Maths + Sorcery = 12% CL 51% EL

[pfim]

What are the chances we win our remaining three games though. I'd say 12% is optimistic.

I don't think that's what the information is saying. It's saying IF Everton win the last three matches, they will finish in the top 4 12% of the time. If they win their last 3 matches, they'll finish 5th 51% of the time, 6th 37% of the time.

 

[Royal Blue Jersey]

The problem with that surely is that football isn't a game of chance. If Bayern Munich played Rotherham they wouldn't have a 33% chance of winning.

It remains a game of chance, as does practically any eventuality in the universe, because things can never be utterly knowable. By that same token, a proper calculation of odds would take into account Munich's advantages in wealth, size, experience, quality, e.t.c all the myriad of things that makes them better than Rotherham, but it'd still just be an expression of chance, even if that would be, what, 90+%?

 

[njligernj]

Cool. Thanks for posting that. This is fairly common in US sports (fairly new too ... last 4 years maybe it's become something people pay attention to).

I don't think that's what the information is saying. *It's saying IF Everton win the last three matches, they will finish in the top 4 12% of the time. *If they win their last 3 matches, they'll finish 5th 51% of the time, 6th 37% of the time.

Correct. It's 12% IF we win the three. It's 0.4% overall. 0.4% chance that we both win the three games and things go our way with other results to get a CL spot. It's not as bad as it sounds -- in 2011 in baseball a team overcame 1 in 278 million odds to make the playoffs: http://sports.yahoo.com/mlb/blog/bi...actual-odds-the-rays-overcame?urn=mlb,wp21395

If (when?) we beat the RS the odds will change and improve (especially if results go our way in other games).

 

[Saliba]

The probabilities will depend on the inputs. If there is a standard Win/Loss/Draw formula, then yes, you'll get unrealistic output like this. If the inputs are based on a club's recent results, home/away splits etc then you'll get a fairly representational output. I'm sure the standard deviation will be fairly high, especially with clubs below ManU and ManCity, and above the two relegated clubs.

I've been studying way too much Econometrics that coming on to this forum for a break and seeing this makes me tear up.

 

[Royal Blue Jersey]

0.4% hhaha sounds about right. Winning all three games increases our chances by 2000% and it's still just about 1 in 9

 

[pfim]

Cool. Thanks for posting that. This is fairly common in US sports (fairly new too ... last 4 years maybe it's become something people pay attention to).

Correct. It's 12% IF we win the three. It's 0.4% overall. 0.4% chance that we both win the three games and things go our way with other results to get a CL spot. It's not as bad as it sounds -- in 2011 in baseball a team overcame 1 in 278 million odds to make the playoffs: http://sports.yahoo.com/mlb/blog/bi...actual-odds-the-rays-overcame?urn=mlb,wp21395

If (when?) we beat the RS the odds will change and improve (especially if results go our way in other games).

Yes, there are four probabilities, treat them all as independent for this purpose:

Win Spurs
Win West Ham
Win Chelsea
Finish top 4

The 12% finishing in top four is given. Let's say you believe there's a 20% chance each of beating Spurs and Chelsea, and a 60% chance of beating West Ham, the probabilities of all four happening are:

.12 x .2 x .2 x .6 = .00288, or .288%.

So basically, it's not happening.

 

[pfim]

I've been studying way too much Econometrics that coming on to this forum for a break and seeing this makes me tear up.

It's fun to apply this to something I enjoy instead of structured products, project scheduling, queuing theory etc.

 

[Royal Blue Jersey]

It's fun to apply this to something I enjoy instead of structured products, project scheduling, queuing theory etc.

Exactly, thats the best way of looking at it. You sound just like Steven Levitt

 

[tasslehoff]

Yes, there are four probabilities, treat them all as independent for this purpose:

Win Spurs
Win West Ham
Win Chelsea
Finish top 4

The 12% finishing in top four is given. Let's say you believe there's a 20% chance each of beating Spurs and Chelsea, and a 60% chance of beating West Ham, the probabilities of all four happening are:

.12 x .2 x .2 x .6 = .00288, or .288%.

So basically, it's not happening.

So your saying there is a chance?

 

[nigelcoyb]

So your saying there is a chance?

"Dumb and dumber" ... great shout that mate given the context of modern day EFC.

 

[mzone16]

Yes, there are four probabilities, treat them all as independent for this purpose:

Win Spurs
Win West Ham
Win Chelsea
Finish top 4

The 12% finishing in top four is given. Let's say you believe there's a 20% chance each of beating Spurs and Chelsea, and a 60% chance of beating West Ham, the probabilities of all four happening are:

.12 x .2 x .2 x .6 = .00288, or .288%.

So basically, it's not happening.

Spurs mate?? We aint got them to play!!