stats based the live score,what do you think?

Football, Soccer - whatever you call it. It is the beautiful game.
Post Reply
MakeThat$$$
Posts: 23
Joined: Tue May 24, 2022 7:16 am

there re some services that provide stats based on the score at ht or goals scored
i.e.
the score is 0-1 at HT
home team scored 60% in the past on second half when losing 0-1 at HT

what do you think abot this kind of stats?
User avatar
Derek27
Posts: 23477
Joined: Wed Aug 30, 2017 11:44 am
Location: UK

MakeThat$$$ wrote:
Tue Mar 07, 2023 1:27 pm
there re some services that provide stats based on the score at ht or goals scored
i.e.
the score is 0-1 at HT
home team scored 60% in the past on second half when losing 0-1 at HT

what do you think abot this kind of stats?
Can't honestly say that stat would be useful. If the home team was the outsider, is a goal down at HT, had 10% possession and is being annihilated, I wouldn't fancy them to get an equaliser because they score 60% of the time.
User avatar
ShaunWhite
Posts: 9731
Joined: Sat Sep 03, 2016 3:42 am

The issue with footy stats is the tiny sample size making any sort of p-value almost meaningless. Eg 60% might only be 3 out of 5 and but for a single decision in that 450mins it could have been 2 out of 5 and 40%.
User avatar
Euler
Posts: 24701
Joined: Wed Nov 10, 2010 1:39 pm
Location: Bet Angel HQ

MakeThat$$$ wrote:
Tue Mar 07, 2023 1:27 pm
there re some services that provide stats based on the score at ht or goals scored
i.e.
the score is 0-1 at HT
home team scored 60% in the past on second half when losing 0-1 at HT

what do you think abot this kind of stats?
I started with this sort of stats back in the 1980's but abandoned it fairly quickly when I realise it didn't tell you anything.

You need to go much deeper to get useful information. You need to model the market.

What was the start price, when was the goal scored, how many goals were expected to be scored. Was it against the run of play, how many shots did each side get. What was the xG.

It's a bit like saying, in golf. I got an eagle on the last hole, so I'll get an eagle on the next, without realising that it's a par 3.
User avatar
Euler
Posts: 24701
Joined: Wed Nov 10, 2010 1:39 pm
Location: Bet Angel HQ

Somebody kindly posted some scraped data here: -

viewtopic.php?f=54&t=14776&start=165

This will allow you to dig around in these sorts of stats.
MakeThat$$$
Posts: 23
Joined: Tue May 24, 2022 7:16 am

ShaunWhite wrote:
Tue Mar 07, 2023 2:11 pm
The issue with footy stats is the tiny sample size making any sort of p-value almost meaningless. Eg 60% might only be 3 out of 5 and but for a single decision in that 450mins it could have been 2 out of 5 and 40%.
no way to have many samples of this kind
we talk about results of the given season with a certain score
weemac
Posts: 1216
Joined: Mon Sep 16, 2013 8:16 pm

MakeThat$$$ wrote:
Tue Mar 07, 2023 1:27 pm

home team scored 60% in the past on second half when losing 0-1 at HT
Who were the away teams? Do you think it's important? The 60% is meaningless if you don't know.
User avatar
ShaunWhite
Posts: 9731
Joined: Sat Sep 03, 2016 3:42 am

MakeThat$$$ wrote:
Tue Mar 07, 2023 2:35 pm
ShaunWhite wrote:
Tue Mar 07, 2023 2:11 pm
The issue with footy stats is the tiny sample size making any sort of p-value almost meaningless. Eg 60% might only be 3 out of 5 and but for a single decision in that 450mins it could have been 2 out of 5 and 40%.
no way to have many samples of this kind
we talk about results of the given season with a certain score
Exactly. That's why others are talking about combining a variety of factors rather than seeing individual stats as useful. Context is key when you can't have the benefit of a large sample.
Atho55
Posts: 637
Joined: Tue Oct 06, 2015 1:37 pm
Location: Home of Triumph Motorcycles

Spread of full time results from Premiership from 2005 to 2019 when 0-1 at HT

HT 0-1.jpg
You do not have the required permissions to view the files attached to this post.
MakeThat$$$
Posts: 23
Joined: Tue May 24, 2022 7:16 am

ShaunWhite wrote:
Tue Mar 07, 2023 2:52 pm
MakeThat$$$ wrote:
Tue Mar 07, 2023 2:35 pm
ShaunWhite wrote:
Tue Mar 07, 2023 2:11 pm
The issue with footy stats is the tiny sample size making any sort of p-value almost meaningless. Eg 60% might only be 3 out of 5 and but for a single decision in that 450mins it could have been 2 out of 5 and 40%.
no way to have many samples of this kind
we talk about results of the given season with a certain score
Exactly. That's why others are talking about combining a variety of factors rather than seeing individual stats as useful. Context is key when you can't have the benefit of a large sample.
thats why the ht stats is an important factor to consider
right?
Michael5482
Posts: 1218
Joined: Fri Jan 14, 2022 8:11 pm

MakeThat$$$ wrote:
Tue Mar 07, 2023 3:57 pm
ShaunWhite wrote:
Tue Mar 07, 2023 2:52 pm
MakeThat$$$ wrote:
Tue Mar 07, 2023 2:35 pm


no way to have many samples of this kind
we talk about results of the given season with a certain score
Exactly. That's why others are talking about combining a variety of factors rather than seeing individual stats as useful. Context is key when you can't have the benefit of a large sample.
thats why the ht stats is an important factor to consider
right?
Stats are important, it's more important how you use them.

Snapshot of my spreadsheet of home teams total shots for and against, along with shots on target for and against and finally goals for and against. I also do the same for away sides then drill further down between HT and FT. I can then model a particular game with a few calcs then trade of it based on what my model is telling me.

So for example the probability of a low scoring game I'll scalp Overs/Unders and let the market do the work for me, higher probability of goals I'll split my stakes and filter it into backing overs whilst watching the game.
You do not have the required permissions to view the files attached to this post.
MakeThat$$$
Posts: 23
Joined: Tue May 24, 2022 7:16 am

ok
so most people said this is bad approach
so i give it up
Post Reply

Return to “Football trading”