XX Draw Selections Information Sheet Summary As the name implies, the XX Draw Selections are matches where my ratings suggest that the draw price is overpriced, and thus value. They are not tips as such, since the majority of these selections will lose, but educated bettors will understand that the key to profitable betting is in finding value, not simply winners. Selections are the result of mathematical computations. It is seven years since I stumbled upon the Betting Exchange concept, and the playing field these days is a lot more level. If you approach sports betting in a disciplined way, it is an activity that can generate consistent profits. History The selections actually came about by accident. At the start of the 2008-09 season I started an Elo based rating system. In contrast to a league where a win is always worth three points, wins using Elo ratings are not all the same. A win against a strong team will boost a rating more than a win against a weak team, and a loss against a strong team will not be as damaging to the rating as a loss against a weak team would be. As time went on, the spreadsheet I use for the ratings became more sophisticated. Rather than just look at the result, I started to factor in the winning margin, so that a two goal win was better than a one goal win for example. This meant that a win didn’t necessarily improve a team’s rating if their margin of victory was less than predicted. Then I started to look further than just the final score, which can be a poor reflection of the game. We’ve all seen games where one team has dominated, only to lose to a breakaway goal. One striking example from the EPL last season (2010-2011) was the match between Birmingham City and Chelsea. Chelsea had 24 shots to Birmingham City’s 1. Chelsea had 9 shots on target to Birmingham’s 1. Chelsea were awarded 14 corners to Birmingham’s 2, yet the result was a 1-0 win for Birmingham. It is my opinion that by including more statistics than simply goals, more accurate ratings are achieved. Birmingham did not deserve to receive a big boost to their ratings that day – the overall statistics did not merit that. The accident previously mentioned, was that as I entered results each week, I realised that, with more work than I initially thought, it would be possible to add a predictive feature to the spreadsheet. In a perfect world, team ratings would be almost constant, but of course, football isn’t like that. Teams change all the time, and Manchester United do not always beat Liverpool who do not always beat Aston Villa who do not always beat Sunderland etc., but if teams were to play true to their rating, what result would mean that both teams maintain their current rating? As I entered results each week, I noticed that when games were drawn, it wasn’t that unusual for the ratings of both teams to remain the same or only change by a point or two. In other words, the Elo ratings ‘expected’ the game to be a draw. I started to track these matches, and the result is the XX Draw Selections. The spreadsheet actually predicts outcomes to 0.1 of a goal and the XX Draw picks are those games where the margin of victory needed to maintain the rating is between +0.1 and -0.1. Previous Results Since I started tracking the outcomes of games where the result needed to maintain the ratings was in the +0.1 to -0.1 range the results have been pleasing to say the least. 260 matches have qualified across the top five leagues in Europe, and with 91 winners, the strike rate is 35.0% (2.86 in Betting Exchange terms). With the draw price on these selections currently averaging 3.478, this is a significant edge. To be honest, I only started tracking the available prices at the start of 2011, but this seems a reasonable price to assume since the draw price remains fairly consistent, and this number will be updated as more data is collected. The table below shows the overall all-time results up to 24th November, 2011 as well as the results broken down by league. The figures in italics are projected numbers based on the average prices from each league, which have only been tracked since January 2011. With the sample size still fairly small, it’s not surprising that we see outliers – France’s Ligue 1 is very strong, but even the poorest performing Serie A offers a reasonable return. Overall, it is a solid enough record. The 2011-12 results to date are actually exceeding the long-term returns. Whether this is because the numbers are becoming more reliable as more data is entered, or just noise in the data remains to be seen. 2010-11 Results January through May Selections Wins Strike Rate Profit ROI 69 24 34.8% 13.45 19.5% 2011-12 Results (to 24th November, 2011) Selections Wins Strike Rate Profit ROI 56 21 37.5% 17.35 31.0% Before You Subscribe Backing draws is not for those looking for a quick buck. Draws are hard to predict, which is why the average price is around 3.47 (implied probability slightly below 29%). This means long losing runs are inevitable. For example, March 2011 had just one winner from ten selections, and there was a losing run of 16 matches early in 2011-12. On the other hand, the occasional round of matches comes along where only a 90th minute goal comes between a five for five (500-1 odds) outcome, such as September 24-25, 2011. I am confident that these selections will remain profitable in the long-term, but can give no guarantee that any given weekend, or even given month, will be profitable. Sometimes it feels that fate is against us. In 2011-12 I have started recording the minute when the draw bet was lost, and already we have seen a 92nd minute penalty take away a win. Handling these ‘bad beats’ is part of the game, and it helps to understand that what goes against you one week can work in your favour the next. What To Expect The XX Draw selections will usually be sent out at least two days in advance – for weekends by Thursday morning at the latest unless there are midweek games, in which case selections will be sent out as soon as possible. The draw prices on these selections typically have plenty of value built in, so if kick-off approaches and you are not yet matched, then taking the available price is still value. The number of selections will vary. Some weeks there may be none. The number of selections will be determined solely by the numbers. I will not feel obliged to offer a selection if there are none, nor will I cap the number of selections on any weekend. Such artificial restrictions make no sense. It should go without saying that your e-mail address will never be shared or distributed and whether you decide to join or not is entirely up to you. There will be no high pressure follow up e-mails, or late night knocks at your door. Using The Selections How you use these selections is up to you. I usually back these selections as single bets, but I can see the appeal of accumulators. Some people may choose to trade out after a certain time, while others may stay in for the full 90+ minutes. Some may want to take me on, and oppose the selections. That’s fine too. These are simply selections that the numbers suggest offer big value on the draw. Unlike most services, I do not analyse the games in great detail and form a subjective opinion. I simply enter in the teams, and the spreadsheet spits out the numbers. If there has been some major news leading up to the game that makes you uncomfortable backing the draw, I won’t be offended. Under 2.5 Goals One other recent observation is that there is an apparent upside to backing these selections on the Under 2.5 goals markets. I started recording these prices at the start of the 2011-12 season, and as of 24th November, 2011, of 56 selections, 34 have been Under 2.5 goals. That is a 60.7% strike rate, with an implied price of 1.65. When you are looking for draws, you obviously want matches where the probability of goals is low, and as I have written on my blog, in a way, it is more satisfying to select a match that ends 1-0 than it is to fluke a 3-3 draw. Under 2.5 Goals 2011-12 Results (to 24th November, 2011) Selections Wins Strike Rate Profit ROI 56 34 60.7% 8.04 14.4% Staking Whether you want to use level stakes or a progressive staking system is also up to you, but it goes without saying to stay within your comfort zone. Only bet with money that you can afford to lose, and whatever you might read on forums or elsewhere, no staking plan in the world can turn a losing system into a profitable one. Prices The prices quoted will be those that should be available on the Betting Exchanges. If the draw price is 3.45 to Back and 3.5 to Lay at the time I send out the selections, I will use the 3.5. This means that I suggest placing a bet asking for 3.5 rather than taking the 3.45 that is available. The draw price seldom shortens significantly, and if your money is up early enough, you will be matched at your asking price more often than not. If other subscribers have acted on their e-mail ahead of you and driven the price down by a tick (.05), the draw is still value. At a strike rate of 35%, a price above 3.0 is still great value, but I would recommend asking for a price more often than you take one, because those .05s add up over a season and make a big difference. Guarantee While I can’t guarantee that the selections will be profitable, I am confident that they will be. Should they fall short of expectations, I will happily refund up to 50% of the subscription fee. I’m not doing this for the money, and losses will hurt me more than any subscribers. If the 2011-12 selections finish with a negative ROI, the refund will be on a sliding scale. Refund Sliding Scale ROI Loss: 0.01% to 4.99% 5% to 9.99% 10% to 14.99% 15% to 19.99% 20%+ Refund: 10% 20% 30% 40% 50% Fees I have decided to set the initial subscription fee for 2011-12 at £99 for anyone subscribing in September, and (another) sliding scale will enable new subscribers to join later in the season for a reduced fee. This will cover the remainder of the 2011-12 season, and there should be well over 100 selections, which makes the cost per selection very reasonable when compared to other full-time services. Anyone joining on or after the 20th of a month will pay the fee quoted for the following month. Sep./Oct. Nov. Dec. Jan. Feb. Mar. Apr. £99 £85 £70 £55 £40 £30 £20 Monthly Fees As there may be a desire from some people to subscribe for a shorter period, a £19 fee per month alternative is offered. Anyone joining on or after the 20th of a month will receive the selections through to the end of the following month. Free Trial There will be no free trial period, for a couple of reasons. One is that the XX Draw selections have been published for several months at Green All Over or Gold All Over, and what you have seen there is what you will get, and second, I don’t need the administrative headache of keeping track of when Free Trials are due to end. Cancellation If anyone on a full subscription wishes to cancel during the season, they are free to do so. The refund amount will depend on the month the request to cancel is received using the above table (in Fees), so for example, a subscriber requesting a refund in January would receive £55. Membership has to be for at least one month. Capping Members Given the strength of the draw price, and the number of outlets available for backing it, I seriously doubt that a few subscribers will make the capping of members something I need to consider. Backing the draw just isn’t a popular bet. About Me Anyone reading this doubtless knows me as the writer of the long running blog Green All Over. I frequently pay the Premium Charge, and concentrate mostly on football, basketball, baseball and American Football. I have been trading and betting on Betfair since 2004. After having betting accounts closed down in the early 1990s, the result of my betting selectively after comparing the prices available in the Racing Post, I lost all interest in betting, and concentrated on a career in IT and raising my children. Besides, back then there wasn’t the wide range of betting opportunities that there are now, although things were starting to change. It was basically football and horse racing or dogs, and without inside information, the over- round that bookies work to made it almost impossible to make much of a profit, which is the whole idea from the bookie’s perspective I suppose! Then in March 2004, I read an article about the Betting Exchanges concept, and realising that this could be a perfect fit for my love of sport and statistics, opened an account with an initial (and cautious) deposit of £100 (actually £98.50 after fees). That remains the only deposit I have ever made. After a slow start -my apprenticeship - I have turned that two figure sum into a six figure sum. Why Am I Offering This Service? Quite honestly, I enjoy the challenge of finding value in the betting markets. Football is not an easy sport to make money from. It’s hugely popular, and the statistics are studied by millions. Finding value is not easy, but I believe that I have found an edge with these selections, at least for the time being. Backing the draw is unfashionable, but that is exactly why there is value. No one watches a game cheering for the draw, so it is under bet. The supposed success of “Lay The Draw” as a system also works in our favour. Plus, in a match where the two teams are equally fancied, money comes in on both teams, yet while logic should mean that the draw price shortens, it rarely drops below 3.15 (one end of season match which was priced at 3.05 in France). By offering these selections, I am not hurting myself. As mentioned, the draw price is very robust, and because the ratings are mine, my bets will be first in the queue anyway! How To Subscribe Payments should be made via PayPal to CalcioCassini@aol.com – there is a Donate button on my blogs which can be used. Thank you for your interest, and please let me know if you have any questions.