The law of large numbers !
often people rave over "systems" or techniques that have short term success. The fact is that we should look a lot further than that. Let's say I invented a "system" where we take the numbers of the last draw and multiply the odd numbers by 2 and divide the even by 2 to obtain the numbers to be drawn next. Or a system where I paint numbers on a wall, and throw darts at the numbers blindfolded in order to know what will be drawn next. (I've seen people enthousiastic about weirder systems) Obviously these systems are ridiculous, and yet, when you start to test them on certain occasions they may even produce you multiple 4 of 6 wins in a row. Does that mean they are valid ??? NO absolutely not, the initial success will very quickly turn to disaster. If you isolate a small number of independent trials with equal probability of success, one may see the balance of achieved success versus expected success (according to the probability) turn in your favor for a short while simply because of an inequal distribution of hits. (Large deviation)
example: let's say the probability of success is 1 in 10
we can expect the result of 30 independent trials to be like this where every zero stands for failure and 1 for success.
000000100000000001000000000100
Or you can get this :
000100110000001000010100000100
Often people conclude from such a situation that the system that produced the second example line is better than the system that produced the first example line. This is not true because the probability that the second example line will be followed by a result like this increases.
0000000000000000000000000100000000000000000000000000000000010000000000000000000
The law of large numbers says that we may see far more success than expected in a small isolated amount of trials, while dventually the larger the amount of trials becomes, the more the amount of success will tend toward the mathematical probability ! If one would do an infinite number of trials, the success should theoritically become exactly the probability. The deviation becomes smaller and smaller when the amount of trials rises. Eventually the deviation will become aprox. zero.
From this we must conclude that an isolated success (even when it is a 5 out of 6 hit) does not mean anything at all. The only valid barometer for the performance of a system is when we measure the amount of success in an amount of trials of at least 10 times the probability of success. If the probability is 1 in 10, we should at least do 100 trials before jumping to conclusions. The larger the amount of trials, the more reliable the results.
Derived from this we should conclude that a system capable of turning for example a probability of 1 in 9,915 (=the probability of hitting at least 3 numbers from a group of 9 in a 6/42 lotto) to 1 in 5 over at least 100 trials is by far more reliable than a system that produces three 4 of 6 hits in 10 trials.
The illusion of combinatorial systems:
The above conclusion only goes for systems predicting number groups, not for combinatorial systems that use all numbers from your lottery. It is fairly easy to create a combinatorial system that produces regular 3 out of 6 hits by using all numbers in your lotto. But these systems cannot improve the odds of winning the bigger prizes nor can they ever result in consistent money making. One simple rule proves this. If you cannot predict numbers, then every line of 6 numbers played from a combinatorial system will have an equal chance of winning the jackpot as a randomly generated 6 number line. Trying to increase the odds for matching 5 or 6 numbers with combinatorial so called tricks will inevitably lead to the need for playing disgustingly large amounts of combinations. Just like playing more random generated 6 number combinations increases your chances. The benefit is always zero with these systems, nomatter what you try! And yet, there are so many systems available on the net that use this misleading illusion !!!!! A few examples of such systems are lottotrix, honest lotto system, delta lotto system,... They are all based on errorous and misleading beliefs.
Bottom line is : Don't jump to conclusions too fast when you have a streak of wins or losses. Only judge over large amounts of trials not single impressive hits from others. A system capable of doubling the odds to match 3 numbers consistently in a large amount of trials will certainly increase the odds for winning bigger prizes. This proves that such system is build on a solid basis. While systems that promote isolated big hits mostly prove nothing at all !!