xG Uncovered

Wield The Power Of Expected Goals To

Elevate Your Football Trading

Football Trading University

Table Of Contents
Introduction

What Is xG?

How Is xG Calculated (simpli ed)

Is xG The Holy Grail For Football Traders?

All xG Models Are NOT Created Equal

How Is XG Calculated - Advanced

xG and Penalties

Where Can You Find Reliable xG Data?

xG Key Numbers & Common Pitfalls

When To Use xG

xG By The Numbers

How To Assess Strength and Supremacy

Predicting Goals and Match Results

Predicting Second Half Goals

Predicting Late Goals

Final Word

The Football Trading University

xG Uncovered

Introduction
If Football had the equivalent of the Oxford Dictionary. word
of the year contenders would be “VAR” and “xG”.

(Since starting to write this guide, a new phrase has entered

the lexicon "PSR" – (Profit and Sustainability) but the less
said about that the better)

xG, nor VAR, are are particularly new terms but they have
certainly become hot topics as the that last 12 months.

We’ll leave the topic of VAR for another day because the
reason you are here is that you want to know how to harness
xG in the right way, to improve your football trading.

If you really must know, the Oxford Dictionary word for 2023
was “Rizz”, which apparently, according to the kids, is a
shortened version of “Charisma”…

(Which is kind of funny because if anyone tried to use “Rizz”

with me, my first thought after (k**b-head) would be that
they are seriously lacking in said Charisma...)

Any-who, let’s get on to the less controversial topics and

get into the weeds of xG or “Expected Goals”

xGUncovered

Page 3
xG Uncovered

What actually is xG? As you would expect, a shot

from the six yard box with
It may come as a surprise to an open goal will have a
learn that xG has, in fact, far higher probability than
been around for a number of a long range shot from
years now. outside the box.

Yet it has only really However, in most instances

started coming into the it can be far more nuanced
public consciousness in the than that.
last couple of years.
How is xG Calculated
In short, xG stands for – Simplified
“Expected Goals” and it is a
measure of the quality of any
given shot on goal.

To put it another way, what

is the probability of a given
shot resulting in a goal.

An xG model uses historical

information from thousands
of shots with similar
characteristics to estimate
the likelihood of a goal
being scored.

Every short will be given a

rating between 0 and 1.

xGUncovered

Page 4
xG Uncovered

For example, if a shot was

rated as 0.10 then it would
be expected, on average, that
this shot, would result in
one goal for every ten shots
of the same type.

As football traders, this

insight could prove
incredibly useful.

We are all familiar with

using the Shots on Target
metric as one of our sign
However, there are several
posts for a potential goal.
factors that you need to
understand regarding how xG is
The more SOTs there have
calculated.
been, the more we expect a
goal to come soon.
Included within this are the
important pitfalls that you need
So….
to understand to avoid applying
xG incorrectly into your football
Is xG the Holy Grail trading analysis
for football traders?
All of which we will highlight
In short, the answer is no.
and explain in more detail
throughout this guide.
Is it useful? Yes,
absolutely!
In addition to all of this, we
will also show you the most
It is very beneficial and can
powerful ways to use xG including
provide helpful insights and
several strategies you can build
positive indicators to build
into your football trading….
into our match analysis when
starting today.
considering a trade.

xGUncovered

Page 5
xG Uncovered

All xG Models Are NOT TIP:

Created Equal Where possible, we’d advise
to try and use the same
The first thing that you need source for xG data to deliver
to understand is that no xG as much consistency in your
model is exactly the same. analysis as you can.

All xG models have their own How is xG Calculated

characteristics and (Expanded) and Why Do
definitions. Models Differ?
This is why you can see
different xG scores when All that said, the main
looking at the same game on characteristic will be fairly
say Flashscores vs SofaScore similar, taking into account
vs FotMob, vs A.N.Other metrics such as:
Distance to goal
Angle to goal
Body part used for the shot
Type of assist or action
leading to the shot (e.g.
throughball, cross, set piece
etc)

Where things start to diverge

in models and ratings will be
whether the xG source factors
in additional metrics such as
Goalkeeper position
Position of other attackers
Positioning of defenders
Shot height and power

In real terms, this means

that one assessment may score
a shot at 0.30, whereas
another rates it at 0.60.

xGUncovered

Page 6
xG Uncovered

How do xG models treat penalties?

As the characteristic of a penalty is virtually the same in
every match, they are all assigned static value of 0.76
(although we round that up to 0.80 for simple mental maths).

If a goal is 1.00, then it is in effect saying that there is

an 80% chance it will be scored, which marries up with the
historical conversion rate of penalties.

From a football traders perspective then penalty itself is

not what is important here.

What is important for us traders is what impact it has on our

assessments of a game.

It is 0-0 at half time and you see the above xG figures, what
would your first assumption be?

xGUncovered

Page 7
xG Uncovered

It would be a pretty reasonable assumption to think that the

away team was dominating and despite the 0-0 score-line, we
should have seen at least 1-2 goals in the first half.

You may now be preparing to enter a trade on goals and/or the

away side for the second half.

What if I told you that there was a missed penalty by the

away side on the stroke of half time.

Suddenly those figures look a lot less appealing!

xG Uncovered

Discounting the penalty, the actual

xG for the first half is a combined
0.34. With the home side, having
the slightly better opportunity.

In fact, there were no shots on

target for either side in what was,
actually, a very dour game thus
far.

Without that knowledge regarding

penalties, you may well have read
this match entirely wrong.

As it happens, Nottingham Forest

punished Wilf Zaha’s missed penalty
with a goal shortly after half
time, with the final score being 1-
0.

Palace’s xG, minus the penalty,

only rose to 0.43 for the whole
xG Uncovered

Where Can You Find

InPlay xG Data?
The four main sources are as
follows and we expect most, if
not all, to be familiar to you:

SofaScore – good level of detail

Flashscore – wider volume of

matches with xG data

Infogol – one of the originals

FotMob – A favourite with

many, although personally I hate
the UX but it can plug gaps for
those not covered by SofaScore
and Flashscore

Bookmakers such Bet365 often also

provide inplay stats and xG data.

Tip:
We advise having a hierarchy of
sources and stick to one or two
as your primary / priorities
resource.

For example, we use the following

in this order:

SofaScore
Flashscore
xG Uncovered

Back up - FotMob is one I

double check if not showing
in either of the above.

Other Important
Factors and Pitfalls
Concerning xG
We’ve mentioned a couple of
these pitfalls already in the
guide which are:

1. Calculations
2. Penalties

There are a few more flaws

that you need to understand
and be aware of if you plan
to incorporate xG into your Big Chance Missed
analysis and football
trading. Another big one (pun
intended) is Big Chance
Missed (BCM), which can
xG Is NOT the Holy Grail. paint a false picture of a
match, particularly, if it
xG, as many other statistics is for an underdog
deliver, is a trend and trend
only. Consider this scenario.

It is not a definitive It has been a tight game

solution. between a big favourite and
perceived much weaker side.
Please do not fall into the
trap of solely relying on and
basing your decisions on xG
alone.

xGUncovered

Page 11
xG Uncovered

Chances have been few but the You may fall into the trap of
underdog, through a mistake thinking that an upset is on
by a defender were presented the cards and underdog is the
with a wonderful chance to value play.
score and fluffed their
lines. But in reality, just like a
penalty miss, that one chance
E.g. heading over from 6 has skewed the picture of the
yards out. game.

You check the match at half Afterall, there is a reason

time and the xG stats look this team is near the bottom
something like this of the table and may not
create another chance in the
game.

In the live example, the game

was Crystal Palace versus
Liverpool and you can see
from the below graphic, that
Palace created only one
chance in the whole half to
contribute to that xG figure.

It is always worth digging a

little deeper into the stats,
particularly if you do not
have access to live pictures,
to try and complete the fully
story of the game so far,
which you cannot get from xG
alone.

xGUncovered

Page 12
xG Uncovered

Disallowed goals and VAR

A disallowed goal or one ruled out for VAR is NOT included in
the xG figures.

I’m sure there are arguments for this on both sides but, for
me, it absolutely should be included in the stats.

What is our base objective with xG?

It is to determine, by quality and probability, how likely a

goal is.

By that logic, if a team has been able to work the play to

the point that they have got the ball in the back of the
next, only for it to be ruled out due to an elbow being
offside.

xGUncovered

Page 13
xG Uncovered

Does that make the probability of a goal any less?

No, it indicates exactly what we want it to indicate, that

chances are being created, of sufficient quality, that the
probability of a goal is high.
xG Uncovered

Conversely, one thing you can do

when assessing a game is to take a
look back at the first half
summary.

This is not fool proof, but most

VAR ruled out goals will show as a
comment.

Depending on where you sit on this

debate; you could decide to add
this to the xG score yourself to
paint, what we feel, is a more
accurate picture of a game. Or
simply choose to ignore it.
xG Uncovered

Missed Shot (A.K.A. Non-

Connected)
A missed shot, or in xG parlance,
a “non-connected” opportunity
would be a chance where the
attacking player did not connect
with the ball.

Think of a striker having a

stinker, the ball goes across the
box, only faintest of touches and
it is a certain goal!!

Yet somehow he contrives to ‘air-

shot’ it and the ball runs free,
and the chance is gone.

This, again, will NOT be

represented in the xG stats.

If you’re of a similar age to me,

the first example that perhaps
sprang to mine was poor old Gazza
and that Euro 96 semi-final vs
Germany.

It is deep into extra time.

Golden goals are in effect for

the first time – score and
immediately win
xG Uncovered

Shearer is in the box and

driven close to the by-line

He whizzes the ball across

the 6 yard box, evading both
the German GK and Defender

There is Gazza rushing in,

just the faintest of touches
will do it.

The goal is at his mercy.

Legendry status awaits…

… yet it is ever slightly too

fast for Gazza and he misses
It by mere millimetres…

Ultimately, there Is not a

whole lot that we can do
about this. We just have to
accept it as one of the
limitations of xG in its
current form.

But by simply knowing that

this is a factor and its
importance, you will
already be ahead of most
What do you think? Should a football traders.
chance like this be included
in the xG stats?

xGUncovered

Page 17
xG Uncovered

This can be particularly When to use xG

advantageous when you are
watching the game with live Naturally with this guide our
pictures; you could even be focus is on live xG data and
making your mental notes we are using this, in
about the “true” xG of the conjunction with our other
game. match metrics to identify
strong potential trades.

Particularly when we are

unable to access live
pictures of the game.

Pre-match xG data is a
completely different kettle
of fish and would need to be
tackled in its own guide.

The key to xG is waiting for

a pattern to develop.

The first 20-25 minutes is

the minimum and ideally we
would be waiting until at
least half time.

The exception of this would

be when using xG to determine
potential first half trades.

xGUncovered

Page 18
xG Uncovered

Even half time is not necessarily optimal due to something

that we call the “bait and switch” in our Football Trading
University course.

In short, one half of football can look very different to the

second.

For example, a goalless first half with plenty of chances,

can very quickly turn into a dour second half, or vice versa.

However, all this is good and well but the later that you
leave things in a match the higher the chance that a goal(s)
will be scored, meaning less opportunities for us as football
traders.

So the art of using xG effectively is being able to spot the

right trend and have the confidence and foresight to act upon
it BEFORE the xG reaches the level of a goal being scored.

xGUncovered

Page 19
xG Uncovered

We’ll cover some of the ways that you can do this in the next
section when we start to understand the key numbers when
assessing xG

xG by the numbers
1xG DOES NOT equate to 1 goal
Perhaps one of the biggest misconception about xG is that
1.00 xG will equate to one goal.

(1 xG when combining both teams current InPlay xG)

With this in mind, most would assume that the below match met
its xG expectations.

1.58 + 0.40 = 1.98 or 2 goals, with the game finishing 1-1.

However, based on our own analysis and confirming this

against several other resources, the average xG for 1 goal is
around 0.65-0.70 xG
xG Uncovered

So if you saw a combined xG in a

game of 1.30-.1.40, you would
expect, on average, for there to
have been 2 goals in said game.

Using the above example, the game

above should of had 2.82 goals – as
an average.

Lets look at another example:

Again, in the Championship, Ipswich

had an xG of 2.80 and Bristol City
had an xG of 0.97.

This would give us a combined xG of

3.77. Divide that number by 0.7 and
we have an expectation of 4.20
goals.
xG Uncovered

Do Not Underestimate The

Value Of A Goal Scoring
Chance
As mentioned earlier in this
guide, a goal scoring chance is
rated between 0.10 and 1.00, with
1.00 being an actual goal scored
and 0.10 meaning the chance,
based on the factors incorporated
into the calculation, has a 10%
or 1 in 10 chance of being a
goal.

Now, our natural behaviour when

it comes to assessing
probabilities is to lean towards
a high probability number.

The misconception here is that it

would need to be a 70% or 80%
chance for it to be classified as
a “significant chance”.

This is not correct.

A significant chance (i.e. a good

goalscoring opportunity) can be
anything from 0.30 or 30% and
upwards.
xG Uncovered

If you see a teams xG move

up by 0.30 or more after a
shot, then you can realistic
count that as the team in
question creating (and not
taking) a very good
goalscoring opportunity.

Two chances like that would

get us up and above our
expected goal target of
0.65-0.70 very quickly.

30% chance multiplied by a

30% chance, actually gives
us a 69% (0.69) chance of
one of those two shots being
scored.

If we expect a goal to be
scored, on average, every
0.65-0.70 xG then monitoring We are conditioned to
chances of 0.3 and higher in thinking that 4 shots on
a game should be our focus. target indicating a goal
is coming. It is by no
Remember, the purpose of means foolproof, but it is
using xG is to predict or a good metric to at least
“anticipate’ when a goal start assessing a game.
will be scored.
So we’ve seen 4 SOTs and
If we have seen two chances we look up at the xG
of 0.30 or better but the figure and we can see that
score remains 0-0, then we those 4 chances were rated
can realistic start to feel 0.20, 0.35. 0.1, and 0.30.
or anticipate that a goal is
coming. That gives us 0.95 xG, so
Let’s put this into a in real terms we’d be
scenario that we are all expecting 1 to 1.5 goals
perhaps more familiar with. to have already been
scored at this point.
xGUncovered

Page 23
xG Uncovered

Important - what this is It could be the final trend

really telling us is that to tick off to enable you to
there is a high amount of enter a trade confidently
chances being created AND
those chances are, How to assess the
generally, of high quality. strength or
superiority of two
This is the power and teams based on xG
objective of xG. It is
allowing us to read beyond Aside from targeting goals,
shots, attacks and the next easiest way to get
possession and determine if comfortable with xG and to
a team(s) Is creating real start finding positive
goal scoring opportunities. trading opportunities is to
focus on the favourite in
matches.

Our objective here is to be

able to do a quick
assessment of a game at half
time and determine if it is
going the way it was
expected from the pre-match
odds or not.

The reason we want to be

focusing on the favourite is
because it is the easier
team to assess, particularly
when first learning to
harness xG.

It is realistic to expect
them to be more able to
maintain their performance
level more consistently than
a non-favourite or underdog.

xGUncovered

Page 24
xG Uncovered

To benchmark this, we would want to see a reasonably strong

favourite (even money or below at kick off) and for them to
be level at the break, yet ahead on the xG by at least
0.60.

We can determine if a team is ahead on the xG by simply

taking the favourites xG and subtracting the xG of the
underdog.

This will leave you with the xG difference or supremacy of

the favourite.

In the below example, we can see that it was 0-0 at Half

Time, with Man City having an xG of 1.59 and Brentford
registering 0.33.

If we then go ahead and subtract 0.33 from 1.59, we can see

that Man City have a supremacy xG of 1.26 (or approx 2
goals).

xGUncovered

Page 25
xG Uncovered

Man City did indeed go on to win the game, registering an

xG of 2.45.

Brentford only added 0.03 in the second half, so City had a

supremacy of 2.10, despite the single goal victory.

It could and should have been much more.

xG Uncovered

We know from earlier in this guide

that we can expect, on average, a
goal for every 0.65-0.70 xG score.

So by selecting 0.60 we are

indicating that the favourite is
performing to expectation and
should be, or close to being, 1
goal ahead.

They have just not been able to

put it in the back of the
net…..yet.

Another good example would be the

recent Liverpool / Luton game.

In a bit of a shock, luton had

gone into the break ahead by one
goal to nil.

However, if we breakdown the half

time xG stats, Liverpool should
have scored twice and been a goal
ahead by this point.

1.25 – 0.65 = 0.60

Liverpool went on to win the game

4-1 with a xG supremacy of 2.4,
roughly 3.4 goals which fits
perfectly within our expected
range.

By focusing on the favourite, we

can see that the trend of the game
has been set, the team is
performing to pre-match
expectations and it should, in
most case, be simply a matter of
time until the favourite finds the
b kth h
xG Uncovered

In our live testing, the

favourite would go on to win the
match 65-70% of the time.

This gives implied odds of 1.43-

1.54

In other words, if you are

seeing a favourite perform to
expectation; are sufficiently
ahead on the xG, then you should
be achieving value by siding
with this team in the second
half.

A note on underdogs

xG can certainly be used to

identify over-performing
underdogs and throw up the
potential for bigger odds bets
and trades.

That said they are far less

frequent and we do not recommend
that the majority of traders
reading this guide focus on
them.

It is better to hone your craft

and wait until you are confident
that your experience and match
reading levels have reach a
sufficient level:
xG Uncovered

Only consider it if you are an experienced trader, looking to add

additional trading opportunities each day.
Only consider it if you are comfortable with long losing runs – your
win rate could be as low as 15-20%
Scenarios where the underdog is winning the xG supremacy are going to
be rare

It takes greater match reading skills and assumptions as you’ll often

be looking for an xG score where both sides have racked up similar
scores, rather than one dominating
There is benefit, however, in being mindful of what a favourable
underdog may look like, as this can help in identifying games to stay
away from

Predicting Additional Goals / Second Half

Goals
There are two ways in which we can approach a second half
when looking to target additional goals.

Remember, as mentioned earlier in the guide, the longer you

want the stronger the trend that you will be able to
follow.

However, counter that, the longer you wait the more chance
of you missing the next goal(s).

It really is a case of balancing you risk / reward position

and using the data we have available to find the optimal
level.

Goal Deficit
This is a simply a case of taking the total xG for both
sides at any point in a game, allowing you to make an
assessment to compare how many goals have been scored
against how many “should” have been scored.

xGUncovered

Page 29
xG Uncovered

The match ended up with

Take this recent example four goals and an xG of
from The Championship. 2.96, in other words, we
expected 4.5 goals
(2.96/0.65).

Three of those goals came

after the 80th minute,
which highlights another
valuable way to utilise
these insights

Targeting Late Goals

Using xG goal deficits can
be particularly useful
when looking for late
goals - which is a highly
lucrative trading period,
At half time, the combined
provided matches are
xG is 0.42 + 0.57 = 0.99.
chosen with care.
As there have been no goals,
In the following match,
that gives us an xG deficit
Sheffield Wednesday scored
of 0.99.
a late equaliser against
high-flying Leicester.
However, we now know that a
goal is scored, on average,
with 0.65 xG.
Could we really
If we now divide 0.99 by have predicted this
0.65, the true xG deficit goal?
becomes 1.50 goals.

This now looks like a game

in which we would be happy
to target 1-2 second half
goals.

xGUncovered

Page 30
xG Uncovered

xGUncovered

Page 31
xG Uncovered

The first half xG looked like this

It was 0-1 to Leicester at the time, with the match

containing an overall xG of 1.11. This would give us an
expected goal total of 1.7.

If we remove the goal scored (0.65), this leaves us with a

goal deficit of 0.46.

In other words, we were edging towards the game having two

goals at this point with the deficit sitting with Sheffield
Wednesday.

xGUncovered

Page 32
xG Uncovered

In the second half, we should have seen at least one goal for
each side with a SH xG of 1.5

Crucially, the overall figures had Sheffield Wednesday with a

significant goal deficit when looking at the combined
figures.

The match had a total xG of 2.62, giving us a total expected

goals of 4.

Sheffield Wednesday alone, "should" have scored 2.3 goals by

the end of this match.

xGUncovered

Page 33
xG Uncovered

If we then combine this knowledge with the momentum of the

match, a late goal, particularly for Sheffield Wednesday,
begins to look very, very likely, which is exactly what
happened.

xGUncovered

Page 34
xG Uncovered

In summary, nobody can predict late goals (or goals at any

time in a match), but by using smart data and insights, we
can increase our chances of being correct significantly.

Second Half xG
The other approach is to disregard what happened in the first
half and solely rely on the second half xG data.

To do this, lets recap some of the fundamentals that we now

know.

xGUncovered

Page 35
xG Uncovered

A goal should be scored, on average, around 0.65-0.70 xG

A “good” chance can be anything from 0.30 and upwards
We need to check if a penalty has been awarded and allow for this in
our calculations
We want to wait as long as possible, but wait too long and we miss the
opportunity. So waiting for the xG to hit 0.70+ is not optimal

If we put all this together, it suggests that we want to be

looking at around 0.35 and 0.40 xG as an optimal entry point
for another goal, particularly if we can see the larger
proportion of that going towards one team.

Seeing this, for example, in the first 10-15 minutes of the

second half would suggest the game is fairly open, chances
are being created and at least one of those would be
classified as a good scoring opportunity.

Additional checks that you may want to conduct, before

entering a trade, is to check the match momentum and
potentially add in the first half xG stats to see if the
broader pattern aligns with what you are seeing.

If everything checks out, you have a potentially a very good

entry point to target a goal or two in the second half.

xGUncovered

Page 36
 xG Uncovered

Final Word
The development of xG provides a massive advantage to football traders
when it is fully understood and used at the right moments.

It is not, however, a silver bullet or the holy grail of football

trading.

Sadly, no such thing exists.

However, if we had to choose a world with or without xG, we definitely

choose one with xG available.

Most average football traders will not be considering xG

Those that doe include it, a large proportion will likely be applying
it incorrectly.

If you absorb and apply the lessons taught today, you will already be
in the upper right percentile of football traders.

xGUncovered

Page 37
 xG Uncovered

LEGAL NOTICE Disclaimer & Terms of Use

This e-book has been put together for educational purposes only; no
guarantees can be provided around profits or losses made from following
this guide or any other material provided by Predictology or 28
Ventures LLC.
The author and 28 Ventures LLC shall in no event be held liable to any
party for any direct, indirect, punitive, incidental or other
consequential damages arising
directly or indirectly from any use of this material which is provided
“as is” without warranties.
It is your responsibility to comply with gaming laws wherever you
reside. In regions where gambling is illegal, this e-book must be used
for informational
purposes only. Any links in this e-book which lead to a third-party
website must only be used when the local laws allow.
You are not permitted to distribute this e-book and it may not be
altered in any way. It may not be shared or sold.
Further information can be found by reading our:
Terms & Conditions | Privacy Policy | Income Disclaimer
Lastly, if you think you may have a problem with gambling, we encourage
you to seek help and support from the many organisations available to
you. These
http://www.gamcare.org.uk/
http://www.gambleaware.co.uk/
http://www.gamblingtherapy.org/
http://www.gamblersanonymous.org.uk/

xGUncovered

Page 38
xG UNCOVERED
We hope you have enjoyed this guide to xG (expected goals)
and it helps to enhance your football trading skills.

If you would like to learn more professional skills to

elevate your Football Trading, please do consider checking
out the Football Trading University.