The race to 38, 68, and 87, 29th November edition

This is a new series that I’m going to come back and update every now and then as the season progresses. The premise is as follows; I’m going to track races to a) avoid relegation, b) get a champions league place and c) win the Premiership, based on the historical performance of Premiership teams over the past 12 seasons.

The race to 38 corresponds to the average number of points that the 17th placed team has achieved since the 2000-01 season, where my data begins, with 68 and 87 corresponding to the average number of points scored by the teams finishing 4th and 1st respectively over the same time frame.

Hitting these targets is no guarantee of finishing in a given position (ask Man United fans from last season or West Ham fans about 2002-03), but will provide us with a line in the sand by which we can judge a teams progress during the season.

The method is relatively simple but I’ll give an example using West Ham. With 14 games played they have a total of 19 points. As such, in their remaining 24 games, they need 19 points in the ‘race for 38’ and, 49 points in the ‘race for 68’, and 68 points in the ‘race for 87′. For each team that has played a season in the Premiership I’m then going to look through their schedule and determine whether they had a 24 game span in which they scored a) at least 19 points and b) at least 49 points and c) at least 68 points. The percentage of the 240 teams that do this will be recorded in the % teams column.

Furthermore each team season actually comprises of a total of 15 ’24 game spans’ (where games 1-24 = sample 1, games 2-25 = sample 2 etc.). I’m then going to calculate how many of these spans return at least a) 19, b) 49, and c) 68 points. The percentage of these 3600 samples (240 team seasons multiplied by 15 samples per team season) will then be recorded in the % spans column.

Nb. In the case of Reading and Sunderland, who’ve only played 13 games, the sample size is 3360 (240 team seasons multiplied by 14 samples of 25 games per season

So without further adieu, the inaugural ‘Race to 38’

So, in a finding that will likely surprise nobody, Manchester based fans can be pretty confident that they’ll be watching Premiership football again next season. To take the case of United – even Derby were able to string together a 25 game stretch where they scored more than 5 points.

I think the ‘spans’ column is the interesting one here as it begins to really tease apart the teams at the bottom – and that’s where this exercise is going to prove useful I think. However you look at it, and @SimonGleave had a look at this a couple of days ago (link), QPR are in trouble. Teams score 33 points over 25 games less than half of the time, and that includes the best teams in the league. I've said it before and I will again before long – unless you believe QPR are one of the best 8-10 teams in the league then they're going to need a hell of a lot of luck to get out of this.

One nice thing to note about this exercise – and something I wasn't aiming to do at all to begin with – is that the percentages in the spans column add up to ~1730%. Given 3 teams out of 20 are relegated then a model should be aiming to have this percentage at exactly 1700%. Using the most simplistic model I could think of (i.e., treating the group as homogeneous and not taking team strength into account at all) I'm not far out here.

Next is the 'Race to 68'

This is where things begin to get interesting. The top two are a long way clear of the pack, which is very spread, to the extent that Spurs are only three times more likely to get the 45 points they need than Villa are to get the 55 they must hit. This time around the sum only comes to ~240%, whereas an ideal model would be hitting 400%. This shouldn’t be all that surprising though – average teams pick up about 52 points a season and even United must score at a faster rate than that to hit the target of 68. This 240% will increase as the season progresses but is unlikely to hit 400% unless the 68 turns out to be a low estimate.

The real race for fourth is going to be between Spurs, Everton, and Arsenal and as of this point it’s really tight. I’d throw my money in Arsenal’s corner at the minute but it’s by no means a sure thing.

Finally here’s the ‘Race to 87’

By this measure there’s only really two teams in it and I suspect it will stay this way for the remainder of the season. (I actually think City are good favourites to win the title but at least United’s start is keeping it interesting).

The six bottom teams have already been eliminated from this race – even if they were to win each of their remaining games they would fall short of 87. In the case of the middle ten teams, well whilst it’s still possible for them to reach 87 they’d have to score points at a rate faster than any achieved previously by another club in order to do so. They’re out of it basically.

So there we are – the inaugural races. Is there anything else you’d like to see thrown in here?


One thought on “The race to 38, 68, and 87, 29th November edition

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s