Expected Goals: How xG Works to Improve Football Predictions Massively
Vložit
- čas přidán 2. 07. 2020
- I use maths to beat the bookies! Can you accurately predict goals in football? Check out my guide to Expected Goals to improve your predictions massively during Football premier leagues. Expected goals (xG) is a revolutionary football metric, which allows you to evaluate team and player performance. In this video, I'll explain what is xG in football & how is the statistic calculated?
0:34 What Expected Goals (xG) means.
2:03 How xG ratings work and are calculated.
5:15 How to use Expected Goals ratings.
Check out our trading courses: caanberry.com/trading-courses
Pro tips to win football bets:
3 strategies to win big bets in football: • 3 Football Betting Str...
Football trading guide: • Football Trading Strat...
#FootballPredictions #SportsBetting
------------------------------------------
Follow me on:
Website ‣ caanberry.com
Facebook ‣ / caanberrytrader
Twitter ‣ / caanberrytrader
Instagram ‣ / caan_berry
------------------------------------------
This video demonstrates how you can use expected goals stats to improve your success rate in football prediction. - Sport
Here's a link to the app used in this video: bit.ly/3lo24Rx . The app's name is Infogol.
Nice1 Caan, didn’t hear of infogol before-looks worth a look thanks for the video & tip;)👍
Defo worth a look. Lots of decent stuff on there.
Another interesting video from Caan, loads of useful videos on all aspects of trading.
Thanks, Nige.
realmente no me canso de decirlo, pero realmente me encanta tus vídeos y gracias por compartirlo
Lovely video. Thanks for clarifying
Glad it was helpful!
Glad I started using Infogol after watching last weekends video. Saved me a right few quid on the Charlton v Millwall match this evening. Thanks Caan👍
No problem, not a bad little App is it.
Good vid. An interesting test would be Expected Goals V Lawro"s predictions each week.
Thanks for the great information. I don't trade much but you explain it in such an easy way to understand it gives me the confidence to try.
Cheers, best of luck!!
Your channel is gonna grow fast with this videos
Good tip, I can see how this would work in conjunction with those other videos, cheers Caan
Information is power and all that. It's not efficient enough yet but certainly helps. I think in the future it will have certainly improved.
Hi Caan, perfect timing with this video! I was wondering if you could explain what “Average Total Match Goals” means for both teams?
Nice little app loads of info on it 👌🏻
At first I didn't realise how much but they seem to have done a pretty smart job.
A point you mention in early videos right back to 2017, is stick to the leagues you know. Through time and casting off the bad from the good, I have probably 4 leagues where xG works really well based on the rolling average number of goals for the those leagues. I keep an eye on the average xG both for and against teams. How do they perform against similar side? This was working really well until after the restart, however the games aren't following the same patterns pre lockdown as post. I now am trading a lot of First Half stuff as the incentive to score is poor and sides look less driven.
Leagues where xG was working well
Bundesliga
Premier League
Championship
Major League Soccer
In Spain, Turkey Portugal and Poland there's an incentive to score but there's goals against the run of play. Slow first halves poor odds and penalties for little or nothing. VAR is used more and as a consequence at the end of the first and second halves added time can tot right up
Thanks your for sharing yout technique.
Gluck for weekend trading
Thanks :)
Very interesting Caan. Thanks
Good one bro
Thanks ✌️
Something most people do not think about and a trap many fall in to is if a match has been profiled for 3 goals is that these goals can happen at any point. Just because a game can go 1-1 inside the first 15 min that does not always indicate a goal fest it is just that the games quota of goals was scored earlier than expected. On numerous occasions I have profiled a match for 3 goals or O2.5 and it has been 0-0 at HT but still finished with 3 or more goals. This happened in the most extreme circumstances between Gorodeya v Belshina on Friday which finished 2-1 and all goals were scored in injury time.
Certainly. Human anxiety comes into the picture when a match isn’t performing to the score line as quick as they’d like
Thanks CB for encouraging me to make predictions. Based on your suggested methodology - I am selecting these 4 conditions for overnight sport.
Date: 7JUL20 (Today)
England Premier League
Game: Watford vs Norwich
Condition 1: Under 4 goals for both teams combined
Condition 2: Over 9 corner kicks for both teams combined
Game: Arsenal vs Leicester
Condition 3: Under 4 goals for both teams combined
Condition 4: Over 9 corner kicks for both teams combined
$100 investment will return around $400!!
*18+ Age, comply with local laws, transact responsibly.
How you getting on now?
Am starting to like this channel though. I hope I cashout using this strategy... Please make a video that's related or quite frankly a video on making a football bet from start to finish.
Thanks. Fan from Nigeria
The best stuff isn't going on CZcams I'm afraid.
Nice video raab. Have you got a strategy we could buy off you
There's lots, but nothing for sale. Just helpful CZcamss on the topic of Football.
Good stuff, I'll be looking at infogol and try apply it to my strategies. Caan, do you think Psychoffs course would offer much more from something like using goalprofits? There seems to be a massive difference in pricing, i just wonder would that reflect in what's learned
I haven't been on them so couldn't say.
Good
I'll check out this angle for selecting matches
Cool. Its not a perfect science yet, but defo going that way I recon.
Expected goals is a useful bit of information but you only know the EG of a match after it's finished, personally you can have all the stats and information you want but the eye test is what will get you an edge in football.
It does update inplay (although not all that accurate yet).
Does xG take into account the speed and the accuracy of the shot itself? After all, a shot hit towards the top corner has a higher chance of scoring than a shot hit straight at the keeper.
I was hoping to see more about shot maps in this video what do you think about them? are they useful or is it just a gimmick that looks nice
Will potentially do a follow up about this as it ties in with the most important statistic for football trading, thanks.
Teach us how to bet on live games
Does xG factor in David Luiz dodgey perm?
In terms of a defender getting in the way, you can probably discount Mr Luiz!
How do I add xg and xga together? to my excel spreadsheet?
Like the sound of this. It should help me with my zilch knowledge of football. Lol. Ask me one on medication lol.
The notion of xG is an intriguing prospect but one that leaves me skeptical. I really struggle to be persuaded that football predictions can be accurately made through the use of largely methodical approaches that are based on mathematics and/or science. In my view, there are surely far too many variables with potential to influence the game that prevent reliance on what is essentially a form of probability being effective to a significant degree.
Are you saying you don’t believe in maths and science?
@@CaanBerryProTrader I do, but have doubts about the extent of their their value in this context. After all, football is more art than science and arguably less quantifiable as a result. Each match is a discrete event upon which past situations (similar or otherwise) may or may not have some bearing to a certain degree or other. I feel that this is the case regardless of the strength of historical stats/data. In my view, things like xG seem to be more indicative after a particular contest rather than have merit as a predictive mechanism.
No matter how often somebody explains expected Goals to me i just don't understand it. Is it an in game statistic or post match or what, i just don't understand.
How to go about trading on matches with low xG..but a goal is scored within the first 10 mins?
If that happens then could be some value yes...
How do you calculate xG in a single match?
For example, one team has 7 shots, 3 on target, score 1 goal and has xG 1.45?
Best looking at Opta or Infogol for their explanation, it gets complex.
how did you do it can you share with me , thank you
How'd do what? Expected Goals? They are on most stats sites.
For me, it only shows the probability of the player scoring after the event has occurred. Therefore, how can you use it in game if it calculates the probability after the fact?
Can you provide some insight to the use of it live? For me, as trader, I have investigated the website and struggle to see the use of the website expect to tell me how likely each teams shots were to hit the back of the net after they've happednded. But too me, all care about is if it hits the net or not.
What do you mean the use of live? Sorry I dont understand... are you talking about live video?
I want to start learning football trading. I'm a beginner.
Sure, check out our recent uploads on the channel page :)
Also I would love to know how they come up with their percentages of the chance being scored. Seems like they just grab any old number. Why 43%? Why not 42% or 44? And why does aguero shooting from 12 yards out centre goal get the exact same percentage chance as Jordan Henderson does from the exact same position? Load of bollocks. xG does NOT take player into consideration. Only where the chance was taken from.
Think there's an explainer on their site that answers this question if you have a look.
All stats should be judged according to their context and how they are used. XG is not designed to predict the likelihood of a particular shot leading to a goal. It's generally designed to measure chance quality according to where the shot was taken from and sometimes how many defenders are in the way, where the goal keeper is etc. This measure of chance quality can be summed over the course of several games to give an accurate measurement of the quality of chances a team is making. It also allows you to objectively compare how deadly different players are by comparing their XG to their actual goal output (over the course of say 12 matches) You could add into the model a modifier to account for the quality of goal keeper based on that goalkeepers goals conceded compared to X goals conceded. Nevertheless most statisticians prefer simple models with more biases that are easier to account for post hoc. As with any stat it provides a framework for discussion and should be discussed in the context in which it was measured.
If pundits don't like it, chances are someone's trying to keep it out of the media....
Of course, just like moneyball, there were a lot of dinosaurs that wanted it held back as it was a conflict of interest with their net worth.
Does anyone know if its possible to make good money betting on bet365 ? I lose all the time.Should I stop depositing ?
See the video on our channel. You'll get banned.
I don't see how knowing the 'location of previous goal attempts' 'angle of previous goals' 'distance previous goals where scored from' going to help predict the outcome of a game?
You don’t see how 10,000 shots from the same position, angle, distance are going to give you an accurate measurement of the likelihood of it happening again?
I don’t know how else I can put this for you really… it’s the same reason Uber know where the best pickup/drop-offs are, Google know the best advertising space is, Amazon know what else you are likely to purchase, Facebook know what content you like etc. The list goes on. It’s big data, being refined and measured with a definable metric, based on things that have actually happened, scored against each other. Probability, historic data, maths, all the things the worlds largest companies previously mentioned use to increase efficiency. But you’re suggesting the old way is better? The opinion of one or two old guys?
I didn’t say it’s ‘now’ at a point of maximum efficiency. I just said it’s clearly coming…
I used the app because it’s one of the best I’ve found, no other reason.
This video how to I guess xg goals in football matches
Yup!
Liverpool been incredibly lucky this season, xG shows City won the league.
City havent ever won anything they just bought it all.
That’s because xG is not accurate and you will go bankrupt if you base your bets just on this. This video is highly misleading
Nice
@@24sumo It's really not, its an informational video, as per my responses on all your other comments :)
xG doesn't always give the full picture, that's why Caan mentions incentive to score, and other stats. You put together two good defenses and they cancel each other out, however Liverpool have had the better defence and Liverpool have scored with a lot of set piece low probability shots. City score a lot closer in so a higher xG. This will probably change next season as Laporte is fit again so they with stop the crosses and the organisation will be better.
Anyone who watched the Spain v Sweden game would have expected at least six goals. It ended 0-0.
haha
39 times you said "expected goals".
Impressive.
@@CaanBerryProTrader what has made you start trading in play football? Would have assumed the market conditions are quite different from what you prefer?
Sampsonite
the luggage company?
No Mary
xG is the biggest load of bollocks ever invented. Every football tip on paddypowers tips blog is from infogols XG and the results are absolutely horrific. Football is a random event with 22 human players. No cpu programme has any influence over the result of a match. Results make stats, stats don't make results. At the start of the season, theo Walcott missed an absolute sitter for Everton. Infogol put chance of a 43% chance of being scored. What an absolute load of absolute tripe. I would score it 95 times out of 100. My grandad would score it more than 43% of the time. Ever since I seen that xG has just been an utter laughing stock to me. People consta rly fail to realise that xG is based of last games. What teams did in past games against different opponents has zero relevance to future games against different teams
Caan I'm a fan of yours but I'm calling you out on this one. Please explain how this app helps football punters at all?? Bookies will also know the likelyhood of a team winning (far more accurately than this app) and will reflect this into their odds. If Liverpool have a 70% chance of winning, every single bookie will know this and offer a price shorter than that. The way you described the 'shot map' was very misleading, knowing the location of shots on the pitch have zero bearing on the outcome of a match... You have clearly been paid to promote this app, which is an injustice to your followers. Following this app will lose a punter money 100%. Please reply Caan.
You don’t see how 10,000 shots from the same position, angle, distance are going to give you an accurate measurement of the likelihood of it happening again?
I don’t know how else I can put this for you really… it’s the same reason Uber know where the best pickup/drop-offs are, Google know the best advertising space is, Amazon know what else you are likely to purchase, Facebook know what content you like etc. The list goes on. It’s big data, being refined and measured with a definable metric, based on things that have actually happened, scored against each other. Probability, historic data, maths, all the things the worlds largest companies previously mentioned use to increase efficiency. But you’re suggesting the old way is better? The opinion of one or two old guys?
I didn’t say it’s ‘now’ at a point of maximum efficiency. I just said it’s clearly coming…
I used the app because it’s one of the best I’ve found, no other reason.
Over the last 2 seasons in the top 5 European leagues the team with the highest xG has won 36% of football matches
It ain't predicting fuck all. It's all utter meaningless dog shit
Imagine actually betting based off xG 😂
Expected goals is not extremely accuratel, it’s not even close. You can’t use it to predict goal scoring potential beyond 5%.
It can be used as a tool alongside other metrics, but it’s nowhere near being some magic number
I like your videos mate but you really Don’t have a clue what you’re talking about here.
It will be extremely accurate over time though... more and more data availalbe and its being optimized. Like google search...
Liar