SMRTILYTICS is a bit of a hobby of mine. For a full explanation of this project, check out the main SMRTILYTICS page. -OG
Week 29
Twenty-nine weeks into the season and I’m forced to ask the question:
Are statistics true predictors of league position?
In terms of points earned, the answer is a resounding “YES!”
But what about the goals conceded and total shot ratio numbers I’ve been carefully collecting to date? This, after all, is an experiment; a hypothesis based on using just two statistics to make overarching statements on league position.
To be honest, there are still 9 weeks of data to be played – and 10 weeks in some cases (Newcastle, Everton, Manchester City and Liverpool) so formulating broad conclusions are premature. To use an analogy I’m fond of, we’re at the 69 minute mark of a 90 minute season; the potential is there still for some changes.
That said, I’m starting to think the numbers I’m tracking are incomplete from a total picture perspective. Leicester are the obvious example of this as they are not top four in any number I track. The trick though is to understand if they are an outlier and I suspect, as I imagine you do, that they are. Over the course of the next 9 Premier League games, I propose that we will see them enter the top four in the combined metric (currently 5th) as their numbers “catch-up” to their league position.
For the first time this season, I’ve predicted them to win the league. My reluctance to date has been a combination of what my numbers were saying and what my experience was telling me – and that is that there is no way Leicester will win the league. That position has changed forcefully as it becomes clearer it is theirs to lose.
As part of this thought process, I’m considering a new measurement which will likely not come into effect until next season (if this experiment persists). The likeliest of measurements from my perspective is a look at form. That thought process is driven by Arsenal’s fortunes this season.
As you likely know, I’ve spotlighted the Gunners in each of my updates and as I looked at individual games, there were signs they were coming out of the wilderness of Weeks 12/13 to 23. I felt certain that based on these indicators, wins against Manchester United and Swansea would be a safe bet. The abruptness of those losses has reverberated across the Arsenal fan base.
The oddity is that based on those losses, and the strength of Spurs’ numbers, I was not optimistic about the North London Derby this past weekend. What happened in that game was both surprising and not. Tottenham played the way I expected them, their pressing game hardly giving the Arsenal players any time to pick their passes – the bulk of the first half was disjointed at best and Spurs dominated at worst.
Click this graphic to see it in full resolution
Yet Arsenal managed a goal in the opening 45 minutes – 2 in 90 from only four shots on target. I’ve levelled a few criticisms against them this year and being clinical is not one of them and yet they were. Their ability to score with 10 men against the best team in the league defensively gives me renewed optimism for the end of the season – the fact it was Alexis is an even better sign. But Arsenal are facing the very real possibility of running out of runway to catch the league leaders, even if they win all of their remaining games.
What of form though, how can it predict these things? Well consistency has been key to both Leicester and Tottenham’s seasons to date and that is indicated by form. I do think if you can capture the past 5 games form guide, weight and number it, you might have a better idea of how a team will perform in the next game – maybe even a points per game stat could be of use. These numbers are a fun hobby, but I want them to be predictive.
All of this focuses on the top of the table but the statistical view of the bottom three has been extremely consistent all season. It’s hard to see a league next year that has Villa, Sunderland and Newcastle in it; Norwich too would be an unwelcome name in the goals conceded column if you’re Stephen Fry or Delia Smith.
One thing for future weeks to look at, is how the statistics from the front half of the season are comparing to the back half in terms of individual match-ups. I’ve already noted some interesting numbers in my weekly data collection but I’d like to save that as a future topic.
Right, on to predictions which look a little different in terms of format this week. Note that Everton, Liverpool, Manchester City and Newcastle have a game at hand.
See you sometime after week 31 – I’m off to drink tequila and talk SMRTILYTICS in Thadland Mexico!
Youknowit!
OG