Are we better at attacking or defending? Decide before looking [;)] (Warning not for stat sceptics)

[Badger 2,405]

The stats suggest that:1. We are better at attacking than defending - busy attack, busy defence. Should be entertaining?2. 6th highest expected goals for; 10th highest ExG against3. Broadly average defensive effectiveness. I''d be interested to see the breakdown of goals vs - relative %ge from set pieces and crosses.A good site for fellow nerds [:D]https://experimental361.com/2018/09/30/scatter-graphics-championship-29-sep-2018/

[nutty nigel 7,509]

If you separated this to halves we''d be quiet/quiet in the first half of games where we''ve scored 3 and conceded 4. But busy/busy in the second half where we''ve scored 11 and conceded 9.

Last 15 mins scored 6 and conceded 3. That''s "Norwich time" where we have busy attack and quiet defence?

[Crabbycanary3 994]

So, if I''ve added all that up together correctly, I think that makes us 5th.

Can anyone confirm that?

[Bethnal Yellow and Green 1,557]

Norwich achieved half of their season so far’s xG in the first three games. Since then they have averaged an xG or about 1 per game.

Their numbers look distinctly midtable and I have a suspicion that the current run they are on is similar to the one about the same time last year. Unless Norwich can start creating more chances in the box then play offs seems very unlikely.

[king canary 7,578]

Interesing take Bethnal.

It does seem like we''re scoring more from close range though, rather than relying on a worldie to beat teams.

[Badger 2,405]

I take your point BYG and am certainly not getting carried away with current form. However, I''m not quite sure of your methodology - which in essence is taking an even smaller sample from an already smaller sample. It is in the nature of trends that at some times you are above trend and sometimes below it (which is basically your point) but equally that very small samples are unreliable. It is even more difficult to extrapolate forwards!Nevertheless, I don''t think that the interpretation of the data as distinctly mid table is entirely justified. The current data suggests top half, rather than mid table.

[Bethnal Yellow and Green 1,557]

King - Shot location does appear to be better, from about half of all shots being outside the box to about 40%. But shot numbers is down also.

Badger - Norwich xG difference this season so far puts them at 11th in the table, which I would say is both midtable and top half. As you say though it is a little early to be drawing too many conclusions. I singled out the first three games as it felt like Farke was trying something different in those games, from Preston onwards he seems to have gone for something a bit more conservative so I wouldn’t expect Norwich to remain so high in the attacking rankings.

[westcoastcanary 173]

[quote user="Bethnal Yellow and Green"]Norwich achieved half of their season so far’s xG in the first three games. Since then they have averaged an xG or about 1 per game.

Their numbers look distinctly midtable and I have a suspicion that the current run they are on is similar to the one about the same time last year. Unless Norwich can start creating more chances in the box then play offs seems very unlikely.[/quote] Yes, but in those first three games we dropped points that, on xG alone, we should have gained. In the subsequent 7 games in contrast, we have garnered points well in excess of what, again on xG alone, we deserved. Overall we have 5 more points than what xG would suggest our performances merit. Penalty box chance creation and higher xG isn''t the be all and end all. A point is a point however it is come by, and good finishing combined with really solid defending are rewarded with points. Interestingly, we have three players in Experimental 3-6-1''s Championship team for September -- Krul, Lewis and Pukki.

[cornish sam 944]

Just out of interest, has the xg against changed after the Preston game as well? Was it just a case of us commuting more to our attacks so being more vulnerable and now we''re trying to balance it more evenly?

[lake district canary 4,531]

I hate stats.   I know this thread isn''t for me, but I can''t let people get away with saying we are likely not to do well this season because of one small area identified by stats when we all know stats are open to interpretation and are limited to the small area they are identifying.  There is always a bigger picture and stats should always be taken with a pinch of salt. 

[westcoastcanary 173]

@cornish sam

From Experimental 3-6-1''s Timeline figures:

xGa games 1--3 = 1.66

xGa games 4--10 = 1.14

[Bethnal Yellow and Green 1,557]

Westcoast - I agree with most of what you say. Expected goals isn’t an exact science and there will always be teams that outperform and underperform those numbers. The numbers available freely are also very basic versions with less reliability. On general though, teams who take their shots in good locations and take a lot of shots will do well, as long as they aren’t leaking even more shots. A large part of Man City’s succes has been down to Pep coaching his players about how to create the best chances (the low cut back across goal).

Cornish - yup, Preston and on saw Norwich reduce their xG against also. Ipswich are the only team to get more than 1.5 against Norwich, with 1.9. Whereas in all three of the first games Norwich leaked more than 1.5 (1.5, 2 and 1.7).

LDC - stats should only be used in conjunction with watching games, and I’ve seen pretty much every match this season. It has been noticeable that Norwich haven’t created much in front of goal in the recent games (which the 1 goal winning margin in all of them suggests). Hopefully Norwich continue to win games, but they will start to be more creative in the long run to get higher up the table.

[westcoastcanary 173]

@Bethnal

No quarrel at all with that (your reply). What I would say is that Man City''s rather faltering start to this season underlines the point that success is built first and foremost on sound defence. "The right balance" between offence and defence doesn''t mean equal emphasis, let alone heavy weighting towards attack! 

[Bethnal Yellow and Green 1,557]

Not sure I’d say 2 dropped points and a goal difference of plus 18 after 7 games a ‘faltering’ start. Also having only conceded 3 goals in the league I wouldn’t say they really have a problem with defence either.

[westcoastcanary 173]

@Bethnal

[:D] Put like that you wouldn''t. But look at who they''ve played so far in the EPL. Charity Shield apart, they haven''t been tested by home-based opposition. The only real tests have come in the Champions League; well beaten by Lyon and run very close last night by Hoffenheim. Sunday will be interesting.

[king canary 7,578]

Experimental 361 is a great site but that team of the month is based on fantasy football point scoring which makes it pretty meaningless in my opinion.

One thing that is worth noting is we''ve seen teams sustain ''fine margin'' success across a season recently (Huddersfield and Reading) although it is very tough.

@westcoast my reading is that early in the season we fell foul of those fine margins whereas now they are falling in our favour.

Longer term we probably do need to create more chances but it certainly feels like Farke has successfully balanced the need to be solid without strangling the attack much better this season.

[westcoastcanary 173]

[quote user="king canary"]@westcoast my reading is that early in the season we fell foul of those fine margins whereas now they are falling in our favour. [/quote] Disagree with that take on the early season king. If Birmingham hinged on fine margins they fell to both sides; WBA was lost due to rank bad defending, nothing marginal about it apart from the score; so I''d say only Sheffield fits your reading from the first three. Preston, any fine margins fell in our favour, and Leeds was again emphatically lost through poor defending. As to exactly what "fine margins" amount to, I''d say last night''s result was indeed fine margins, but falling both ways.