I briefly covered the NxALT error in an earlier essay on “AWALT” (all women are like that), but as it seems to be catching on in various domains relevant to, or sphere adjacent, it is time for a dedicated essay. Whenever I view characteristics of a population, I tend to make the initial assumption that it follows a normal distribution similar to the bell curve depicted in this essay.
Such a distribution is characterized by the fact that the values cluster around the mean, and the further away one gets from the mean, the smaller the population will be. For instance, in regards to IQ, 68% of the population are within 1 standard deviation either above or below the mean, meaning that they have an IQ in the range 85 – 115. 95% of the population are within 2 standard deviations either above or below the mean, meaning an IQ in the range 70 – 130. When one enters the outliers, meaning an IQ either below below 70 or above 130, this totals a mere 4.2% of the population. The extreme outliers, those people with an IQ either above 145 or below 55, are a mere 2% of the total population.
The normal distribution is present in many observations of human traits, height, weight and IQ being among them. In Gendernomics I argue that sexual market value should be viewed as a normal distribution, as this would be the distribution that ensured the maximal chance of “pairing off” when one takes hypergamy and the female pareto attraction into account. If all men are 10s, then it becomes impossible for hypergamy to select the highest value males, likewise if all women are 10s, then it becomes impossible for women to ensure that they have optimized hypergamy.
To summarize, in a normal distribution the majority of observations are within 1 – 2 standard deviations of the mean value, and the further one gets away from the mean the lower the amount of observations one makes. Thus it follows, that the probability of making an observation that is within 1 – 2 standard deviations of the mean is much higher than to observe an outlier.
Samples and Representation
One of the keys to make a normal distribution work is the sample sizes one is working with, once the number of observations is sufficiently large the central limit theorem under certain conditions state that the samples of random variables independently drawn from independent distributions follow the normal distribution. Thus, the total number of samples is very important to drawn inferences from the data. Census data is the gold standard for such inferences, because the sample size is the entire population. For instance, if one knows the height of every single member of a group, one can with certainty state the average height of said group.
However, in many cases where NxALT raises its ugly head, we are rarely dealing with gold standard research, where one has sampled an entire population, or multiple populations. In order to generalize from a sample of a population to that entire population, one must ensure that the sample is sufficiently large and representative of the population. Many election forecasters during the 2016 Presidential Election made the mistake of electing non-representative samples, for instance over-sampling people who lived in cities and under-sampling people in rural communities.
For instance, when it comes to the number of sexual partners an article in The Daily Mail  states that the average number of sexual partners for women were 5.81, the total number of female participants was 104, in the age range 18 – 35, with an average age of 21. Obviously, these women did not each sleep with exactly 5.81 men, some had notch counts above the mean, some below the mean. However, as the sample size was fairly small, the generalizability of the data to the general population is limited. Thus, the study in itself proves very little about anyone except those who responded to the survey, and one can never fully trust self-report surveys.
When one talks about many of the things one does in the manosphere, such as AWALT, Hypergamy and other behavioral observations that have been made about females, Beta males or other groups, one makes the observation of the behavior from a sample, and assumes that such behavior exists in the entire group, and follows a normal distribution. Thus, every member of the population has the capacity for a given behavioral trait, but the manifestation thereof varies. For instance, one could argue that one outlier is the woman who has never cheated on her partner, and one outlier is the woman who has cheated on 100% of previous partners. One outlier is the woman who stays committed to her husband for the duration of marriage, and one has had 7 husbands in 10 years, always trading up.
The argument, therefore is not “All members of this group have the exact same propensity to engage in behavior A”, but rather “Behavior A has been observed in a sufficient number of group members, that it is probable that it exists in all group members to some extent” and from this one can conclude that in interactions with members of this group, behavior A is likely to be observed again with some frequency.
NxALT in Practice
What constitutes a “NxALT fallacy” is the presentation of outliers as actually being representative of the mean, or alternatively outliers as examples of there being no mean. For instance, if one challenges a person with the statistic that 40 – 50% of couples divorce , one may face the argument “My grandparents were married for 70 years!” which doesn’t affect the probability of the statistic at all. It is merely the reporting of an outlier.
It is for all intents and purposes, the equivalent of responding to “The average American male is 5 ft, 9.5 inches tall and weighs 195.5 lbs  with the statement “My buddy Kevin is 6.5 and 260”. That’s great for Kevin and he is obviously an outlier as far as height and weight is concerned, but it doesn’t affect the means. Such an average is calculated the total value of a population is divided by that population, in order to find out the characteristics of the population. Thus, in a sample of two, Kevin at 6.5 and Dave at 5.4, their height is added up, then divided by two, resulting in the average height of 5.95.
If one is making a decision, for instance if I was running a company producing clothes for men, my best bet for maximizing revenue over cost would be to produce clothes within 1 – 2 standard deviations on either side of the mean, as this would allow me to sell to 95% of the population. When someone makes the argument that “There are people who use smaller and bigger sizes that you are missing out on selling to”, that is something I would be fully aware of, yet the cost of stocking those sizes, would have a negative effect on my overall profit.
Likewise, if a man is trying to decide whether to get married or not, he should base his decision on the 40 – 50% probability that the marriage will end in divorce, rather than the fact that his grandparents were happily married for 70 years. The former is a much larger sample. The tendency to ignore the mean, and instead prefer anecdotal examples as the guideline is linked to female solipsism, wherein a woman will usually neglect empirical data in favor of her own lived experience.
The practice of observing patterns, codifying patterns, then testing them through experimentation, is a foundation of grounded theory. This approach has been utilized within the manosphere for decades, and is an accepted academic method of research. Like it or not, people made decisions based on the patterns they observed well before they could dredge up citations  from academics from google.scholar to support their argument.
When Grog the caveman was wondering if he should head east or west to hunt, he likely observed the weather, used his knowledge of migration paths, previous hunting history to the east and the west, perhaps other factors, then he made his decision. For instance, he may have had a heuristic of “if the sun is shining, the animals are more thirsty so they will gather by the stream to the west, but if it is windy, they seek cover by the mountains to the east”.
This is somewhat where we reach the Mexican standoff between AxALT and NAxALT, in many cases no “Gold Standard” research exists that can give us perfect information about the distribution itself. In the article I used as my first source, the average number of sexual partners that the women had were 5.81, however a general heuristic is that women tend to leave out that drunken hookup, the guy she banged on vacation, the TA she banged to get a better grade, and perhaps reply with the number of boyfriends she’s had rather than the number of sexual partners.
This type of deviation tends to occur when people are ego invested in how they themselves or other people would perceive them if they gave an honest answer. Perhaps the best example of this (shout-out to my old research methodology professor) are the studies on penile length and the variation between self-reports and measurements taken by a medical professional . There is also the old observation that when asked only a minority think they are below average intelligence .
Summary and Conclusions
The challenge that is presented by AxALT and NAxALT, is that such statements are rarely statements of fact, but rather they are heuristics used in decision-making processes. Regardless of what internet debaters think, a great majority of decisions are made without consulting the latest academic research into the topic, and without engaging the mighty internet in a game of “let me find sources that support what I have decided to do/believe”. People are generally much better at rationalizing a decision after it has been made than they are at arriving at a conclusion through research.
This means that both “AxALT” and “NAxALT” become a Pascal’s wager of sorts. For those who are not familiar with this argument, Blaise Pascal made the argument that if you were an unbeliever, you had every incentive to act and think as a believer, as one had everything to gain, and nothing to lose by doing so. The reasoning being that if you act and think as a believer you will be welcomed into Paradise upon your death, if you do not you will go straight to hell.
Thus, the questions become:
A) If I adopt AxALT as my position what is the upside and the downside?
B) If I adopt “NAxALT” as my position, what is the upside and the downside?
A is clearly the most efficient route, as it permits decisions to be made with little investigation and investment. Furthermore, it shields one from idealism and naivete, as one requires a person or group to demonstrate through their actions that they are in fact outliers from your initial observation, or that your perception of the mean is an erroneous one. The vulnerability of A is that one may err in judgment in those cases where one assumes that a situation will reflect the mean, which is correct from a probability perspective, and the situation is in fact an outlier.
B, would be the more laborious option as it requires every decision to consist of data collection, data evaluation/analysis and then a decision to be made on a case by case basis. It also creates a situation in which your past observations cannot be adopted as a guideline for your present or future actions, as you must experience each case on its own. The major vulnerability of B however, is that the probability of error becomes much higher than in A. In A, one would have a 5 – 32% chance of being wrong, whereas in B one would have a 95% to 68% chance of being wrong.
For instance, a man man who adopts position A, who considers getting married for the second or third time, will evaluate the proposition as informed by his past experiences with divorce, and the consequences thereof. Whereas a man in the same situation who adopts position B, would not adopt his past observations as a guideline for his present decision-making, as “not all women are like my ex-wife“, and thus his previous experience is in no way relevant information for his future decisions.
The unstated premise of “NAxALT” is simply that it is preferable to deal with the consequences of treating someone as a maximally positive outlier, as opposed to the consequences of treating them as being average. The Soul-mate myth is a prime example of this, in that if one treats all women as queens, nobody is offended, if one treats all women as common wenches, everyone is offended.
I recently launched a Patreon page where I will be posting additional content every month for those who support me and I will do a Google Hangout for the highest tier Patrons (limited to 10 people).
I’ve also had some requests for consults, which I’ve declined up until now, but due to demand I’ve chosen to open up for doing some consults on request. For details please check out my Consulting and Patreon Page