**GBS PROJECT OVERVIEW**

The bookmaker’s overround is the single most important disadvantage the sports bettor will face and **any successful betting system must be able to overcome this margin through the principles of value betting and a methodical staking strategy**, with a view to preserving the bankroll and maximizing the profits in the long run. Kelly staking, combining both fixed profits and percentage bank approaches, maximizes the growth of a betting bank whilst minimizing the risks. An effective betting system should, however, be able to estimate its advantage fairly accurately in order to benefit from this strategy. It’s our belief that our recent astrological studies on midpoints when related to the starting time of a sports event can provide superior edges when focused on under/over bets.

**PURPOSE**

The vast majority of betting systems identify value bets through a set of ratings of a match, to be then translated into a probability distribution for the possible results. A reliable rating is usually the match form of the two teams in terms of scoring and conceding goals. Once identified fair odds for a match, comparing them to those of the bookmakers finally determine a value bet. Where bookmaker’s odds are superior to the fair odds provided by an accurate rating system, we have potentially gained an edge, meaning that the bookmaker has underestimated the probability of the result and is offering higher than the mathematically fair odds. **Our purpose is to add to this set of ratings a peculiar and innovative forecast analysis based on our recent astrological studies, giving a clear advantage over any other betting systems**. A superior edge translates automatically in a superior yield, an enhanced ratio of profit to the total amount wagered. Our aim is to provide a prosperous betting system that allows a punter to recycle profits already won into future wages, ensuring that much of the initial betting capital is not actually used. We believe that our betting system may easily fulfill the common and ambitious target of doubling a bankroll during the course of a season, using stake sizes in line with a proper risk assessment.

**The Betting Bank**

A betting bank or bankroll can be described as an amount of capital that has been set aside exclusively for the purpose of betting. Every time a bet is placed, a small proportion of the bankroll is staked. The smaller the stake as a proportion of the total bankroll, the less significant the impact after either a win or a loss. In terms of risk management, smaller stakes involve less risk. In terms of growing the bankroll, smaller stakes will naturally contribute smaller profits, and growth, when a betting edge is accessible, will take longer. Traditionally, the size and growth of the bankroll has been viewed as the primary measure of betting success. In fact, two key pieces of information are required: the yield, or profit over turnover, and the betting rate, or number of bets over a specified time. Betting yields cannot be fixed in advance as they fluctuate continuously during intermittent periods of success but they provide an objective and comparative measure of the strength of a betting system. The number of wagers required to double a bankroll of 100 points (the magnitude of a point could be $1 or $1,000) for a series of betting yield-stake-size relationship is shown here:

Source: Fixed Odds Sports Betting: Statistical Forecasting and Risk Management by Joseph Buchdahl. Summersdale Publishers Limited, 2003

**Early Test of the Over/Under Geocosmic Betting System (O/U GBS)**

For a value betting system, the betting yield provides a meaningful indicator of success, because it can be related directly to the size of its edge over the bookmaker’s odds. The yield from a succession of bets already settled does provide a good idea of the average advantage a betting system has secured. The O/U Geocosmic Betting System (GBS) has been tested on 26(N) Serie A and Bundesliga games from May 20^{th} to August 1^{st}, returning (F) 107.86 points from a starting bankroll (B) of 100, with an average of 1.24% stake size(S) and for an yield(Y) of 30.23%, meaning that it has, on average, gained a +30% edge over the bookmaker for each of these bets. The following equation **F=B+(YSN/100) **describing the simple relationship between betting yields(Y), bet number(N), stake size(S) and starting bankroll(B), where (F) denotes the final bankroll after the betting sequence, suggests that 267 level stakes of 1.24 points each are needed to double a bankroll from 100 to 200 points when the betting system has an average betting yield of +30% for each bet, as performed by this test whose detailed succession of bets is illustrated in the next table.

**RISK FACTORS**

Sports betting is about managing probabilities, probability is about chance and chance is about risk. Betting, like other forms of financial investment, amounts to speculating on the future with a view to making a profit. No speculation is completely free of risk. In under/over bets, a losing fixed odd bet involves a full loss of stake. A betting system spreads the risk across a series of bets, with each stake a small proportion of the total funds at its disposal. The larger the size of the stakes as a proportion of the bankroll, the greater the chance of a total loss of initial betting resources, if things go wrong. Kelly staking maximizes the growth of a bankroll whilst minimizing the chance of failure.

**KELLY STAKING**

The Kelly staking is a money management plan that combines both fixed profits and percentage bank approaches. By taking into account the expected rate of return and the risk, the Kelly utility function provides an economically justified and mathematically precise way to compute optimal bet sizes the maximize the overall growth of a bankroll. The mathematics of the Kelly Criterion are rather complex. For sports betting, however, it is important only to know how to calculate the Kelly stake size. This is given by the following relationship:

**K = (E-1)/(O-1)**

where K is the size of the Kelly stake as a decimal percentage of the bankroll, E is the decimal edge as presented throughout the next sections of this project and O denotes the decimal odds. With the Kelly Criterion, one is always betting percentages of the bankroll, so as the bankroll grows, so does the size of the stakes. Likewise, when the bankroll shrinks, the stake size will shrink too. Next table shows some examples of Kelly stakes. There are inherent dangers with Kelly staking, where the advantage or edge is overestimated by the betting system, consequently wagering more than ought. Maximizing the growth of the bankroll should be the most important objective and safety the greatest concern, so a fractional Kelly approach may be used instead. O/U GBS will bet one-third of the suggested Kelly stakes, of course slowing down the expected growth of the bankroll but with the benefit of reducing dramatically the probability of bankruptcy. Kelly staking is also highly dependent on odds: shorter odds will attract larger stakes, longer odds smaller bets, with chances of not returning a profit decreasing for shorter odds, despite stakes being proportionally larger.

Source: Fixed Odds Sports Betting: Statistical Forecasting and Risk Management by Joseph Buchdahl. Summersdale Publishers Limited, 2003

**DETERMINING THE EDGE – STEP ONE – VALUE BETS**

Betting odds can be seen as an aggregated expert opinion reflecting both the judgment of bookmakers and the betting behavior of bettors. Recent studies have shown that in general betting odds possess an excellent predictive quality and perform better in forecasting soccer results than various quantitative models. Also, betting odds are routinely used as a quality benchmark for testing the predictive quality of mathematical models, while it would be advisable mixing the power of both approaches to create new forecasting possibilities. So far, hardly any study has tried to revert the forecasting process using existing forecasts (from betting odds) to draw conclusions about the qualities of the teams, obtain team ratings and thus contribute to the performance analysis. That’s why GBS O/U will focus preferably on Under/Over bets with shorter odds, backing this way mainly the favorite outcome as indicated by bookmaker’s odds. Furthermore, bookmaker’s profit margin (ovverround) is not proportionally spread across the range of possible outcomes, but is concentrated on the higher odds. With superior value in the shorter prices, the potential for gaining an edge, if not additional profit, may very well be greater, since the disadvantage to be overcome is smaller.

Source: Fixed Odds Sports Betting: Statistical Forecasting and Risk Management by Joseph Buchdahl. Summersdale Publishers Limited, 2003

Betting odds are widely used to derive forecasts as they are simply transferrable to probabilities and have proven their quality in a large number of different studies. If no bookmaker margin was contained in the betting odds, the inverse betting odds for any possible outcome of a match could be interpreted as its probability of occurring. To eliminate the bookmaker margin from the odds, i.e. ensure that the derived probabilities sum up to 100%, we applied the most widely used approach of basic normalization. For instance, on under/over bets where both have equal chances, a bookmaker would probably go 1.83 and 1.83 rather than 2.0 and 2.0. That equates to a probability of 54.65% (100/odds) for each selection, and added together the book is 109.3%. This additional percentage is the overround. Then, the actual probability can be calculated as **p=1/(odds*overround)** resulting in probabilities for each outcome summing up to 100%. Total goals betting in soccer, where there are usually only 2 possible outcomes (over 2.5 goals or under 2.5 goals), attracts overrounds that are commonly less than 110%.

Once determined the bookmaker’s actual estimation of each selection, we need to compare it to the fair probability as suggested by our preferred set of ratings of a match, for the purpose of identifying value bets. Forebet, as an example, is a system for predicting soccer matches on the basis of statistical analysis, where for each possible outcome a probability is calculated by a complex mathematical algorithm working on the football Big Data. The process is quite large and various data mining techniques are used, such as statistical modeling, machine learning, data mining that analyze current and historical data. Next table shows Forebet’s assessment of fair under/over probabilities for the Serie A match Juventus FC – AS Roma.

The average odds for this match were 1.52 (Over) and 2.48 (Under) as reported by BetExplorer. So, the bookmaker’s estimation for an Over outcome, on the basis of our previous considerations, was 62% while the fair probability, according to the Forebet analysis, was 69%, giving a ratings edge of **7.0%**. This represents our first edge factor, to be added successively to the proper forecasting factors determined by two astrological studies, as outlined in the next sections of this project.

**DETERMINING THE EDGE – STEP TWO – MIDPOINTS**

An experiment was made attempting to demonstrate any statistical relevance in the coincidence of the planetary aspects defined astrological midpoints (according to the terminology of their theorist Reinhold Ebertin) with the total number of goals scored in each game of Premier League and Italian Serie A in the seasons from 2011-12 to 2015-16. Table 2 published in our ebook Bet and win on soccer with midpoints: A breakthrough in sports astrology and football predictions shows the percentage differences (greater than 0.6%) obtained from all the midpoint combinations (with an individual point ASC or MC as focal point), ordered starting from the highest ‘negative’ value on the left side (which highlights the combinations of midpoints that appear most often in the games with 0-2 goals), and the highest ‘positive’ value on the right side (which therefore highlights the midpoint combinations that appear most often in games with more than 2 goals). Table 2b shows the midpoint combinations with both the individual points as factors.

Let’s take, e.g., the aforementioned Italian Serie A match Juventus FC – AS Roma that took place on August 1, 2020, in Torino at 8.45 PM. The relevant midpoints (involving ASC and/or MC as focal points with an orbital degree of 1°50’) at the start time + 2 minutes (8.47 PM) of this game were the following:

ASC | S.Square | MOO/MC | 1 | degree | -0,94 | |

ASC | Square | NEP/NOD | 0,8 | degrees | 0,58 | |

MC | Square | MOO/MAR | 0,5 | degrees | -0,90 | |

MC | Opposite | VEN/URA | 0,4 | degrees | 2,02 | |

MC | Ses.Square | VEN/ASC | 0,6 | degrees | 0,56 | |

MC | Ses.Square | NOD/ASC | 1,3 | degrees | 0,79 | |

2,10 % | Total |

Total shows the percent deviation of the summed up detected combinations. It also indicates that the astral balance tips towards an Over outcome, giving a midpoints edge of **2.10%**. This represents our second edge factor.

**DETERMINING THE EDGE – STEP THREE – PUTTING IT ALL TOGETHER**

To sum it up: the ultimate betting edge for each selection is then determined by the ratings edge together with the forecasting edge based on our astrological studies. How to merge these factors is a matter of risk management. Our inclination is for a cautious attitude, since the preservation of the betting bank is paramount for building a steady and consistent income stream. In light of that, our suggested solution is one-third of an arithmetic average of the previously measured factors, giving a more conservative **E** value for our Kelly staking formula: **K = (E-1)/(O-1)**.

Following up our example, the stake (K) of the selected match Juventus FC – AS Roma would have been **0.97%** of the bankroll, when the edge (E) is determined by one-third of an arithmetic average of the edge factors (((7.00+2.10)/2)/3)=1.516) giving a decimal edge of 1.01516, and when the odds (O) for an Over outcome are rated as 1.52 by bookmakers. So K=(1.01516-1)/(1.52-1)=0.02916 or 2.916%. Since our conservative approach is for a fractional Kelly, as outlined before, O/U GBS would bet one-third of that K value, that is **0.97%** of the bankroll.

**PICKING BETTING SELECTIONS**

In line with a general conservative attitude, combining a sound risk management with an investment idea of betting, selections should be picked when all boxes are properly ticked, that is when a value bet that has been identified through a fair ratings estimation, as outlined in a previous section of this plan, is then confirmed by the edge provided by our forecasting method based on astrological studies. This approach is highly recommended in a view to maximize the growth of a bankroll whilst minimizing the chance of a partial loss of initial betting resources.

**CONCLUSIONS**

It’s our belief that GBS betting system, due to its fundamental flexibility, may be easily adapted to any sports competitions that envisage Over/Under betting selections, such as Us sports, provided that each sports competition is properly studied through an astrological analysis of midpoints.

**REFERENCES**

*Fixed Odds Sports Betting: Statistical Forecasting and Risk Management*by Joseph Buchdahl. Summersdale Publishers Limited, 2003*The Betting Odds Rating System: Using soccer forecasts to forecast soccer*by Wunderlich, Memmert. PLoS ONE, 2018*Bet and win on soccer with midpoints: A breakthrough in sports astrology and football predictions*by Massimo Moras**.**Amazon Media EU, 2020