This is the Football in Cider (FiC) Girl creating your Top Tips

The Football in Cider Way

How confident are you when it comes to knowing which football team to back? Do you normally get it right, go on a hunch, use a form guide maybe? The ability to see the future and predict the winner not only gives fantastic bragging rights, but can also make you very rich.

In this article we are going to discuss how predictable a football game can be and how we at Football in Cider use technology to get the best probability of the game's outcome.

 

Are Football Matches Predictable?

Before we get into the nitty gritty details of our prediction system we really should ask the basic question of whether it is actually possible to predict the outcome of a football match? Many argue that getting 22 humans onto a small field to kick an inflated hide of a dead cow gives rise to unpredictable results, and with some justification.

 

Are Football Games Random

There are plenty of examples of what appears to be randomness in football. Leicester winning game after game, having previously fought off relegation, to win the Premiership in 2016. And do you remember Denmark in 1992 winning the European Cup after failing to qualify.

 

1988 FA Cup Final Liverpool v Wimbledon

For me my favourite example of the unexpected result has to be Liverpool v Wimbledon in the 1988 FA Cup Final. Liverpool had already won the league and were fully expected to win a second double at Wembley (at the time aka Anfield South). They were in fine form. Wimbledon had only been promoted a year earlier and were expected to finish mid table at best. However, in a game that provided plenty of action and entertainment, Wimbledon survived the Liverpool onslaught, including saving a penalty, and managed to win the game and the FA Cup. Against all expectations.

 

Boring Boring Arsenal

However.

There is another side to this story. There happens to be some pretty hard facts that suggest the outcome of a football game is as good as over before it even kicks off!

  • Sir Alex Ferguson has won 12 Premier League titles out of a possible 24!

  • Only six teams have won the Premiership title:

    • Manchester United (13)

    • Chelsea (4)

    • Arsenal (3)

    • Manchester City (2)

    • Blackburn Rovers

    • Leicester City (1)

  • In 2003-2004 Arsenal went the whole season unbeaten (mainly winning 1 - 0)

  • Sunderland (05-06) and Derby (07-08) both only won one home game in a season

  • Derby went an amazing 32 games without a win in 2007-08

  • A third of the clubs that were in the division last season were in either exactly the same position as last season or just one place different (at time of writing).

 

Rochdale v Barcelona

Unfortunately for Rochdale fans, I think we would all agree, if The Dale went over to play Barcelona in a European Championship Final, Barcelona would win.

If they played 100 times, then Barcelona would win at least 95 times, possibly more, and the other 5 games would be a draw at best or shockingly Rochdale may sneak, against the run of the play, a remarkable history making win (Maybe the ref was having a bad day!).

All things considered, football matches are, to some degree, predictable. But like most things in life, nothing is certain.

 

Championship Manager 2 - The best game ever!

What exactly determines the probability of a football match? Before you answer this yourself let me roll on: Championship Manager 2 (1995 edition) listed skills you could search on when looking for a player. To me, this is a good list of influences that may affect a football match:

  • Confidence
  • Fitness
  • Aggression
  • Creativity
  • Determination
  • Dribbling
  • Flair
  • Heading/Aerial Ability
  • Influence
  • Injury Proneness
  • Intelligence
  • Marking
  • Off the Ball
  • Pace
  • Passing
  • Position
  • Set Pieces
  • Shooting
  • Stamina
  • Strength
  • Tackling
  • Shot Stopping
  • Technique

 

Smoky Pub

Joking aside, this is a pretty good start. There’s obviously a lot more discussion we can have here. We will all have a different opinion on this. All we are really missing is a smoky pub, some pints of beer (or cider) and a long afternoon with nothing else to do but emulate Alex Ferguson.

  

Footballing Facts

What is critical here is not actually the list of all things that affects a football match, but rather the “Facts” that affect a football match.

We need to separate facts from fiction
We need to separate the truth from opinion.

We need to realise that facts can not be disputed, whilst opinion is derived from fact and is totally subjective.

Previous results are fact. Future results are opinion.
How many injuries a team has is fact, how much the team will miss the players is opinion
How many missed chances a player had is fact. Why he missed is opinion
Etc.

 

Diego Costa

By focusing solely on the facts, only then can we form an opinion. It sounds simple, but many people form an opinion, treat the opinion as fact and then base their judgment on the opinion. “We’re never gonna win, Diego Costa is so shit he couldn’t even score in a brothel!” The opinion here is that Costa is shit. If instead you announced at full volume to everyone at the pub

 

“We’re almost certainly not going win, Mr Costa has only managed to score one goal in the previous 10 games compared to a league average of 1 goal every 1.6 games for all other premiership strikers that have averaged over 70 mins of play”

 

Then, and only then, do you have a point.

 

You will also have a broken nose.

You will also be outside picking yourself up from the pavement

You will also be on your own.

  

Best Leave The Facts To The Computer

Most of us should not be announcing boring facts to our footy mates when discussing the game at the pub on a Saturday afternoon. We should leave this to the computer. In the pub, opinions are fine. In the computer, facts are fine. Don't get this mixed up.

 

Footballing Geek Time

Predicting games with computers is simply a case of putting the facts in and letting the computer detect and test for patterns.

 

Ooh Patterns

Patterns are established by combining facts with other facts by using operators (both, or, one of etc) and regression testing.

Regression testing is where we go back in time and apply the tests to previous results repeatedly. If a link is found between a fact and another fact then the link is reinforced 

 

By reinforcing connections just a little for 1000’s of regression test we allow some flexibility with the patterns, ie maybe they are true 40 out of 50 times, This flexibility likens these combinations to our very own human opinions:

 

“Costa hasnt scored in the last 10 games when Hazard played on the left (except twice) which means I dont think he will score today as Hazard is again on the left.”

 

Football in Cider’s Neural Network

We are now at a point where we can explain how Football in Cider uses facts as inputs to predict football results.

 

The idea is simple: We allow the computer to observe the facts.

These facts are then combined together to see if any patterns emerge.

The computer then repeats the combination process again. Applying combinations with combinations to get another layer of combinations.

Again It looks for patterns.

We actually perform, in total, 3 layers of combinations.

 

Layer Cake

After the last layer of combinations have been regression tested the system is ready and primed for the current games facts to be entered.

The facts will trigger the strongest links that will eventuate in a game outcome.

 

Sometimes, the network outputs different outcomes, but on other occasions the outcomes all agree.

 

Each path through the network will also offer a probability that is defined by the strength of its connections. This probability is the key as to whether to place a bet or not.

 

Beating The Bookies

Finally it is a simple case of comparing our probability with the bookies probability and purchasing a bet if it is favourable.

 

Making Money

What this means to us, football in cider members, is that we can see a Prediction and a probability. For example the Predictions Page may declare Arsenal will win 2 of every 3 games (66%) at home to Spurs and the bookies are offering odds of 1.8. This means on average if we put $10 on Arsenal to win we will win $36 at a cost of $30. Making a profit of $6 from our original $30 which is a 20% return. This is the ROI value you can see alongside every prediction..

Only predictions with a high ROI and strong patterns associated with both teams are included as Top Tips.

 
 

Final Whisltle

The final whistle has blown and hopefuly you now have a slightly better understanding of how Football in Cider achieves its accurate predictions. If you have any questions or would like further informationi please do not hesitate to contact The Prediction Guru on out Contact Us page. Have fun with your betting!