We know that, like us, many of you readers have made your own predictions.
We also know that you enjoy seeing how well our predictions fared and that a few of you are eager to know how our opinions fared against the power of two dice and randomness.
So we now proudly present the results of the Soccer Mad Boffins (and our two random dice) WC2018 Group Stage predictions.
GOAL TALLY
The first thing to mention is the number of goals scored so far is 122 in 48 games. Dividing the goals by the number of games gives a figure of 2.542 goals per game, which sounds about what we might realistically expect - mainly 2-0 or 2-1 score-lines, with a few 0-0 bore draws here and there off-set by a few 4-0 thrashings here and there to provide the headlines.
So how does this compare with the score-lines that we predicted? Here are the results:
Dr Alex G. Gillett (AG): 136 goals (that's +14 than the actual total of 122)
Dr Kevin D. Tennent (KT): 95 goals (-8)
Random Dice: 178 (+56!!! That would have been quite some World Cup)
All well and good, but how did that play out in terms of how accurately we predicted the outcomes (who would win? Or would they draw?) or even the number of goals that each team scored in the various games? How accurately did we predict the final group standings and which teams would progress to the next round? And for more fun, what exactly did the dice predict?
GAME & GROUP TABLE PREDICTIONS
Let's take a look at a group-by-group breakdown, and use the same points based competition as in our previous blog post about these predictions, in which each predictor (Alex, Kevin and Dice) could win up to 5 points per game, following the simple rule of:
Correct outcome (win / lose / draw)= 3 points
Correct 'Home' team goals = 1 pt
Correct 'Away' team goals = 1 pt
GROUP A
Game
|
Actual
Score
|
AG
|
KT
|
Dice
|
Russia V Saudi Arabia
|
5-0
|
2 - 0
|
2 - 1
|
1 - 2
|
Uruguay v Egypt
|
1
- 0
|
2 - 1
|
1 - 0
|
3 - 0
|
Russia v Egypt
|
3
- 1
|
2 - 1
|
1 - 0
|
1 - 3
|
Uruguay V Saudi Arabia
|
1
- 0
|
3 - 0
|
3 - 0
|
0 - 6
|
Saudi Arabia V Egypt
|
2
- 1
|
0 - 0
|
0 - 1
|
0 - 0
|
Uruguay v Russia
|
3
- 0
|
2 - 1
|
2 - 0
|
0 - 0
|
TOTAL POINTS
|
17
|
20
|
4
|
Actual Final Table
|
AG
|
KT
|
Dice
|
Uruguay
|
Uruguay
|
Uruguay
|
Saudi Arabia
|
Russia
|
Russia
|
Russia
|
Egypt
|
Saudi Arabia
|
Egypt
|
Egypt
|
Uruguay
|
Egypt
|
Saudi Arabia
|
Saudi Arabia
|
Russia
|
AG and KT both predicted same table,
correctly identifying Uruguay and Russia as first and second place,
respectively.
KT just edges it on points due to more
accurately predicting goals scored by each team in particular games.
The random dice score a paltry 4 after
wrongly predicting that Saudi Arabia would dominate the group undefeated,
starting and ending with a 2-1 win over
Russia and 0 – 0 draw with Egypt, via a
6-0 thrashing of Uruguay. None of this
actually happened, in fact Saudi Arabia began their campaign with a 5-0
drubbing at the hands of Russia, lost 0-1 to Uruguay, although they did beat
Egypt 2-1. But they still finished the group in bottom place.
GROUP B
Game
|
Actual
Score
|
AG
|
KT
|
Dice
|
Iran v Morocco
|
1-0
|
2 - 1
|
0 - 0
|
0 - 1
|
Portugal v Spain
|
3-3
|
1 - 1
|
1 - 1
|
0 – 0
|
Portugal v Morocco
|
1
- 0
|
3 - 0
|
2 - 0
|
4 - 5
|
Iran v Spain
|
0
- 1
|
1 - 3
|
0 - 2
|
2 -3
|
Iran V Portugal
|
1-1
|
0 - 3
|
0 - 2
|
0 - 3
|
Spain v Morocco
|
2
- 2
|
2 - 1
|
2 - 1
|
0 - 1
|
TOTAL POINTS
|
13
|
12
|
6
|
Actual Final Table
|
AG
|
KT
|
Dice
|
Spain
|
Portugal
|
Portugal
|
Morocco
|
Portugal
|
Spain
|
Spain
|
Portugal
|
Iran
|
Iran
|
Morocco
|
Spain
|
Morocco
|
Morocco
|
Iran
|
Iran
|
AG and KT both correctly favoured Spain and
Portugal to qualify from Group B, although overestimated the powers of
Cristiano Ronaldo, as the Portugese actually finished in 2nd place
behind Spain.
AG did best from this Group scoring 13
points.
The dice again favoured a long-shot team to
win the group – this time Morocco – although in reality they finished in last
place, notching just 1 point (somewhat ironically from a 2-2 draw with Group B
champs Spain).
GROUP C
Game
|
Actual
Score
|
AG
|
KT
|
Dice
|
France v Australia
|
2-1
|
3
- 0
|
3
- 1
|
2
- 3
|
Denmark v Peru
|
1
- 0
|
2
- 1
|
2
- 1
|
2
- 2
|
France v Peru
|
1
- 0
|
3
- 0
|
2
- 0
|
6
- 0
|
Denmark v Australia
|
1
- 1
|
2
- 1
|
1
- 1
|
2 -
1
|
Australia v Peru
|
0
- 2
|
2
- 1
|
2
- 1
|
2
- 2
|
Denmark v France
|
0
- 0
|
1
- 2
|
1
- 1
|
3
- 0
|
TOTAL POINTS
|
11
|
19
|
7
|
Actual Final Table
|
AG
|
KT
|
Dice
|
France
|
France
|
France
|
Denmark
|
Denmark
|
Denmark
|
Denmark
|
Australia
|
Peru
|
Australia
|
Australia
|
France
|
Australia
|
Peru
|
Peru
|
Peru
|
KT confidently won this round by a clear 8
points over AG. Once more the dice were
way behind.
Both AG and KT correctly predicted the top
two placings, whilst the dice opted for Australia to go through at the expense
of France (actually based on the first game which featured both of these teams,
the neutral observer might have been tempted to bet on that happening).
GROUP D
Game
|
Actual
Score
|
AG
|
KT
|
Dice
|
Argentina v Iceland
|
1
- 1
|
2
- 1
|
1
- 0
|
0
- 0
|
Croatia v Nigeria
|
2-
0
|
2
- 1
|
1
- 2
|
0
- 4
|
Argentina
v Croatia
|
0
- 3
|
2
- 2
|
2
- 1
|
3
- 1
|
Nigeria V Iceland
|
2
- 0
|
0
- 2
|
1
- 1
|
2
- 2
|
Nigeria v Argentina
|
1
- 2
|
0
-3
|
1
- 2
|
0
- 5
|
Iceland v Croatia
|
1
- 2
|
2
- 2
|
2
- 0
|
0
- 5
|
TOTAL POINTS
|
9
|
6
|
10
|
Actual Final Table
|
AG
|
KT
|
Dice
|
Croatia
|
Argentina
|
Argentina
|
Argentina
|
Argentina
|
Croatia
|
Iceland
|
Nigeria
|
Nigeria
|
Iceland
|
Nigeria
|
Croatia
|
Iceland
|
Nigeria
|
Croatia
|
Iceland
|
Group D was where the dice’ random
predictions bore fruit, scoring 10 points to beat AG and KT.
Interestingly only AG correctly identified
that Croatia and Argentina would progress to the next round, as KT and the dice
put too much faith in Iceland and Nigeria respectively.
In truth, this was a very close group which
went down to the final games.
GROUP E
Game
|
Actual
Score
|
AG
|
KT
|
Dice
|
Serbia v Costa Rica
|
1
- 0
|
1
- 1
|
2
- 0
|
2
- 0
|
Brazil v Switzerland
|
1
- 1
|
3
- 0
|
1
- 0
|
5
- 6
|
Brazil v Costa Rica
|
2
- 0
|
3
- 0
|
2
- 1
|
2
- 2
|
Serbia v Switzerland
|
1
- 2
|
1
- 2
|
1
- 1
|
1
- 4
|
Switzerland v Costa Rica
|
2
- 2
|
1
- 1
|
1
- 1
|
1
- 5
|
Serbia v Brazil
|
0
- 2
|
0
- 4
|
0
- 2
|
2
- 0
|
TOTAL POINTS
|
17
|
17
|
8
|
Actual Final Table
|
AG
|
KT
|
Dice
|
Brazil
|
Brazil
|
Brazil
|
Costa Rica
|
Switzerland
|
Switzerland
|
Serbia
|
Switzerland
|
Serbia
|
Costa Rica
|
Switzerland
|
Serbia
|
Costa Rica
|
Serbia
|
Costa Rica
|
Brazil
|
Again, AG got the top two places right, as
KT overlooked Switzerland but like AG tipped Brazil to win the group.
In contrast, the dice had Brazil as bottom
of the group – predicting a 5-6 defeat at the hands of the Swiss amongst other
results that did not actually occur.
When we rolled the dice for the rest of the
tournament, the dice ultimately predict Switzerland to win the World Cup, so
watch this space.
GROUP F
Game
|
Actual
Score
|
AG
|
KT
|
Dice
|
Mexico v Germany
|
1-0
|
1
- 3
|
0
- 1
|
0
- 3
|
Sweden v South Korea
|
1
- 0
|
3
- 0
|
2
- 0
|
1
- 2
|
South Korea v Mexico
|
1-
2
|
2
- 2
|
1
- 1
|
1
- 3
|
Germany v Sweden
|
2
- 1
|
2
- 1
|
3
- 2
|
3
- 3
|
Mexico v Sweden
|
0
- 3
|
2
- 1
|
0
- 1
|
1
- 0
|
South Korea v Germany
|
2
- 0
|
0
- 4
|
1
- 2
|
1
- 3
|
TOTAL POINTS
|
11
|
12
|
4
|
Actual Final Table
|
AG
|
KT
|
Dice
|
Sweden
|
Germany
|
Germany
|
Germany
|
Mexico
|
Mexico
|
Sweden
|
Mexico
|
South Korea
|
Sweden
|
Mexico
|
South Korea
|
Germany
|
South Korea
|
South Korea
|
Sweden
|
Group F totally threw all involved. EVEN THE RANDOM DICE PREDICTED GERMANY WOULD WIN THE GROUP. But after a lackluster showing, the Germans finished in last place meaning that nobody did particularly well here, KT scraping it by a single point over AG, with the dice coming last by scoring a paltry 4 points.
However, both AG and the dice noticed
Mexico’s potential. In fact the dice think Mexico will reach the final and come
second place to Switzerland. Only time will tell…
GROUP G
Game
|
Actual
Score
|
AG
|
KT
|
Dice
|
Belgium v Panama
|
3
- 0
|
3
- 0
|
2
- 0
|
3
- 1
|
Tunisia v England
|
1
- 2
|
1
- 2
|
0
- 0
|
6
- 2
|
Belgium v Tunisia
|
5
- 2
|
4
- 0
|
2
- 1
|
0
- 3
|
England v Panama
|
6
- 1
|
3
- 0
|
1
- 0
|
3
- 0
|
England v
Belgium
|
0
- 1
|
1
- 3
|
0
- 2
|
5
- 3
|
Panama v Tunisia
|
1
- 2
|
0
- 1
|
1
- 1
|
1
- 3
|
TOTAL POINTS
|
22
|
15
|
11
|
Actual Final Table
|
AG
|
KT
|
Dice
|
Belgium
|
Belgium
|
Belgium
|
Tunisia
|
England
|
England
|
England
|
England
|
Tunisia
|
Tunisia
|
Tunisia
|
Belgium
|
Panama
|
Panama
|
Panama
|
Panama
|
Both AG and KT correctly predicted the
final placings of Group G, although AG’s exact predictions of the score-lines
for two games (Belgium v Panama, and Tunisia v England) helped him to best in
this round, with 22 points.
Panama made a valiant effort in their first World Cup Finals but predictably finished bottom place despite a good effort.
Panama made a valiant effort in their first World Cup Finals but predictably finished bottom place despite a good effort.
GROUP H
Game
|
Actual
Score
|
AG
|
KT
|
Dice
|
Columbia v Japan
|
1
- 2
|
2
- 0
|
1
- 0
|
0
- 2
|
Poland v Senegal
|
1
- 2
|
1
- 1
|
1
- 2
|
0
-0
|
Japan v Senegal
|
2
- 2
|
0
- 0
|
0
- 1
|
2
- 1
|
Poland v Colombia
|
0
- 3
|
1
- 1
|
0
-1
|
1
- 0
|
Senegal v Columbia
|
0
- 1
|
0
- 2
|
0
- 0
|
3
- 4
|
Japan v Poland
|
0
- 1
|
1
- 2
|
0
- 1
|
0
-2
|
TOTAL POINTS
|
11
|
16
|
12
|
Actual Final Table
|
AG
|
KT
|
Dice
|
Colombia
|
Colombia
|
Senegal
|
Poland
|
Japan
|
Poland
|
Colombia
|
Japan
|
Senegal
|
Senegal
|
Poland
|
Colombia
|
Poland
|
Japan
|
Japan
|
Senegal
|
Both AG and KT underestimated the Japan
team, wrongly predicting 4th place for the team which actually
finished as runner-up in Group H.
However, both correctly calculated Columbia’s qualification to the next
round.
Interestingly the dice got Japan right but
pitched Poland for top prize, which was unfortunate because they cam 4th.
Still though, the dice beat AG on points
for Group H, but KT’s superior predictive skills meat that he took the Group
with 16 points.
Dr Tennent celebrates his Gold Medal for winning First Place |
Furthermore, we are not the only experts to try to predict the World Cup, and some have tried to do it in an even more sophisticated (although perhaps not more accurate) way. The Financial Times reported that Goldman Sachs updated their computer assisted statistics driven 'soccermeter' to to take into account goal differences up to June 26, and predicted that England 'might' reach the final and play Brazil, beating Mexico and Spain on the path to the World Cup Final! Meanwhile another sophisticated 'machine learning' technology simulated the outcome of the tournament 100,000 times using the bookmaker's odds. They agreed with us - that a Brazil-Germany final was most likely, though Spain were strongly favored in their model! Needless to say, this demonstrates that the most sophisticated Artificial Intelligence is fallible when it comes to soccer!
We have a FREE research paper published by Project Management Journal available here.
Our book 'Foundations of Managing Sporting Events' was published in paperback earlier this month - and you can buy it for a special price direct from the publisher by quoting discount code: FLR40. Hardback and E-book also available.
No comments:
Post a Comment