Basing how likely you are at succeeding at something on statistics(let's call it the frequentist worldview) is part of one of the worst types of ideas, it's the type of idea that is very wrong and has very bad consequences but simultaneously sounds reasonable and scientific.
The best argument against it is pure observation, for example, if we view a company succeeding as pure luck, then the people at the top of these companies should be in general proportional to the general population(or the upper class), but it isn't, these people have features that are common in their groups, but uncommon in the general population(or even the upper class), a good chunk of them is some kind of immigrant(WASPs founders are a dying breed), a good chunk of them has a weird worldview, etc.
The problem with the frequentist worldview is human diversity, having or not having some characteristic significantly alters the probability of the event measured, me having a computer and knowing english significantly altered the chances of i writing this, but this isn't measured by frequentism(bayesianism is kinda what i'm describing here, but in the real world, good luck measuring every little thing accurately, what are the chances of me having a computer? Unless you're a frequentist, measuring this isn't straightfoward, so we need more bayesianism, and this spirals out into a incoherent mess).