If one of your value bet loses, it doesn't mean the bet doesn't have value. A punter must learn to accept that not all bets may be winning bets. The decisive factor is to discover value in your picks. Of course, the more the number of value bets, the larger the profit you can achieve.

Simply keeping a record of how each of the hundreds of tips we make actually perform against the eventual result just isn't enough, what we need now is a way of analysing that data and grouping it logically to get the best from it. Results are not absolutely the same, to put it differently a tip that shows one possible outcome for match A and the same possible outcome for match B will definitely not produce the same result (i.e. a correct prediction or a wrong prediction). Why is this? Well you will find hundreds of reasons why and also you will never be able to account for them all, if you could you would without doubt be a millionaire. When trying to predict the outcome of a match you might look-at such qualitative things as the present injury number of each team, the team sheet, morale of the players, etc. We can also look-at Quantitative factors using our statistical methods to predict the outcome of the match, so we may look at things such as past performance, position in the league, or maybe more tried and tested statistical methods for example the Rateform method. We can utilize all of these details to predict the outcome of match A as well as the outcome of match B but still not possess the same result, a component of the reason for this is, as explained before, that we can not account for all the factors in a match, it's impossible. But there is another thing, something we can account for which we have not yet thought about.

Once we look at one match in isolation we only look-at the factors concerning each of the 2 teams in the match, but why not expand this to look-at how the other teams they have played will also be performing? 'Why would we want to do that?' I hear several of you say. Because results are not always the same. Let's say our prediction for match A and match B is a home win (forgetting about the predicted score for the moment). What else can we think about to enhance the prediction of a home win? We can look at the performance of all the home win tips designed for the exact same competition that the match has been played in and after that make a judgement based on that new information. This really is great as it gives us an extra factoring level to take into account that we did not have before.

Looking across all the home win predictions in just one league will give us a portion success rate for home wins for that specific league, but we can improve on this even further. We can do this by doing the exact same exercise across a variety of leagues and obtaining a portion success rate for each league. Therefore we can now look for the league which produces the very best overall home win prediction success rate and look for home win predictions for the coming fixtures. By default we realize that that league is more more likely to produce a successful outcome for a home prediction than some other. Of-course we can employ this technique for away win and draw predictions also.

How Tight Is the League?:

Why does this distinction among the leagues occur? As with trying to predict the outcome of a single match there are plenty of factors that make up this phenomenon, but there are just a couple of major factors that influence why one league should produce more home wins by way of a season than another. The most obvious of these might be described as the 'tightness' of the league. What do I mean by 'tightness'? Within any league there is usually a gap within the skills and abilities of those teams consistently at the very best of the league and those at the bottom, this is often expressed as a 'difference in class'. This difference in class varies markedly between different leagues with some leagues being a lot more competitive than others due to a closer level of skills throughout the league, 'a tight league'. In the matter of a tight league the instances of drawn games will be more noticeable than with a 'not so tight league' and home wins will probably be of a lower frequency.

As a result, let's say we are considering predicting a home win, armed with our new details about the 'tightness' of leagues we could make predictions for matches within a season for as many leagues once we can manage, and watch how those predictions perform in each league. You shall find that the success of the predictions will closely match the 'tightness' of a particular league, so where a particular league produces more home wins then we shall have more success with our home predictions. Don't be misled, this doesn't mean that just because there are actually more home wins we have been bound to be more accurate, what I am taking about is a success rate in percentage terms of the range of home predictions made which has nothing directly to do with how many actual home wins you can find. For example, let's say we make one hundred home predictions in league A and one hundred in league B, and let's say that seventy five percent are correct in league A but only sixty percent in league B. We have made the exact same number of predictions in each league with differing results, and those difference are probably as a result of the 'tightness' of each league. League B will be a 'tight' league with more teams having similar levels of 'class', whereas league A has a wider margin of class in relation to the teams within it. Therefore we should pick out the most effective performing league concerning home wins and make our home win selections from that league.

We Have To Be Consistent:

Of course there is more to it than that. It's no good just taking each tip and recording how it performed we have to apply the same rules to each and every tip made. You will need to make certain that the parameters you set for each predictive method you use (e.g. Rateform, Score Prediction, etc.) remain constant. So choose your best settings for each method and stick to them for just about every prediction, for every league, and then for the entire season. You need to do this to be able to retain consistency of predictions within leagues, between leagues, as well as hop over to this site time. There's nothing stopping you using a number of different sets of parameters as long when you keep the data produced from each separate.