Card Counting......A New Angle......?

[spreadman]

Card Counting - A New Angle....? I was watching a programme recently about the American mathematics student card counters who capitalised on an existing mathematical formula that would tip the odds for a game of Blackjack in the player's favour. I wondered if this concept could be adapted and applied to sport's betting, and more specifically, betting on televised football matches. Imagine each match is a different 'card' in a pack. If during the football season there are, say, 250 televised matches, the pack would consist of 250 different cards, each with a different face, unique, but with one common numerical denominator - the number of goals scored during the match. This pack of cards would have more 'aces' and 'zeros', than 3's or 4's, but we would know, from previous seasons' goal averages, that in the pack , before it was 'dealt', (televised), there would be, a certain per centage of 0's,1's,2's, 3's, etc., just like in a real packof playing cards. Only, in our football deck, there wouldn't be four cards of each number. Which doesn't matter to us, and the blackjack card counter doesn't need to keep track of the cards' numerical values, he just assigns values, negative or psitive, to low value card numbers, (2,3,4,5,etc.), medium, (7,8,9), and high,(10 through to ace). All we would do would be to keep a record of how many 1 goal, two goal etc. matches,or 'cards' had been dealt to us, (televised), and bet accordingly. We would be applying the card counters strategy to games of football. When the counted number value is in the counter's favour, he plays, when it is not, he doesn't. Simple as that. If there are more high cards left in the dealer's pack, you have a better chance of '21', or at least a better hand than the dealer. So, stretching our imaginations, could this be applied, successfully to betting, and in particular, spread betting on low or high scoring matches? If we know during a season there will be 20% of matches with 1 goal scored, 25% with 2 goals etc. 10% with 0 goals, that tips the odds in our favour for an 'under 2.5' bet. The score% may vary, but at the moment, the Premier goal average is 2.1 per match, and most televised games that I trade on are Premier matches. Now here is the questionable premise : if a series of 3 or 4 televised match results are higher than average, e.g. 3-0,4-1,2-3,6-0, is it likely that the next match will be low scoring? To use the card counting analogy, if a sequence dealt is J,J,K,A,10,Q, will a 2 or 3, or a lower card number, be dealt next from the deck? This certainly is the premise for the card counter, who would consider sitting the hand out, or beting low stakes. The danger here, as I see it is the 'gamblers fallacy' - the misguide belief that in a lottery, roulette or two outcome random event such as coin tossing, because four heads have just been tossed, a tail will follow next. This is untrue, in a 'fair' situation, with a sound coin, there is an even chance for a head or tail to be the next outcome. What is true, however, is that the likelihood of a 3 heads sequence becoming a 6 heads sequence, in a limited coin tossing sequence of, say 250 tosses, is less likely, I make it 8/1 against, (2x2x2), though someone may correct me. So, in theory, with a set number of total goal scores expected in 250 matches, it may be possible to predict when a low score result is due, and bet accordingly, knowing the 'deck' to be dealt contains 'cards', namely 'results to come', that are low in (goal) value, and thus in our favour. I can back this theory up with my past results recorded data. I have noticed that after a set of 2 or 3 high scoring results, the next match is a low score 1-0 or 2-0. The longest losing run that I have recorded is 3 consecutive high score matches, followed,. obviously, by a low score match. Indeed, a graph of the goals scored over a season of televised matches has a 'wave' or 'rollercoaster' pattern, with very few 'plateaus'. Below I have recorded the goal totals for this season's televised Sky matches, English leagues, so far : 5,3,0,3,2,4,2,0,2,1,1,3,1,3,1,1,3,0,5,0,3,2,4,1,2,4,0,5,2,3,2,0,4,1,1,2,0,3, 4,2,1,0,4,4,1,2,1,1,2,2,1,5,4,3,2...... Interestingly, you will notice that after a high goals score, generally a low score follows. It is this factor that I hope to benefit from, just as the card counter benefits from a deck in his favour...... I will endeavour to use this 'card counting' system for a spread betting market called 'Total Goal Minutes, a highly risky, volatile market, but with the odds, (pack!), loaded in our favour, I hope to make it profitable. :hope That's enough for now, more details will be posted in a day or two's time. Look out for a new thread, probably called, 'Spreadman's TGM'S System', or something equally unoriginal. Hope the 'TGM' doesn't stand for 'Totally Gone Mad'...:\ I await your response, questions and abuse (!), thank you for listening. :ok

[davidc]

Re: Card Counting......A New Angle......? Hi Spreadman - interesting strategy, good luck with it. Just a couple of small points. Your analogy between blackjack and football looks good on the face of it but there could be a problem with using historical goal stats versus a definite (fixed chance) strategy in blackjack. Using historical stats can be problematic as there is always the possibility that the stats will change over time, for instance the premier league goal stats for the first few rounds this season were lower than in previous seasons. In blackjack, when following a strategy like card counting the odds for the next card to be dealt can be calculated exactly and you can then bet based on the odds of hitting the right cards with a reasonable chance of winning over time. Also the cards in the decks are always the same and the only outside influence on these cards is when they are shuffled whereas with your TGM system there are many outside influences which can affect the result. Have you checked the odds for the under on live sky matches this season and if so do you have an edge. There were 34 from the 55 matches listed which finished under and (if I have worked it out correctly) you would need to have odds of 1.62 or better to give you the advantage for level stake fixed odds betting. It may be a good idea to back test the system with some refinements to see if these figures can be improved. I haven't done any research but you might find that a large proportion of the over games may be teams near the top playing teams near the bottom and you could therefore rule out these games to give you an increased edge. I realise that you are not looking at the fixed odds market but I have no experience in the spread betting markets. Good luck.