The Luck Index (ESPN) - ManUtd the Luckiest EPL Team of 2017-18

Shane88

Actually Nostradamus
Joined
Jan 12, 2011
Messages
35,173
Location
Targaryen loyalist
Look at them there, all around the boardroom table, spoofing away...

"Can you believe we're being paid for this?"

"Shh, someone will hear."

Luck index, for feck sake.
 

SER19

Full Member
Joined
Apr 15, 2008
Messages
12,565
This abomination is being discussed and roundly mocked already. A juvenile, flawed ‘study’
 

ZupZup

Full Member
Joined
Jan 8, 2014
Messages
2,399
Location
W3104
From 380 matches... they only used 157 incidents to come up with this table. So they deem there to be less than 1 significant potentially match changing incident in every two Premier League matches. Get in the sea.
 

Šjor Bepo

Wout is love, Wout is life; all hail Wout!
Joined
Mar 29, 2011
Messages
15,642
study and luck in the same sentence :lol: id love to hear what is luck in football.
 

The United Irishman

"Martial is championship material at best"
Joined
Apr 30, 2014
Messages
2,870
Location
Birmingham
A lot of my Liverpool associates have been posting this on facebook and twitter...I haven't even bothered reading it.
 

The United Irishman

"Martial is championship material at best"
Joined
Apr 30, 2014
Messages
2,870
Location
Birmingham
https://www.redandwhitekop.com/forum/index.php?topic=340752.0

The pick of the bunch:

"It's great timing for me after making a comment in the Sgt Shitcoats Spawny thread about how I thought the refereeing in the whiskeynose era was dodgy. That got waylaid by all sorts of tangential comments about respecting him and all that, but something I didn't mention there was the way Fergie used to look at a number of refs near the end of a game they were drawing and start pointing at his watch. Some games had 8 added minutes for them to score in. Looks like maybe they still have refs in their pocket after all."
 

DavidDeSchmikes

Full Member
Joined
Jan 20, 2013
Messages
17,225
It jumps from -43 for Arsenal to -73 for Liverpool.
It guess luck includes dry pitches, referees, offside goals at Old Trafford, teams that don't play 'football' and long balls
 

EyeInTheSky

Full Member
Joined
Sep 20, 2015
Messages
9,992
Location
On my sofa enjoying pineapple on its own
Here is what we can establish from this.

  • Every injustice they have benefited from does not count
  • Time and space has to stop at the moment they go ahead or are winning in a game and that would be the final result(it's only fair)

If I didn't know any better I would have said that this is surely not a pool fan since it is an obvious attempt to humiliate them and make them look pathetic. Then I think of things like RAWK and the Alternate league table and it all makes sense.

Got to laugh at the expected goals thing as well since they use that to say "Hey we should be winning every game with these predictive stats" irony is they fail to realise that if they actually believe that then it surely means that Klopp is not getting the best out of their "steller" players with failing to win feck all time and again.

The underlying theme here which everyone seems to have gone unnoticed is that they think they have a divine right to win and everyone else has either cheated them out of it or have serendipity on their side.
 

Handré1990

Full Member
Joined
Jun 25, 2013
Messages
4,818
Location
In hibernation
From 380 matches... they only used 157 incidents to come up with this table. So they deem there to be less than 1 significant potentially match changing incident in every two Premier League matches. Get in the sea.
This is exactly why it’s such a joke. They should leave ‘studies’ to competent people. My thesis advicer would throw me out a window if I came up with some shit like this. Like someone else posted, they probably couldn’t believe their luck in being allowed to ‘work’ on this.
 

Hammondo

Full Member
Joined
Aug 21, 2015
Messages
6,822
If this was showing the opposite then this forum would be loving it.
 

Keeps It tidy

Hates Messi
Joined
Jun 6, 2013
Messages
17,638
Location
New York
People always scoff at stuff like this but, I think we all agree our point total did not really match our performance over the season.
 

breakout67

Full Member
Joined
Apr 17, 2017
Messages
9,050
Supports
Man City
U need to watch that game again because the scoreline was flattering to deceive.
You don't win 4-0 by luck, it's 4 bloody goals. Our late goals were also quite well constructed.

I am quite confident that they deem it lucky because Pogba 'should' have been sent off when it would have been quite a harsh second yellow (City get away with those fouls every game).
 

GifLord

Better at GIFs than posts
Joined
Jun 7, 2013
Messages
22,898
Location
LALALAND
You don't win 4-0 by luck, it's 4 bloody goals. Our late goals were also quite well constructed.

I am quite confident that they deem it lucky because Pogba 'should' have been sent off when it would have been quite a harsh second yellow (City get away with those fouls every game).
https://ok.ru/video/346636225109
here's the extended HL
At 0:0 Swansea almost scored a fluke goal - if that went in Swansea would have gone full park the bus mode. They were already playing really defensively the whole match till the 80th min when the tried to go for it and opened up at the back - thats when the goals started coming for us.
 

breakout67

Full Member
Joined
Apr 17, 2017
Messages
9,050
Supports
Man City
https://ok.ru/video/346636225109
here's the extended HL
At 0:0 Swansea almost scored a fluke goal - if that went in Swansea would have gone full park the bus mode. They were already playing really defensively the whole match till the 80th min when the tried to go for it and opened up at the back - thats when the goals started coming for us.
So if Swansea scored a fluke goal we would have struggled? Do you know the implications of that? It means we would have been unlucky.

I think you are also missing the detail of us scoring from a corner at the end of the 1st half. Dead ball situations are much less affected by the game state than open play, even if Swansea were up and parked the bus that set piece goal would have probably happened anyway since we would be the ones attacking.
 

GifLord

Better at GIFs than posts
Joined
Jun 7, 2013
Messages
22,898
Location
LALALAND
So if Swansea scored a fluke goal we would have struggled? Do you know the implications of that? It means we would have been unlucky.

I think you are also missing the detail of us scoring from a corner at the end of the 1st half. Dead ball situations are much less affected by the game state than open play, even if Swansea were up and parked the bus that set piece goal would have probably happened anyway since we would be the ones attacking.
And if Pogba got a second yellow - which he deservedly should have? what then?

edit: You should also know we're pretty bad at coming back when behind
 

MikeKing

Full Member
Joined
Aug 7, 2017
Messages
5,125
Supports
Bournemouth
This is just great prep for the boom-bust cycle. Meltdown incoming immediately after the first "wrong decision", and it will be blown out of proportion. Now with a right class case study to back up their claims of unjust, how is it not something to entirely focus on during this season.
 

BluesJr

Owns the moral low ground
Joined
May 15, 2013
Messages
9,052
Stats in football is getting out of control. This is nonsense.
 

breakout67

Full Member
Joined
Apr 17, 2017
Messages
9,050
Supports
Man City
And if Pogba got a second yellow - which he deservedly should have? what then?

edit: You should also know we're pretty bad at coming back when behind
Maybe you just ignored what I wrote since I already addressed that?

That was a debateable yellow card; it was at the edge of Swansea's box and we had plenty of cover. If Pogba 'should' have been sent off then City would get a couple of red cards a game since they do those type of challenges several times a game unpunished.
 

El Zoido

Full Member
Joined
Jun 7, 2013
Messages
12,324
Location
UK
Liverpool least lucky. :lol: Did Klopp fund this study himself?
 

B20

HEY EVERYONE I IGNORE SOMEONE LOOK AT ME
Joined
Aug 23, 2003
Messages
27,591
Location
Disney Land
Supports
Liverpool
Business as usual then.
 

Fingeredmouse

Full Member
Joined
Aug 10, 2014
Messages
5,638
Location
Glasgow
Stats in football is getting out of control. This is nonsense.
It's also not statistics. It's ludicrously flawed, invalid, potentially cherry picked drivel of the highest order. I'd quite possibly sack one of my analysts if they delivered this excrement.
 

WR10

Correctly predicted France to win World Cup 2018
Joined
Jul 19, 2009
Messages
5,644
Location
Dream
So if we were lucky, why were we unlucky to be put in a position that we had to rely on luck?
 

paulscholes18

Full Member
Joined
Jan 4, 2013
Messages
20,173
https://ok.ru/video/346636225109
here's the extended HL
At 0:0 Swansea almost scored a fluke goal - if that went in Swansea would have gone full park the bus mode. They were already playing really defensively the whole match till the 80th min when the tried to go for it and opened up at the back - thats when the goals started coming for us.
Maybe they felt Pogba should have got a 2nd yellow when the score was 0-0
 

Snow

Somewhere down the lane, a licky boom boom down
Joined
Jul 10, 2007
Messages
33,317
Location
Lousy Smarch weather
Ok, so 3 people who are coders determine just by watcing the games which incidents are possibly incorrect. Straight away I have questions. How are they what is controversial? Also only 3 so that's roughly 126 matches each so not only is there room in error regarding if they catch an incident or not but also how they determine them.

From the thousands of incidents they randomly select 157. That's not a very big number and I don't think that if you're going to do this at all that should be randomizing a sample size but rather look at them all. Being random you could end up with every deemed incident from a single club and omit a lot from another.

Lastly a single person (a referee) decided on his own what was wrong and right. Refs aren't always right and even with hindsight they can disagree.

Lastly what this whole thing is really based on is this.

A detailed mathematical forecasting model was built by the team at the University of Bath to predict the outcomes of each game controlled for:

o Time of incident
o Penalty conversion rates
o Red card coefficients

• Other factors taken into account for forecasting were:

o Team strength
o Form
o Home advantage

• Each game was then simulated 100,000 times using a Bayesian Hierarchical Model, with the median score probability taken as the new result for each match.
Honestly you could probably have gotten a better result by using Football Manager than this model. These things don't take into consideration the mentality of players at all. There can be such a massive difference, a shift in mentality when something changes in a match, be it a red card or a goal. I don't think these metrics can take them sufficiently into account. Does it take into account the difference of conceding/scoring in the 2nd minuted and the 44th minute?

Lastly they say this
The team had 2 months to collect, analyse and forecast data from all 380 Premier League games
Just watching the games takes up 26 of those 60 days or almost half the time. Take away the time spent sleeping, eating and not working and how much time are you left with? This increases the chances of error even more.

I hope the team that did this isn't doing it for their doctorate.
 

theriddler

New Member
Joined
Aug 6, 2018
Messages
118
Supports
Liverpool FC
I'm a Liverpool fan, and to be honest, I don't believe this stuff one bit.

While I think there were one or two decisions that went against us, every side can have the same greivances.

The study doesn't account for bias in judgement of all the factors that cause the 'swing' in points. Further, simulation models can be highly tinkered with to give you the results you want to see by moving/removing variables - as far as I understand, which I think I do quite well considering I have a Masters in Math and make a good living from making simulation models :smirk:

Given my backgrounds, the use of statistics and numerical analysis in football is very interesting to me; however, this accounts for some things that only people who live in computer labs would ignore - emotion, momentum and bias. Also its the University of Bath, while quite decent, its not exactly Oxbridge. (Yes, I'm a nerd, sue me :drool:)