Value in betting
Hi, iv been reading the forums for a while now. I see alot of information regarding getting value for your bets by working out if the bookie has mispriced it if I use formula a to come to my conclusion and somebody else uses formula b on say over 2.5 goals mine says there's a 40% chance there's says that is 50%, the bookies have priced it at 45% chance. Who's to say which holds more value? Hopefully this isn't a stupid question
In the above scenario there are three possibilities. You're conclusion is most accurate and profits, the other guy's conclusion is more accurate and profits, or the bookie's got it right and you both lose.Sam1988 wrote: ↑Sun Apr 14, 2024 6:02 pmHi, iv been reading the forums for a while now. I see alot of information regarding getting value for your bets by working out if the bookie has mispriced it if I use formula a to come to my conclusion and somebody else uses formula b on say over 2.5 goals mine says there's a 40% chance there's says that is 50%, the bookies have priced it at 45% chance. Who's to say which holds more value? Hopefully this isn't a stupid question
- wearthefoxhat
- Posts: 3235
- Joined: Sun Feb 18, 2018 9:55 am
Value is important when you take other factors into account.
On my excel sheet(s) I'd only consider a runner that indicates value (create a tissue price and compare with market) and then check the overall profile.
In the example below, by using the IF(AND) conditions formula, it automatically highlights any that align.
Not every runner will be value as they are not well rated and suggests the price is inflated.
On my excel sheet(s) I'd only consider a runner that indicates value (create a tissue price and compare with market) and then check the overall profile.
In the example below, by using the IF(AND) conditions formula, it automatically highlights any that align.
Not every runner will be value as they are not well rated and suggests the price is inflated.
You do not have the required permissions to view the files attached to this post.
- wearthefoxhat
- Posts: 3235
- Joined: Sun Feb 18, 2018 9:55 am
Here's how the Kempton 8.30 worked out.
If backing at ISP, 7/1, the winner would not have been in the value range, but the BSP of 9.73 triggered the value signal.
The top two rated looked interesting but had question marks about class/form overall.
If backing at ISP, 7/1, the winner would not have been in the value range, but the BSP of 9.73 triggered the value signal.
The top two rated looked interesting but had question marks about class/form overall.
You do not have the required permissions to view the files attached to this post.
-
- Posts: 7
- Joined: Tue Apr 16, 2024 7:45 am
I don't think this is a stupid question. I have built models for several sports/markets. For some sports I have several models. For some sports my best model is a hybrid model. By hybrid I mean that it is taking the percentages or odds (percentages and odds is really just different ways of showing the same thing) from several models and combine them to get a more accurate price.
Someone else in the thread replied, lots and lots of data points. I think this is a good answer. If you have enough odds and result data so that you have a huge sample. Checking what holds most value/is more accurate is simple math.
The problem is if you do not have a large enough sample size. When you do not have enough data it tends to get very tricky.
Someone else in the thread replied, lots and lots of data points. I think this is a good answer. If you have enough odds and result data so that you have a huge sample. Checking what holds most value/is more accurate is simple math.
The problem is if you do not have a large enough sample size. When you do not have enough data it tends to get very tricky.
Peter had written some blog posts about value. Here is a good one: https://www.betfairtradingblog.com/blog ... ok-like-2/
(That was from 2013, back when the BA blogging platform was nice enough to include the year in the date of the post!)
(That was from 2013, back when the BA blogging platform was nice enough to include the year in the date of the post!)
- wearthefoxhat
- Posts: 3235
- Joined: Sun Feb 18, 2018 9:55 am
As usual I tend to go off tangent.
My horse racing model showed the outcome of comparing my interpretation of value v the exchanges highly efficient market. I posted it to show there needs to be flexibility when intepreting ratings and the need to have other factors to compare them with.
The OP was asking about the O2.5g market and comparing his findings book % against the bookies book % and what you'd constitute/interpret value.
It does depend on the starting point. With Xg stats and normal league stats there are differences. If the OP's model leans more towards one or the other, there may be an edge if looking at the Last 4, 6, 8 games at home or away for each team. Over time, you'll see a trend/pattern that can be exploited.
In conclusion, I'd trust your own findings against the market any time.
My horse racing model showed the outcome of comparing my interpretation of value v the exchanges highly efficient market. I posted it to show there needs to be flexibility when intepreting ratings and the need to have other factors to compare them with.
The OP was asking about the O2.5g market and comparing his findings book % against the bookies book % and what you'd constitute/interpret value.
It does depend on the starting point. With Xg stats and normal league stats there are differences. If the OP's model leans more towards one or the other, there may be an edge if looking at the Last 4, 6, 8 games at home or away for each team. Over time, you'll see a trend/pattern that can be exploited.
In conclusion, I'd trust your own findings against the market any time.
get it right and you officially join the elite club.
full of insights ,always learn something new from the tangents
- wearthefoxhat
- Posts: 3235
- Joined: Sun Feb 18, 2018 9:55 am
Formlines and all other associated variables will require a trained eye to make an educated guess.wearthefoxhat wrote: ↑Mon Apr 15, 2024 7:37 pm
Not every runner will be value as they are not well rated and suggests the price is inflated.
If not careful one could fall into data fitting and gamblers fallacy.
It's great to see you horn in those skills.
The integrity of the data collected could also be a point of concern.E.g the BSP is said to be a bit off.
( it contains some in running data) Not unless you account for that 'error'. Some false positives might creep into the Tissue price.
But of course , experience is always the best teacher and with your hands on approach, it's more likely you've already seen it in your data in one way shape or form .
I personally prefer to turn up and work with what the market is throwing at me , I find myself having very little time to look at the data I've collected, bought or even used .
Time constraints leave me with a narrow choice of cold trading and in-play metrics.
-
- Posts: 1081
- Joined: Fri Nov 20, 2015 9:38 am
Looks like you are identifying significant value on the lay side too?