July Mailbag
5 min read

July Mailbag

July Mailbag

It is the middle of the summer, Arsenal are still three weeks away (wow that is a short time) from their first real match of the season and there isn't a new signing to break down yet.

That means it is the perfect time to turn to others to help provide topics for discussion.

Could Zinchenko cover at the 6 or only the left 8?

The easy answer to this question is that I don't know but I don't want to fall back on that easy answer. To actually put myself on the spot, I would say the slightly less slippery answer of I don't think that would be my first choice for how or where he plays.

Given that in the Premier League he played as a full back and with his national team has played more of as an eight I think that it would be hard to project that he could easily slot into a different role.

I think perhaps the more interesting question might be what if he is coming into play as the starting left back and the writing might be on the wall for Kieran Tierney as the undisputed starter. Zinchenko would certainly offer Arteta the ability to field a player that is more inverted fullback/auxillary midfielder than Tierney.

Would you be content with Zinchenko ending our midfield business? I think his skill set could suit a 6/deeper 8 role in the team (while obviously covering Tierney). His defensive attributes being used higher up the pitch appeals to me but I worry about our midfield ceiling not being raised, especially with the shadow currently lingering over he who shall not be named.

To answer the first part, no I don't think that I would be content with that, I still think that the left eight is the area that has the biggest potential for a marginal improvement in the squad now that the center-forward position has been filled. I wouldn't be surprised if Zinchenko was given that chance in that role but I still see it as risk where the team is projecting on a player changing positions and roles.

That being said I can see the allure of projecting his fullback production to the midfield.

Stick with the Zinchenko questions: Looking at our midfield, is Zinchenko enough? Or do we need a goal scoring midfielder like a Paqueta or Tielemans?

I would still like to see a more attacking player brought in but could understand why that might not happen. The potential for Vieira and Smith Rowe to give a more attacking option are still there so it isn't like this is impossible to fill with players already in the club. My personal opinion is that given the age, experience, and value I favor Tielemans for the club still.

How are quality of competition and quality of teammates adjustments done in soccer analytics (if at all)? I imagine a PSG player that only plays against bottom half of the table Ligue 1 sides could be skewed for example. Is the data adjusted at the individual match level?

Generally speaking most stats are left unadjusted. It is a task that is not simple or easy. I have built a simple tool that uses Club Elo ratings to adjust for relative strength differences between teams and leagues and I think puts things in the right ball park but it isn't so good I would present a paper on it at a conference or anything.

One of the methods I find interesting was done by Ben Torney here:

Counting Across Borders - Stats Perform
This article is a written version of Ben Torvaney’s 2018 OptaPro Analytics Forum, which aimed to provide a unified model for the effect of league and age on player output. Read Ben’s initial submission here. Ben’s analyst mentor on this project was LucyRead More ›

Others do adjustments but they are more black boxes that don't have great explanations for what is going on.

How many goals does Arsenal need to score (statswise) to get into the top 4? And do the squad have enough goals in them atm?

The average goals needed to finish in 4th place is 70.3 goals (2008-9 to 2019-20), with the lowest total being 62 and the highest total being 84.

My target for the team is more than 63 non-penalty goals (with maybe 6-8 penalty goals along the way) that would put Arsenal right about at the average needed for 4th place.

Now the big question is do they have enough and would getting to the goals level mean the defense was good enough to turn that into points. On the first part I think yes, when I projected out the goal scoring for the team I think that it is plausible with the current talent to get 9 more goals from the players currently on the roster.

I also do think that the defense should be good enough to where the team approaches a goal difference of around +30 that is needed to be in the conversation for 3rd/4th. The goals and goal difference numbers here are my expectation for the team and I would be disappointed with anything less than that.

Is there a metric that has a strong correlation to winning? I guess after xG difference, what’s the next-best predictor?

This is an interesting question and one that I have not done too much research on. I did last week look at the relationship between wages and places and that is strong and positive. I would speculate that a measure of access to dangerous locations like penalty box touch difference would have a strong correlation to winning as well. This is a place where it would be really interesting to go and pull the different correlations between points and stats.

Statistically, does pineapple belong on pizza? The people need to know.

People are free to put whatever they want on their pizza is my take. I am west coast guy and we have a very anything-goes attitude to pizza where all sorts of different ingredients are put on pizza and I think that they are all great.

My local brewery has a pizza called toes on the nose with ham, pineapple, sweet onions, and bell peppers and I think it pairs excellently with a Hazy IPA.