Server and to Qlikview and play around with both.

I am looking at some free scraping tools now and see if I can get something into Excel tonight or in the morning.

Hollister11 · 24-10-2017 08:39PM

Hollister11 wrote: »

United to win @ 1.61

Leicester Over 1.5 Team Goals @ 1.57

Double winner

jbt123 · 24-10-2017 08:41PM

jbt123 wrote: »

Swansea - Man Utd

Utd to score 2 or more & 6+ corners 5/2

BTTS, Both have 3+ corners & Both have 2 cards each.

Both taken under Paddy specials.

Good luck all.

One United corner short for the 5/2 shot

A Swansea goal short for the 14/1 shot.

seanhynes · 24-10-2017 08:52PM

City will go score a few in et no doubt :rolleyes:

mailburner · 24-10-2017 09:01PM

45/1 for the 0-0 in the city game :eek: seemed pretty unlikely to be fair when they're putting 5/6 past prem teams and silva aguero jesus up front

Danick · 24-10-2017 09:53PM

seanhynes wrote: »

City will go score a few in et no doubt :rolleyes:

Thanks for mentioning it - threw a few quid on them for et but not luck so backed under 3.5 goals for Wolves and it came in - I would never have even thought about this game if you hadn't mentioned - like you said - the odds on this ending in a draw were crazy - I didn't even bother including it in my games on flashsport app - anyway thanks for this post - you saved my night - I had been on a bad streak all evening

Danick · 24-10-2017 09:54PM

seanhynes wrote: »

City will go score a few in et no doubt :rolleyes:

Thanks for mentioning it - threw a few quid on them for et but no luck so backed under 3.5 penalties for Wolves and it came in - I would never have even thought about this game if you hadn't mentioned it - like you said - the odds on this ending in a draw were crazy - I didn't even bother including it in my games on flashsport app - anyway thanks for this post - you saved my night - I had been on a bad streak all evening

The Daily bets, Banter & Accumulators thread IV

Comments