Tottenham v Arsenal: Performance at Close Game State

An analysis of the North london Clubs ahead of today's big game.

There's a pretty big game today at 4pm UK time. Tottenham host Arsenal in what some believe to be a crucial fixture in the battle for fourth, Tottenham are in great form powered by the impressive form of Bale and Arsenal are coming into this fixture off the back of 3 vital PL wins, albeit against weak opposition.

So let's call this article a preview of sorts, even though it is more about the two clubs' progression throughout the season. Instead of listing basic and all encompassing stats, I want this article to be about something called Close Game State (really needs a better name??).

What is a close game state? Essentially what happenens in terms of shots, SoT, Goals and scoring and save%'s when a team is drawing, leading by a goal (+1) or losing by a goal (-1). Close game state accounts for 80% off all events in a football match, and solely discards the events that happen when one team is being hammered or doing the hammering. We eliminate blow-outs, basically.

Get it? If not, there's tons of information available if you click my profile name on this very page.

Remember we are looking at the two clubs solely at Close Game State to see how they have progressed throughout the season and if one club is superior to the other.

We are using moving averages in all the tables below and not culmulative totals.

Goal Difference at Close GS

Close_gs_nl_derby_medium

This is Tottenham's and Arsenal's goal difference per game at close state seen as a moving average.

Arsenal have had a couple of impressive peaks in terms of CloseGS goal difference but seem to be settling at around + 0.40 goals per game at close game state. It's a good number. But it's not as good as Tottenham's +0.63.

Tottenham, after a low point on game 12 , have steadily improved their CloseGS goals difference per game.

Simply put, if a team is outscoring the opposition by an average of +0.63 goals per game when losing by a goal, drawing or winning by a goal, that team is simply going to win more football matches than a team with a lower number.

Tottenham are better right now in terms of goals at CloseGS, but why?

Shots Differential at Close Game State

Close_gs_nl_shots_medium

Total Shots differential at close game (per game, moving average) shows us some interesting things. Firstly, that Tottenham are pretty good right now, but it took a good deal of improvement mid-season to lift that low number to the impressive place it sits at now. Tottenham outshoot the opposition by 5.52 shots per game, and it is recent form which drives that number.

Arsenal, meanwhile, have seen their shot differential number slide for most of the season from a game 9 peak of 7.11 to a low point on game 22 of just 2.41. It has picked up in the last 3 games but Stoke, Aston Villa and Sunderland will do that for a team.

The question is: How have Arsenal managed to continue picking up points despite their shots differential at the crucial Close Game State having crashed?

Scoring% at Close Game State

Close_gs_nl_scoring_medium

Remember Arsenal's game 9 peak in teams of Shots Differential, and their subsequent slide ever since? That aforementioned slide should of brought about a bigger dip in form, but it didn't for Scoring% came to the rescue!

From game 10 onwards Arsenal's scoring% has improved 7.42% to 10.32% as of today.and that is a 39% increase. Some improvement, indeed. And it was improvement that was vitally needed considering what was happening to their CloseGS shot differential.

Tottenham, on the other hand, were kicking around at 10.5-11% for a spell but have seen that number dip slightly to a shade over 9% as of today.

Save% at Close GS

Close_gs_save_medium

Here we have the save% information, and a rough estimate would hold par save% at 90%.

This graph perfectly illustrates positive and negative regression toward the mean at Close Game State. Arsenal had a super strong Save%, powered by early season clean sheets. That save% was always likely to regress toward the mean given time, and it has.

Tottenham, started slowly in terms of Save% but have been pretty consistent, at ashade under 90%, for most of the season.

Let's add scoring% and save% together now:

pdo

Close_gs_nl_pdo_medium

As stated, pdo is scoring% and save% added together. It is usually seen as a heavily luck driven metric, and anything too far above 100, depending on team quality, can be seen as 'lucky' and likely to regress back towards the mean. Conversely, anything below 100 can be seen as 'unlucky'.

Arsenal sit at 100.3. Tottenham at 98.05. Tottenham may have been the (very) slighly unluckier of the two teams, but both pdo's are so close to the mean that it may be fair to say that both teams' performances and results this season have been neither lucky, nor unlucky and instead are a 'true' reflection of team talent.

Arsenal have 47 points and Tottenham have 51 points.

Tottenham have the better goal difference at CloseGS as well as the better shot differential.

Arsenal have the better scoring% and Save% numbers at CloseGS.

Tottenham are the better team in terms of skill driven metrics, and Arsenal are the better team, albeit very slightly, at partially luck driven metrics.

Better to be good than lucky, I say.

X
Log In Sign Up

forgot?
Log In Sign Up

Forgot password?

We'll email you a reset link.

If you signed up using a 3rd party account like Facebook or Twitter, please login with it instead.

Forgot password?

Try another email?

Almost done,

Join Bitter and Blue

You must be a member of Bitter and Blue to participate.

We have our own Community Guidelines at Bitter and Blue. You should read them.

Join Bitter and Blue

You must be a member of Bitter and Blue to participate.

We have our own Community Guidelines at Bitter and Blue. You should read them.

Spinner.vc97ec6e

Authenticating

Great!

Choose an available username to complete sign up.

In order to provide our users with a better overall experience, we ask for more information from Facebook when using it to login so that we can learn more about our audience and provide you with the best possible experience. We do not store specific user data and the sharing of it is not required to login with Facebook.

tracking_pixel_9353_tracker