Colossus Bet
The Colossus website lets you choose your pick of results from a list of 7 games that they select. It is similar to picking lotto numbers. Then based on the actual results of the matches one lucky person may win the jackpot (or share in the jackpot if there is more than one correct pick). The jackpot is a whopping £10,000,000.
Now unlike the lotto each pick is not equally likely to come up. You should use this to your advantage and only pick the most likely scores. Below chart shows the results for the English Premier League since August 1993. We use the AOD (any other draw), AOA (any other away win) and AOH (any other home win) to classify the results in the same way that Colossus does. This chart gives an indication of exactly how likely different scores are.
As you can see 1-1 draw is the most frequently occuring result, followed by 1-0 home win and a 2-1 home win. These 3 scorelines make up 30% of all match outcomes. Least likely is 3-2 home or away and a 3-0 away win, those prediction would be unlikely to win going by past history.
We can go one step further and estimate the actual probabilities of each specific match. We know that the bookmakers make a lot of money by being right (otherwise they would be out of business), so lets take their published odds to estimate the probability of each match. Experts and pundits like Michael Owen understand football but generally don't have expertise in probability theory. The bookmakers are experts in both fields and also have a huge bank of statistics to back up their odds.
So take for example Sunderland vs Man Utd. One popular bookmaker is offering the following odds.
The outcomes above are given with the Colossus classification in brackets. The chance of a scoreless draw is 9/1 and a 1-1 draw is 6/1. To get the chance of Any Other Draw we would add 13/1 + 55/1 + 325/1. Fractional odds are hard to add together, or multiply together. So, we can translate the odds to percentages to get the implied probability for each outcome. This will make it a lot easier to understand and to add or combine the chances. For example 13/1 means 13 unfavourable outcomes for each one favourable outcome, so a 1 in 14 chance or a 7.1% implied probability.
It is important to understand that these implied probabilities are just estimates. The actual probability will be a little less. The bookie will have built in a bias by exaggerating the actual probability to ensure that they have an edge. In the Correct Score market the total bookies implied probability can add up to as much as 130%. The total probability by definition should be 100% so the bookie has a 30% edge in this market. Not very profitable for the punter! For this reason I will adjust the bookies implied probabilities by dividing by 130% for the purpose of realistic calculations.
Once you have calculated the probabilities for each result, it is then very straightforward to pick the most likely result - i.e. the one with the highest probability. Repeat this for each of the 7 matches to get your picks. Below table lists the optimal picks. These picks will give you the greatest chance of being right.
The great thing about converting the odds to probabilities is that we can now multiply the probability of each match to get the probability that ALL our scores will be correct. In this case:
12.8% x 11.0% x 12.8% x 12.8% x 12.0% X 11.8% x 16.4% = 0.0000054%
Now unlike the lotto each pick is not equally likely to come up. You should use this to your advantage and only pick the most likely scores. Below chart shows the results for the English Premier League since August 1993. We use the AOD (any other draw), AOA (any other away win) and AOH (any other home win) to classify the results in the same way that Colossus does. This chart gives an indication of exactly how likely different scores are.
Result of all premiership matches since Aug 1993 classified by scoreline.
As you can see 1-1 draw is the most frequently occuring result, followed by 1-0 home win and a 2-1 home win. These 3 scorelines make up 30% of all match outcomes. Least likely is 3-2 home or away and a 3-0 away win, those prediction would be unlikely to win going by past history.
We can go one step further and estimate the actual probabilities of each specific match. We know that the bookmakers make a lot of money by being right (otherwise they would be out of business), so lets take their published odds to estimate the probability of each match. Experts and pundits like Michael Owen understand football but generally don't have expertise in probability theory. The bookmakers are experts in both fields and also have a huge bank of statistics to back up their odds.
So take for example Sunderland vs Man Utd. One popular bookmaker is offering the following odds.
Odds offered by one popular bookmaker in the Correct Score market for Sunderland vs Man Utd.
The outcomes above are given with the Colossus classification in brackets. The chance of a scoreless draw is 9/1 and a 1-1 draw is 6/1. To get the chance of Any Other Draw we would add 13/1 + 55/1 + 325/1. Fractional odds are hard to add together, or multiply together. So, we can translate the odds to percentages to get the implied probability for each outcome. This will make it a lot easier to understand and to add or combine the chances. For example 13/1 means 13 unfavourable outcomes for each one favourable outcome, so a 1 in 14 chance or a 7.1% implied probability.
It is important to understand that these implied probabilities are just estimates. The actual probability will be a little less. The bookie will have built in a bias by exaggerating the actual probability to ensure that they have an edge. In the Correct Score market the total bookies implied probability can add up to as much as 130%. The total probability by definition should be 100% so the bookie has a 30% edge in this market. Not very profitable for the punter! For this reason I will adjust the bookies implied probabilities by dividing by 130% for the purpose of realistic calculations.
Once you have calculated the probabilities for each result, it is then very straightforward to pick the most likely result - i.e. the one with the highest probability. Repeat this for each of the 7 matches to get your picks. Below table lists the optimal picks. These picks will give you the greatest chance of being right.
Optimal picks based on bookies implied probability.
The great thing about converting the odds to probabilities is that we can now multiply the probability of each match to get the probability that ALL our scores will be correct. In this case:
12.8% x 11.0% x 12.8% x 12.8% x 12.0% X 11.8% x 16.4% = 0.0000054%
This equates to less than 1 in a million chance of being right!
The overall prize of £10,000,000 definitely makes this a game worth playing though. For your £2 stake per line, by picking the most likely outcomes you have an Expected Value for your payout of £5.40. That means that if you could continue to play this game billions of times in your life time that you would win nothing a lot, and win the 10 million very rarely. It would average out at a £5.40 average win. That is a very healthy positive expected value. It definitely beats the National Lottery for value.
That can be overly theoretical, especially considering there is nowhere near enough weeks even in a very long lifetime for this long run average to be achieved. So feel free to use your own gut feeling on whether a measly £2 punt is worth the chance of a life changing payout.
For first 72 hours of bets matched with free bets used this promo code when signing up. Promo Code: RAFNAS13L0
If you liked this article you can find more of the same on my blog at fog2014.blogspot.ie or follow me on twitter @FoG_BLoG.
For first 72 hours of bets matched with free bets used this promo code when signing up. Promo Code: RAFNAS13L0
If you liked this article you can find more of the same on my blog at fog2014.blogspot.ie or follow me on twitter @FoG_BLoG.
Comments
Post a Comment