‘Top four attacking unit’ – Stoke City’s real chances of promotion this season


We’re almost a quarter of the way through the season, entering an international break, before the relentless wave of matches continues and Stoke City are handily placed, just outside the playoff positions.

Faith in the current league table runs from the often heard “the table never lies” to the opposite extreme where it is “too early to call”, but the truth, with a little bit of statistical manipulation, lies somewhere between these two standpoints.

It’s hugely important to know where your team comes in the league pecking order. Primarily, to manage expectations.

An overly optimistic projection, based wholly on current league position, may be due to a soft early schedule and a liberal dose of good fortune. A complacent January status quo then may leave a side exposed to a post-New Year relegation fight when more accomplished opponents are faced and good luck recedes to more usual levels.

Similarly, a team languishing at the foot of the table may have readily identifiable reasons for the slump in results and have an otherwise a solid foundation to climb the table. They can hold off from hitting the panic button.

Actual match outcomes define the present league table, but a team’s underlying defensive and attacking process is more influential in deciding the direction a side is going and this can greatly inform any major policy decisions in the upcoming months.

Evaluating a team’s overall worth requires a large body of performance related evidence. Data is drawn not just from the present season, but also earlier matches. But to ensure evolving trends are captured, more recent key performance indicators are given greater weight than those from months ago.

Video Loading

Video Unavailable

Stoke are currently the eighth best team in the Championship based on their ability to created good quality scoring chances and restrict the opposition from doing the same. We’re a top four attacking unit, but just above average on the defensive side of the ball.

Brentford currently have the best combination of key performance indicators, followed by the three relegated former Premier League teams, while Wycombe are only gradually coming to terms with a step up from League One.

Armed with these underlying attacking and defensive numbers we can estimate the chances that any remaining game will end in a win, loss or draw.

For example, when Stoke next take to the field at home to Huddersfield, Infogol’s team ratings for both sides estimate that Stoke will win 48 out of every 100 encounters and 26 will be lost or drawn.

If we wish to simulate a single match outcome based on these implied probabilities, we could randomly draw a number between one and 100.

So 48 or below would represent a Stoke win, 49 to 74 indicates a draw and 75 to 100 a Huddersfield win.

In isolation, this is a mundane exercise, but if we repeat the process for the remaining 11 games on the upcoming weekend, we can assign the points won by the sides from these hypothetical outcomes to their current standings and calculate a feasible Championship table after 12, rather than the current 11 matches.

However, we don’t need to stop there.

We can repeat this not just for one week, but for each of the remaining game days until the final ball is kicked on May 8. Nor need we just run a single iteration of a season, we can run thousands, each time adding accumulated points to the current totals.

This has a great many advantages. It gives a performance based projection for the range of finishing positions, underpinned by an objective assessment of the quality of Stoke and their 23 Championship rivals.

It quantifies how likely we are to finish in the top two, make the playoffs or slip down into a relegation contest, if we maintain our current level of performance.

Just as importantly, we can project the effect of improvements in the squad. By how much would our chances of automatic promotion increase if we managed to improve our defence by say five per cent or would it be a better move to add more attacking power or split resources and strengthen at both ends of the pitch.

An analytically driven team could then look at the team’s underlying numbers to see where squad or tactical improvements are needed to try to turn these projections into a reality.

Here’s where Stoke City are projected to finish using all these probabilistic tools.

As with the Huddersfield example, Infogol has run the simulations and counted how many times out of every 100 trials Stoke finish in the important positions.

It’s testament to how Michael O’Neill has worked with the players available, made shrewd investments and begun to restructure the club towards a sustainable model, underpinned by both youth and experience, that currently 11 seasons out of 100 ends in automatic promotion and just once do we suffer relegation.

There’s much to look forward to and little to fear.

Mark Taylor is football data analyst for InfoGol. You can follow him on Twitter at @MarkTaylor0 or visit the InfoGol website at www.infogol.net. All data from https://www.infogol.net/en





Source link

Leave a Reply

Your email address will not be published. Required fields are marked *