Liverpool FC Forum

Really interesting piece.

It’s all in the numbers: why Liverpool’s great start to the season makes perfect sense

Daniel Finkelstein Fink Tank
Last updated at 12:01AM, September 21 2013

Liverpool have made an impressive start to the season, no doubt about it. And I am not that surprised. I expected an excellent start to the season.

The first reason for this is that the fixture list was pretty kind to them. They have scored one more point than we expected them to at this stage.

The second reason for my lack of surprise is more encouraging for them. The Premier League title is much more open this season. There are now five clubs with a 10 per cent or more chance of becoming champions. Manchester United (25 per cent), Chelsea (21 per cent), Manchester City (20 per cent), Arsenal (and note this because it’s a properly big chance, 15 per cent) and Tottenham Hotspur (10 per cent).

Liverpool are only just behind this with an 8 per cent chance of winning the title. They have a 49 per cent chance of securing a Champions League place. So, of course they have done well so far. They are a good team.

There is one more reason for my lack of surprise. They did well in the transfer market. The most important reason for this is that they did not sell Luis Suárez. According to the Fink Tank Dec Tech Player rankings, the Uruguay striker is the third best performing player in Europe and the best player in the Premier League. Trouble he might be, but be is very, very talented.
Partly as a result of his contribution, Liverpool are stronger in attack than defence, so it is impressive that they have added to their team one of the best defenders in Europe. Mamadou Sakho is one of the best 60 players in European leagues according to our rankings. They have also bought, in Simon Mignolet, a better goalkeeper.

Looking back over the past four managers, the picture is pretty clear. Rafael Benítez was a good manager constrained by lack of money. He did consistently well, and the moment he left the team dropped off considerably. During Roy Hodgson’s period as manager the team had very poor results, which started to improve, slowly, under Kenny Dalglish.

This improvement has continued under Brendan Rodgers without increasing in pace. Yet the improvement has had a rather odd shape. The team have sharply improved their attack, but this has been almost (but not quite) cancelled out by a deterioration in defence.

The improvement means that Liverpool have just popped back above Everton, having dipped below them. In 2008 Liverpool were 180 per cent as good as Everton, now they are only 113 per cent as good.

Daniel Finkelstein has been awarded the Royal Statistical Society Prize for excellence in statistical journalism for the Fink Tank

-------

Given this piece was written before the Southampton result, however it should not detract from the fact that we in general are both playing better football and accruing more points under Rodgers. Ignoring the stats about our title winning chances (no one in their right mind thinks we'll win the title), our start can only be a massive step in the right direction to finishing in the top four.

Onwards and upwards.

The chance of a top four finish rated at almost 50/50 is encouraging seems it's not just us rose tinted brigade showing optimism 

woof woof ! » Mon Sep 23, 2013 9:39 am wrote:The chance of a top four finish rated at almost 50/50 is encouraging seems it's not just us rose tinted brigade showing optimism 

Agreed. The article is not based on opinion whatsoever, purely probability.

SouthCoastShankly » Mon Sep 23, 2013 9:42 am wrote:

woof woof ! » Mon Sep 23, 2013 9:39 am wrote:The chance of a top four finish rated at almost 50/50 is encouraging seems it's not just us rose tinted brigade showing optimism 

Agreed. The article is not based on opinion whatsoever, purely probability.

Encouraging, although it's still a model though and models can be wrong.  I'm not sure we've a 50% chance of finishing above 2 of United, City, Chelsea, Arsenal and Spurs. I'd say more 15-20%?

Also I'd say our chance of winning it are 1-2%.

We have a strong first 11 when everyone is fit, our options off the bench are not good enough for top 4. So let's hope we Don't encounter any more injuries to key men.

fivecups » Mon Sep 23, 2013 5:38 pm wrote:Also I'd say our chance of winning it are 1-2%.

The difference being your numbers are pure speculation. Whilst I also don't think a league win will materialise, those numbers in the article cannot be challenged with opinion. Probability is the exact same method bookmakers determine odds and make millions each day. Like the phrase says 'the house always wins'.

These stats will change over the course of the season, I'll try and update them periodically.

There are only two stats that count at the end of the seasons - points gained and position

Teams don't win anything down to their "probability"

They win down to their squad and their results.

Off course "probability" can be challenged in sport - probability itself is speculation based on numbers.

Benny The Noon » Tue Sep 24, 2013 6:46 am wrote:There are only two stats that count at the end of the seasons - points gained and position

Teams don't win anything down to their "probability"

They win down to their squad and their results.

Off course "probability" can be challenged in sport - probability itself is speculation based on numbers.

I never said it couldn't be challenged, I said it couldn't be challenged with opinion, and by that I mean unsubstantiated opinion. You cannot say with any level of credibility we have a 1-2% of winning the league without evidence to support that claim.

The point that the Fink Tank makes, is they predicted through their own modelling that we would be where we are now in the league (give or take a point). Would you have made that prediction prior to the season?

Their predictions were made on a combination of most recent form and summer signings, the latter you have openly criticised as a failure.

So first thing - where is his article containing his prediction is printed "before" the event as opposed to after ? Surely if he so confident with his "probability predictions" we should be able to see his prediction before the event as opposed to after

And does he actual post the workings of this model ?

Can we see how he actually arrives at these stats ?

Benny The Noon » Tue Sep 24, 2013 11:38 am wrote:And does he actual post the workings of this model ?

Can we see how he actually arrives at these stats ?

I'm torn between it being FIFA or a Football Management game, it's an absolute nonsense to suggest there is an exact undeniable 'probability' attached to the outcome of a sporting event. Bookies odds change according to betting patterns, where they get them from originally is neither here nor there. One bookie reputedly made a complete pig's ear of their odds, offered Hilfenhaus at long odds (200-1?) to be top wicket taker in the Ashes and had to pay out when he ended top wicket taker in 2009.

Benny The Noon » Tue Sep 24, 2013 11:38 am wrote:So first thing - where is his article containing his prediction is printed "before" the event as opposed to after ? Surely if he so confident with his "probability predictions" we should be able to see his prediction before the event as opposed to after

And does he actual post the workings of this model ?

Can we see how he actually arrives at these stats ?

They have scored one more point than we expected them to at this stage

Model can be found here - http://www.dectech.co.uk/football_sites ... p_info.php

Cue the weak objections...

So do you have the prediction they made "before" the actual results ?

And it's nice they don't take into account players being injured or suspended

What a load of waffle

Send it to our Septic Owners - they would love all that nonsense and would file it alongside Moneyball.