Here are three key reasons why this newsletter scam works:
1. Selection bias - The newsletter only publicizes the correct predictions and hides all the incorrect ones. With hundreds of stocks, a few will inevitably move as predicted by chance.
2. Small sample size - Six predictions in a row could easily happen by chance given the large number of stocks. More predictions are needed to establish a valid pattern.
3. Regression to the mean - Extreme stock movements in one period will tend to be followed by a reversal in the next period just through normal fluctuation. This is exploited to create false predictions.
Rather than being impressed, one should be skeptical of any claims of predictive powers without transparent disclosure of all predictions made. Chance