Experimental 3-6-1

[Stato]

For those of you who like a bit of Stat action Experimental 3-6-1 is well worth a glance.

https://experimental361.com/2016/09/15/scatter-graphics-championship-15-sep-2016/

At this early stage he has this comment to make about us:

"Brighton’s underlying performances have been the strongest in the division, so going one step further than last season and securing promotion looks to be within their power. However some below-par finishing looks to be letting them down at the moment."

Fair comment if you consider the stats, but my view of the last couple of home games is less that our finishing is below-par and more that we are attempting too many more difficult chances, from distance and through crowds of defenders.

 

[FatSuperman]

For those of you who like a bit of Stat action Experimental 3-6-1 is well worth a glance.

I LOVE experimental 3-6-1. Surprised we don't see more charts on here (whilst we are performing anyway). Across the season, if you follow it he is saying that team X over or under performed against the average or was lucky / unlucky based on the type of shots they took. Currently it sound as if we've been quite unlucky (we've taken a lot of shots that normally score) or aren't nearly good enough at finishing. Will be very interesting to keep track over the season.



Edit: drunk

 

[Hampster Gull]

 

[severnside gull]

Fascinating but isn't it the sort of stats base that saw us sign O'Grady et al?
Not being dismissive. It's interesting. But I'm nervous of taking it too much into account

 

[GT49er]

Hmmm......nailed on automatic then?

 

[FatSuperman]

None of the above. It doesn't mean anything is or is not going to happen, it's just about the number of quality chances (according to this dude) versus the defence and how well a given l team has done. If you look at the little charts that show chances per game it's quite interesting (for me). As an example, statistically we should have scored 1.8 goals versus Brentford's 1.1. So either we were unlucky and they were lucky (they hit 2 out of 1.2) or we can't finish for shit (given we got zero).

I've the course of the season (assuming it's luck) then we should do ok. Obviously the stats vary wildly based on the latest performance, but for me I just treat it as something to while away the hours until the next game. A bit like my nervousness of going into the work toilets in case that old boy is violently smashing one out in Trap two

 

[LlcoolJ]

Very cool site. Saw it via Twitter last season and it's really interesting.

I’ve had to manually hack this graphic because Derby were ruining it. With a return of just one goal from 98 attempts, the other 23 teams were crammed into an incomprehensible mess at the bottom, so I’ve taken the step of boxing them out separately.

 

[Stato]

Very cool site. Saw it via Twitter last season and it's really interesting.

Derby's situation is mad. They have scored the same number of goals in 7 matches as they scored against us in 77 minutes last May. In these games they have variously started:

Darren Bent career average a goal every 225 minutes
Chris Martin career average a goal every 288 minutes
Matej Vydra career average a goal every 302 minutes
Johnny Russell career average a goal every 306 minutes
Tom Ince career average a goal every 315 minutes
Nick Blackman career average a goal every 387 minutes
James Wilson career average a goal every 468 minutes
Abdoul Camara career average a goal every 738 minutes
(They also have Andreas Weimann (a goal every 504 minutes) coming back from injury)

This season, assuming any two of the strikers and Ince are on the pitch at any one time, they have scored nothing in about 1890 minutes (3 players x 7 matches x 90 minutes) with Derby's only goal coming from defender Craig Forsyth.

 

[Icy Gull]

Derby's situation is mad. They have scored the same number of goals in 7 matches as they scored against us in 77 minutes last May. In these games they have variously started:

Darren Bent career average a goal every 225 minutes
Chris Martin career average a goal every 288 minutes
Matej Vydra career average a goal every 302 minutes
Johnny Russell career average a goal every 306 minutes
Tom Ince career average a goal every 315 minutes
Nick Blackman career average a goal every 387 minutes
James Wilson career average a goal every 468 minutes
Abdoul Camara career average a goal every 738 minutes
(They also have Andreas Weimann (a goal every 504 minutes) coming back from injury)

This season, assuming any two of the strikers and Ince are on the pitch at any one time, they have scored nothing in about 1890 minutes (3 players x 7 matches x 90 minutes) with Derby's only goal coming from defender Craig Forsyth.

Just pleased that we are not playing them again anytime soon because you just know that they would get a glut against us. Ever since I've watched the Albion, every time a stat like this or one like a striker hasn't scored for 43 games crops up then inevitably the record ends when they meet the Albion. It does also work the other way, if a team hasn't lost at home for a high number of games, we are almost certain to win there.

I keep reminding myself that we have never beaten Sheffield Wednesday at Hillsborough and this HAS to come to an end soon....

 

[symyjym]

Those graphs sound look pretty accurate. Busy attack, quiet defense, energetically wasteful, but formidable.

 

[Brighton Mod]

At least 3-6-1 makes sense of our lack of striker signings this summer. But the stats can be deceiving, there is no average or mean, a goal is not scored for every ten or twelve shots and long range attempts or speculative shots pose no threat. We have taken and lost many points on teams scoring from their only attempt.

 

[Stato]

His latest analysis makes for interesting reading:

https://experimental361.com/2016/09/25/e-ratings-update-championship-25-sep-2016/

Let's hope he is right, as it seems we may already be safe from relegation.

 

[larus]

Another interesting point for me is the number of "top" teams we've played so far.

If you split the league into 3 groups of 8 :

We've played 4 of the top 7 (8 including us).
2 of the next 8.
3 from the bottom 8.

So, in theory, we've play on average, harder fixtures already than is to come in the rest of the first half of the season.
You'd hope that if we keep the key players fit, we could have a decent run now.

But I guess we've still not played anyone good yet.

 

[FatSuperman]

At least 3-6-1 makes sense of our lack of striker signings this summer. But the stats can be deceiving, there is no average or mean, a goal is not scored for every ten or twelve shots and long range attempts or speculative shots pose no threat. We have taken and lost many points on teams scoring from their only attempt.

True, not all shots are equal - his model takes this into account, it values shots based on how likely that sort of shot is to go in, you see that penalties are massive jumps on the graphs as much more often than not, they end in a goal. Long range speculative shots aren't worth anywhere near as much, although occasionally they go in. If you look at the match by match graphs you'll see some teams with loads of shots, but they are all terrible (tiny jumps on the timeline).

It's interesting that we have the best offence and defence according to this model, it's not borne out in some of the results of course, but on balance it looks good. A,a zingy Derby are second, partly thanks to a solid defence, but they are also top six offensively. Instinctively that makes sense as you'd have to say that team are massively underperforming, despite that I can't see them turning it around anytime soon.

Everyone should read all the championship articles on that site!

 

[Bozza]

Another interesting point for me is the number of "top" teams we've played so far.

If you split the league into 3 groups of 8 :

We've played 4 of the top 7 (8 including us).
2 of the next 8.
3 from the bottom 8.

So, in theory, we've play on average, harder fixtures already than is to come in the rest of the first half of the season.
You'd hope that if we keep the key players fit, we could have a decent run now.

But I guess we've still not played anyone good yet.

I'd be surprised if all of Huddersfield, Barnsley, Reading, Birmingham and Wolves finished in the top 8.

 

[FatSuperman]

I'd be surprised if all of Huddersfield, Barnsley, Reading, Birmingham and Wolves finished in the top 8.

Yeah I agree. The site posted a difficulty matrix based on leading bookies odds, based on this we've played two tough games (Derby and Newcastle), three easy and four other. Brentford are ranked 13th based on odds.

https://experimental361.files.wordpress.com/2016/06/ch-16-17-frd1.png
Bristol have had the toughest run so far, so they've done pretty well really.

 

[larus]

I'd be surprised if all of Huddersfield, Barnsley, Reading, Birmingham and Wolves finished in the top 8.

Of course the league will change and some teams will have played easier games and maybe had a run of things going their way.

However, of the "bigger" clubs below, how many have got their act together for this year? (I realise that bigger is subjective).

Wednesday.
Leeds.
Fulham.
Ipswich.
Forest.
QPR.
Villa.
Derby.

 

[Peter Grummit]

"Burton have impressed in their debut season at this level so far but, while Brighton’s winner came late on, they can have few complaints about this result. Apart from a short spell midway through the first half the hosts were unable to threaten consistently here, although to their credit it took a penalty to break them down."

"A dominant display from Brighton saw them record a comfortable looking victory, made all the more impressive by their ability to keep Barnsley relatively quiet. The Tykes had taken almost as many shots as their hosts this season but were mainly reduced to speculative efforts here."
Match summaries based purely on the quality and quantity of shots taken by each side.

 

[Postman Pat]

Updated after yesterday's games:
View attachment 78381

View attachment 78382

 

[Stato]

I don't want to criticise too much as I do really like his website, but yesterday's game highlighted the limitations of his method.

Looking at the possession and number of chances he surmised that we scraped past Wednesday, whereas supporters from both sides that were at the game seem to suggest that Brighton looked fairly easily the better side and that Wednesday's chances were mostly not of good quality. I noticed the same about the Brentford game when his method suggested that we should have got something from the game. Being there, I didn't see us particularly troubling Brentford at all.

Its worth noting that his method showed Brighton scoring fairly well when Hyypia was in charge, so may need some adjustment that better takes into account the quality of chances. Working a ball to a free Glenn Murray five yards out needs to be treated very differently to Gary Gardner shooting from distance every time he is passed the ball.

.