Why are there so many games that end a draw in football ?

[jonnyg]

This is quite a hot topic in the football data analytical world and I was wondering what your views are .

An area that I have looked at for a number of years so rather then bombarding you will loads of data I just wanted to give a summary at this stage re my thoughts and look forward to hearing your thoughts .



The mainstream view which I do not agree with is that the reason why there are so many games that end draw is because of teams " fighting back " so we will see a game > 1-0 / 0-1 > 1-1 and my view is that there is in general a low goal production in football which produces the draw bias .


Now the key is not to take a blanket approach and look at individual leagues and I want to show you ligue 2 which has a high amount of draws season on season and I think I have worked out why that is >

In general an early goal ( an area that I have looked at for a number of years ) will accelerate goal production < but do not take a blanket approach > and if there is acceleration of goal production then there will be lowered expectation of a draw > and it gets even more sophisticated when you look below >

” teams score more goals when they are being led but fewer goals when they are leading,” is a common conclusion via the mainstream football data analysts.

The above conclusion implies that the pathway > 1-1 in a game of football would in general be more likely then 2-0/0-2 and the limitation with reference to the conclusion is the need to look at the effect of the time of the opening goal in an individual league in terms of expectation of the other team fighting back.

There have been 60 games since last season in Ligue 2 where the home teams opened the scoring in the 0-10 minutes time band and 23% ended a draw.

GAME PATHWAY

7/60 1-0 FT

28/60 > 2-0

25/60 > 1-1

In games in Ligue 2 where the home team open the scoring in the 0-10 minutes time band > 1-1 pathway will be as likely as > 2-0 pathway.

There have been 85 games since last season in Ligue 2 where the away teams opened the scoring in the 0-20 minutes time band and 29% ended a draw

GAME PATHWAY

9/85 0-1 FT

24/85 > 0-2

52/85 > 1-1

In games in Ligue 2 where the away team open the scoring in the 0-20 minutes time band > 1-1 pathway will be far more likely then > 0-2 pathway which is why there is draw bias in this profile in ligue 2

HT Summary

If you read the academic research in the area of the draw in a game of football , a common conclusion is that a high % of games end a draw because of teams ” fighting back ” and the actual reason is because in general football will produce a low goal production which will provide some draw bias .

The reason why a high % of games in Ligue 2 end a draw is because of two reasons;

1. low goal production in ligue 2 in general season on season will produce draw bias

2. If you consider that in general that an early goal will result in acceleration of goal production which means that as a result there will be lowered draw bias , this will not apply to ligue 2 in games where there is an early goal metric for the simple reason that 1-0 half time to the home team in Ligue 2 added to the opening goal in the 0-10 minutes time band and 0-1 half time in Ligue 2 added to the opening goal in the 0-20 minutes time band and lowered expectation of the teams + 1 goal winning the second half which is where the draw bias will be strongest and you should be looking to trade around the draw in play in that profile in ligue 2 / backing against the team + 1 to win the 2nd half of the games.

FT Summary

If you have built a model which triggers bets on the draw in games before they start then consider how the time of the opening goal in an individual league will effect expectation of the other team fighting back which should motivate you to trade games in play by reacting to the events that occur during a game .


http://www.socdoc.co.uk/why-are-there-s ... ligue-2-2/



If you have read the information above and are still awake > football data and analysis is still emerging > most of the analysis is around research circa 1998 so along way to go >

[oliver123]

I would not advise anyone to get involved in football trading, and for one simple reason, the whole affair lasts 90 minutes. Now, on the donkeys, the most you have to wait is about seven minutes maximum. Your trade is done and dusted and you get on with the next race.
I know many people lay the draw and wait for a goal in play but again it's a waiting game - sorry but I just don't have the patience. Having said that, a very good post, Cassini.

[jimibt]

I would not advise anyone to get involved in football trading, and for one simple reason, the whole affair lasts 90 minutes.

in essence, i think that's what makes football trading so good. you have a dozen related markets and countless ways to rationally (with the comfort of time) execute strategies with planned exit paths. that said, whether it be horses or football, i'm 95% automation based, so the actual enduring of the market isn't an issue for me. horses for courses (i'll get my coat! )

[megarain]

I wasnt sure where to ask this, but, has anyone looked at Backing the draw, and laying over 2.5 goals, when the prices fit a certain criteria ?

U should get off-setting wins/losses, which, would help comm churn, for PC payers.

Thx