Football in play

Football, Soccer - whatever you call it. It is the beautiful game.
Post Reply
gstar77
Posts: 34
Joined: Fri Dec 04, 2020 10:52 pm

firlandsfarm wrote:
Fri Feb 12, 2021 11:55 am
gstar77 wrote:
Mon Feb 08, 2021 6:34 pm
As we know the exchange is efficient and nearly all markets are near 100% perfect and show true odds.
Derek27 wrote:
Fri Feb 12, 2021 4:30 am
Toss a coin and offer 4/5 heads, 5/4 tails and that's a 100% overround, but they're not true odds.
gstar, please tell me you are not mixing a 100% book with a 100% market efficiency! Derek is quite right in pointing out that a 100% book does not mean that the market is 100% efficient in it's implied probabilities. If that were the case there would never be value to be had in a market with a 100% book.

I do not have Bf odds for football at kick-off but if I take the maximum Home, Away and Draw odds from the football-data.co.uk spreadsheets for the last 5 Premier League seasons (including this season so far) as being a reasonable approximation to the Bf odds I get an average book of 99.73%, very close to a pure 100% book. The return (ROI) I get if I place a bet on all 3 outcomes in all matches is +6.38%, -6.02% and -0.09% respectively. If the odds were 100% efficient in their probabilities it would be 0% for all 3 (will a very small adjustment for the average underround book). Interestingly the advantage of backing Home would have been much higher in the first 4 seasons but this season the bias is going to the away team. I suspect that is because the market has overcorrected for the previous Home bias.
My thinking is that sports trading is the only way to make a consistent long term profit rather than out and out gambling. But then I thought that if I could mirror the bookmakers model and either back or lay at better odds than the true probability of that event occurring this would yield long term profits. I must admit I did confuse a 100% book to true probability which is why I asked about gaining an edge in play when the odds would have moved. Luckily I am only paper trading this but thanks for all the sound advice
User avatar
Derek27
Posts: 23468
Joined: Wed Aug 30, 2017 11:44 am
Location: UK

gstar77 wrote:
Fri Feb 12, 2021 12:03 pm
firlandsfarm wrote:
Fri Feb 12, 2021 11:55 am
gstar77 wrote:
Mon Feb 08, 2021 6:34 pm
As we know the exchange is efficient and nearly all markets are near 100% perfect and show true odds.
Derek27 wrote:
Fri Feb 12, 2021 4:30 am
Toss a coin and offer 4/5 heads, 5/4 tails and that's a 100% overround, but they're not true odds.
gstar, please tell me you are not mixing a 100% book with a 100% market efficiency! Derek is quite right in pointing out that a 100% book does not mean that the market is 100% efficient in it's implied probabilities. If that were the case there would never be value to be had in a market with a 100% book.

I do not have Bf odds for football at kick-off but if I take the maximum Home, Away and Draw odds from the football-data.co.uk spreadsheets for the last 5 Premier League seasons (including this season so far) as being a reasonable approximation to the Bf odds I get an average book of 99.73%, very close to a pure 100% book. The return (ROI) I get if I place a bet on all 3 outcomes in all matches is +6.38%, -6.02% and -0.09% respectively. If the odds were 100% efficient in their probabilities it would be 0% for all 3 (will a very small adjustment for the average underround book). Interestingly the advantage of backing Home would have been much higher in the first 4 seasons but this season the bias is going to the away team. I suspect that is because the market has overcorrected for the previous Home bias.
My thinking is that sports trading is the only way to make a consistent long term profit rather than out and out gambling. But then I thought that if I could mirror the bookmakers model and either back or lay at better odds than the true probability of that event occurring this would yield long term profits. I must admit I did confuse a 100% book to true probability which is why I asked about gaining an edge in play when the odds would have moved. Luckily I am only paper trading this but thanks for all the sound advice
I can assure you that it is possible to win through out and out gambling (more commonly known as betting ;) ) and even more so on the exchange with a 100% book and no bookies overround. You can't mirror the bookie's model on the exchange because with a near 100% overround your lays will always be undercut and therefore inaccessible.

Football is a very efficient market but one of the inefficiencies in-play is the over-reactions when a side's attacking, red-cards, free kicks outside the penalty box, etc.
Atho55
Posts: 637
Joined: Tue Oct 06, 2015 1:37 pm
Location: Home of Triumph Motorcycles

If you are looking for a template to help get your head around your proposal this is what I did.
Football Model.jpg

Off the view have selected the Home Team Odds for the Leics v Liverpool game 3.6. This brings up just 18 matches that started with those odds.


Leics v Liverpool 3.6.jpg
Liverpool odds at 2.25 (highlighted green) Row AJ inline with the odds offered by BF. My data uses Pinnacle odds. Leicester odds a lot less than data suggests. Draw odds from BF better than data.

The odds will change before the off but you can get the gist of what can be done
You do not have the required permissions to view the files attached to this post.
gstar77
Posts: 34
Joined: Fri Dec 04, 2020 10:52 pm

Atho55 wrote:
Fri Feb 12, 2021 3:14 pm
If you are looking for a template to help get your head around your proposal this is what I did.

Football Model.jpg


Off the view have selected the Home Team Odds for the Leics v Liverpool game 3.6. This brings up just 18 matches that started with those odds.



Leics v Liverpool 3.6.jpg

Liverpool odds at 2.25 (highlighted green) Row AJ inline with the odds offered by BF. My data uses Pinnacle odds. Leicester odds a lot less than data suggests. Draw odds from BF better than data.

The odds will change before the off but you can get the gist of what can be done

Thanks for this really appreciate your help
eightbo
Posts: 2154
Joined: Sun May 17, 2015 8:19 pm
Location: Malta / Australia

gstar77 wrote:
Fri Feb 12, 2021 12:03 pm
My thinking is that sports trading is the only way to make a consistent long term profit rather than out and out gambling.
But then I thought that if I could mirror the bookmakers model and either back or lay at better odds than the true probability of that event occurring this would yield long term profits. I must admit I did confuse a 100% book to true probability which is why I asked about gaining an edge in play when the odds would have moved. Luckily I am only paper trading this but thanks for all the sound advice
As I try to improve this yr I find myself moving away from what I used to think of as trading and heading more towards value betting.
"Trading out" is only important to me when the worst outcome of a market would result in more than a certain % hit to my bank (e.g. 1-2%).

Whether you're gambling or trading you still have to place bets at points of value on average if you want to profit — it's just the money mgmt is different.

When trading, both the entry and exit contribute to your goal of "yielding long term profits" so you have to keep that in mind and think ahead in order to find a neutral or better average exit in terms of value with the time you have left.  Sometimes you'll get it wrong and be placing orders at points where it's not value but your objective is just to get it right most of the time, so when the dust settles you come out ahead.

PS I believe it's the case that those SPs are accurate on average but they can deviate both sides in individual cases.
If the real price of something is 2.00 and it played out 100 times with prices ranging from 1.80 - 2.20 when the event started, it wouldn't mean you are wrong to lay the 2.18 if you also intended to back it at 2.20, as if that was your strategy (to lay and back 1st price of the book wherever it is) and you were able to do that, you would come out ahead.  Many of your bets will be worse than the true value unbeknownst to you, but many of them will be better than the true value, too so it all balances out.
User avatar
Morbius
Posts: 492
Joined: Thu Feb 13, 2020 3:38 pm

out of curiosity, does anyone know how many football matches have in play markets on Betfair or do they all have them or is it only selected matches
User avatar
Morbius
Posts: 492
Joined: Thu Feb 13, 2020 3:38 pm

Derek27 wrote:
Fri Feb 12, 2021 2:30 pm
gstar77 wrote:
Fri Feb 12, 2021 12:03 pm
firlandsfarm wrote:
Fri Feb 12, 2021 11:55 am


gstar, please tell me you are not mixing a 100% book with a 100% market efficiency! Derek is quite right in pointing out that a 100% book does not mean that the market is 100% efficient in it's implied probabilities. If that were the case there would never be value to be had in a market with a 100% book.

I do not have Bf odds for football at kick-off but if I take the maximum Home, Away and Draw odds from the football-data.co.uk spreadsheets for the last 5 Premier League seasons (including this season so far) as being a reasonable approximation to the Bf odds I get an average book of 99.73%, very close to a pure 100% book. The return (ROI) I get if I place a bet on all 3 outcomes in all matches is +6.38%, -6.02% and -0.09% respectively. If the odds were 100% efficient in their probabilities it would be 0% for all 3 (will a very small adjustment for the average underround book). Interestingly the advantage of backing Home would have been much higher in the first 4 seasons but this season the bias is going to the away team. I suspect that is because the market has overcorrected for the previous Home bias.
My thinking is that sports trading is the only way to make a consistent long term profit rather than out and out gambling. But then I thought that if I could mirror the bookmakers model and either back or lay at better odds than the true probability of that event occurring this would yield long term profits. I must admit I did confuse a 100% book to true probability which is why I asked about gaining an edge in play when the odds would have moved. Luckily I am only paper trading this but thanks for all the sound advice
I can assure you that it is possible to win through out and out gambling (more commonly known as betting ;) ) and even more so on the exchange with a 100% book and no bookies overround. You can't mirror the bookie's model on the exchange because with a near 100% overround your lays will always be undercut and therefore inaccessible.

Football is a very efficient market but one of the inefficiencies in-play is the over-reactions when a side's attacking, red-cards, free kicks outside the penalty box, etc.

+1

The inefficiency in the football markets are much more apparent in play simply because pre-off quantitative analysis is largely using historical data but the huge achilles heel in this is that there is a huge unknown amount of data that the market cannot factor in. In horseracing compare the RP forecast prices with SP for example.

Their compilers are very good but same problem as big differences result from the public knowing more than them and SP is more accurate than any one individual quant irrespective of how sophisticated their model is.

However odds are still wrong in football pre-off but that takes experience to know that really.
Post Reply

Return to “Football trading”