Away Match Day Thread 12/09/20 L1 - Lincoln City v OUFC

[LowerSouth]

High profile game. Televised. Multiple caneras. Nowhere to hide.

One camera at Lincoln v Oxford. No chance.

Overturned-just goes to show can never tell ?

 

[Shosho]

I'm not so sure, the ball will get played inside where the stats will probably show there's a greater chance of conceding. Also, allowing the oppo to have the ball is important for counter attacking. This is just my feeling though, @Shosho will probably be able to provide a stats based answer if there is one.

This is interesting and something that clubs are slowly becoming aware of. Even just a basic knowledge of xG can guide you play style.

Simple xG principles:
- closer to goal = greater chance of scoring
- central shots = greater chance of scoring
- shots with feet=better than headed attempts

This obviously works both ways when attacking and defending - you look to create those central 6 yard attempts whilst limiting your opponent to wide areas. A chance outside the 18 yard box but central? About 5-10% chance of scoring....central 6 yard box = 40%+ chance of scoring. There is a host of things that can impact this - number of defenders between the ball and the goal, the type of attempt (head/foot), the type of assist etc.

For crosses specifically, the type of cross plays a big role. For example, a cut back, along the floor between the 6 yard box and penalty spot for someone to shoot with their foot will carry a higher probability of scoring than a floated cross for a header into the same area:



A header, from a cross, carries ~5-10% of scoring...scoring from crosses is tough! To this end, relentlessly crossing is not a particularly clever strategy (unless you focus on cutbacks...Man City the perfect example!).

To put this into the context of the match on Saturday - Those two headers we conceded would have ~5% or less chance based on their locations:



To look at the full match, both Oxford and Lincoln shots (Oxford = blue):



Aside from the headers, we limited Lincoln to shots from wide areas and distance (Oxford = 9 shots, Lincoln = 8). Compare that to Oxford's chances, 2 in the six yard box and a further two between the 6 and 18 yard box - the angle on the six yard shots decreases the probability but still carries 20-40% chance (guess-timation!). Those Oxford shots from range are not ideal and where possible would prefer to see through balls etc, but simply going on the above I don't think we should be hugely concerned. We had decent shot volume and (largely!) decent shot quality.

To look at Oxford last season:

Shots taken:




Shots conceded:



Pretty clear in the comparison above, but Oxford created loads from the penalty spot inwards whilst limiting opponents (love seeing a sparse six yard box! Plus, any attempts from that area are wide). Unsurprisingly, limit the opponents to bad chances and create good ones yourself and we will be all good.

To break that down by shots taken by players:



You can extrapolate loads from this! If you are an opponent one look and it tells you most that you need to know...Rob Dickie had lots of chances, most probably headers so tough to convert, but clearly a threat from set pieces...let Rob Hall have the ball and take pot shots...Matty Taylor is a penalty box god...Sykes regularly gets into positions on the corner of the six yard box (as he did, twice on Saturday)...when Brannagan shoots from in the box he is clinical...

Anyway, it's gone off on a tangent, but smart clubs (not just Brentford!) easily have access to the above and it can inform how to play, recruitment etc. Specifically when it comes to crosses and our strategy, allowing a cross in the air is preferable to getting beaten in behind allowing cutbacks...how you defend the cross is up to you...battle in the air or do as Burnley do and pack as many players as possible between the ball and the goal!

I know this probably feels like the evil number wizards are taking over, but this stuff is pretty basic and as such can be used to guide/supplement a host of strategies.

 

[Sarge]

This is interesting and something that clubs are slowly becoming aware of. Even just a basic knowledge of xG can guide you play style.

Simple xG principles:
- closer to goal = greater chance of scoring
- central shots = greater chance of scoring
- shots with feet=better than headed attempts

This obviously works both ways when attacking and defending - you look to create those central 6 yard attempts whilst limiting your opponent to wide areas. A chance outside the 18 yard box but central? About 5-10% chance of scoring....central 6 yard box = 40%+ chance of scoring. There is a host of things that can impact this - number of defenders between the ball and the goal, the type of attempt (head/foot), the type of assist etc.

For crosses specifically, the type of cross plays a big role. For example, a cut back, along the floor between the 6 yard box and penalty spot for someone to shoot with their foot will carry a higher probability of scoring than a floated cross for a header into the same area:

View attachment 4597

A header, from a cross, carries ~5-10% of scoring...scoring from crosses is tough! To this end, relentlessly crossing is not a particularly clever strategy (unless you focus on cutbacks...Man City the perfect example!).

To put this into the context of the match on Saturday - Those two headers we conceded would have ~5% or less chance based on their locations:

View attachment 4598

To look at the full match, both Oxford and Lincoln shots (Oxford = blue):

View attachment 4599

Aside from the headers, we limited Lincoln to shots from wide areas and distance (Oxford = 9 shots, Lincoln = 8). Compare that to Oxford's chances, 2 in the six yard box and a further two between the 6 and 18 yard box - the angle on the six yard shots decreases the probability but still carries 20-40% chance (guess-timation!). Those Oxford shots from range are not ideal and where possible would prefer to see through balls etc, but simply going on the above I don't think we should be hugely concerned. We had decent shot volume and (largely!) decent shot quality.

To look at Oxford last season:

Shots taken:


View attachment 4600

Shots conceded:

View attachment 4601

Pretty clear in the comparison above, but Oxford created loads from the penalty spot inwards whilst limiting opponents (love seeing a sparse six yard box! Plus, any attempts from that area are wide). Unsurprisingly, limit the opponents to bad chances and create good ones yourself and we will be all good.

To break that down by shots taken by players:

View attachment 4602

You can extrapolate loads from this! If you are an opponent one look and it tells you most that you need to know...Rob Dickie had lots of chances, most probably headers so tough to convert, but clearly a threat from set pieces...let Rob Hall have the ball and take pot shots...Matty Taylor is a penalty box god...Sykes regularly gets into positions on the corner of the six yard box (as he did, twice on Saturday)...when Brannagan shoots from in the box he is clinical...

Anyway, it's gone off on a tangent, but smart clubs (not just Brentford!) easily have access to the above and it can inform how to play, recruitment etc. Specifically when it comes to crosses and our strategy, allowing a cross in the air is preferable to getting beaten in behind allowing cutbacks...how you defend the cross is up to you...battle in the air or do as Burnley do and pack as many players as possible between the ball and the goal!

interesting info that ^^ @Shosho ... thanks

 

[ZeroTheHero]

Yes thanks - interesting.

One thing though. I understand that attempts (and especially headers) from crosses are not as likely to score as those from shots. But the question here is 'How many times when a cross is not allowed to come in does that result in a shot on goal from elsewhere?'. Because, yes, if you allowed a cross to come in *instead* of a direct attempt on goal that would (by your stats) obviously be less dangerous. However, many times when a cross is *not* allowed, that move does not result in a direct shot - the player will play it backwards, skew it out for a goal kick, lose the ball over the line for a throw, be tackled and lose possession etc.

So (and I think it would be difficult) you have to know how many times when a cross is not allowed does that result in a direct shot. Does that make sense?

 

[Shosho]

Yes thanks - interesting.

One thing though. I understand that attempts (and especially headers) from crosses are not as likely to score as those from shots. But the question here is 'How many times when a cross is not allowed to come in does that result in a shot on goal from elsewhere?'. Because, yes, if you allowed a cross to come in *instead* of a direct attempt on goal that would (by your stats) obviously be less dangerous. However, many times when a cross is *not* allowed, that move does not result in a direct shot - the player will play it backwards, skew it out for a goal kick, lose the ball over the line for a throw, be tackled and lose possession etc.

So (and I think it would be difficult) you have to know how many times when a cross is not allowed does that result in a direct shot. Does that make sense?

Yeah I know what you mean - you would probably need tracking data to see how many times a cross is 'rejected' - the probability of scoring is obviously only counted when a shot is taken...so you are looking a 5-10% of shots taken, via a header on the penalty spot...this obviously misses all the cross that were blocked, unsuccessful etc.

Would be interesting for the clevers at Liverpool to look into - they've now calculated the probability of scoring in x amount of events from any position on the pitch

 

[Concretebob]

Overturned-just goes to show can never tell ?

I'm very happy to have been proven wrong!