To give the background for the experiment, I’ve never been much into online dating because when I started practicing game, the only material was largely centered on clubs. I’ve been using day-game quite a lot over the past few years because I travel a lot for my day job and often end up spending 5 days to two weeks in a location. However, lately I’ve found it a challenge to do day game due to spending less time in a city, having fewer repeat visits and having to compress the process into a much shorter time period.
I noticed that Anthony (@beachmuscles) was having a lot of success using various online platforms including Tinder, and concluded that this could be a very useful avenue for me as well and I don’t like leaving pussy on the table. I’ve also noticed a trend where people tend to go out more in groups, or on already planned dates, as opposed to going out just to meet people, so it seems that the internet has lead to distribution and logistics innovations in the sexual market place as well.
The first issue I ran into when I downloaded Tinder was to find good pictures of myself, I’ve had photography as a hobby for years but I’ve never spent much time in front of a camera, and I don’t really have a lot of pictures of myself suitable for a dating profile. I also know that people tend to be very poor judges of their own pictures, so I figured that I had to find a way to get 3 – 5 good pictures of myself.
This was what caused me to conduct this experiment because in absence of other pictures, I used my professional headshots for my Tinder profile along with filling out the default spaces in the profile including my alma matter and my job title, then overwhelmingly got responses from women ages 30 – 35 with professional jobs and University degrees. This is hardly my ideal demographic, so being that I know a thing or two about red pill theory I formulated an experiment. It was only intended to give me insight into online dating in general, and Tinder in particular, but when I mentioned it on The Red Man Group many men requested a write up, so here we go.
Tinder was initially released in September of 2012 and has an estimated 50 million users on a global basis. The app has gained a reputation as a hook-up app, mostly owing to the location based nature of the service, and the fact that it is largely based on first impressions. Services that promise to improve your success rate on Tinder frequently mention having great pictures as a critical success factor on the platform. The range of recommendations for pictures vary a little bit between different sources OKcupid  and Photofeeler  have different perspectives, where one recommends smiling and eye contact in pictures, the other recommends the opposite. There are also different recommendations from a varied range of sources [3, 5, 6] that include number of pictures, context of pictures and various other factors as having an influence on matching rates.
In addition the “Smart Pictures” feature of Tinder constantly crunches the numbers on which picture people swipe right on the most to determine the picture that appears first on your profile. There has been some criticism of this feature, for the reason that a sub-set of users will look through all the pictures before swiping right, however many users also swipe on the first picture. Thus, if your last picture is your worst, but is preceded by 4 stellar pictures, your worst picture may appear first.
This experiment sought to test these recommendation in an exploratory fashion to see what effects these recommendations have on number of right swipes. The experiment was conducted using the same Tinder profile, with different pictures and different standard information added to the app. An additional variable beyond determining what profile pictures to utilize, was to test the effect on number of matches and the demographics of matches for each profile.
In Red Pill theory the female strategy of Alpha Fucks/Beta Bucks, is the idea that women classify men according to two categories, and act against these men based on how they are categorized, the innate behaviors and tactics that support and permit women to enact this strategy are collectively known as hypergamy . The most classical manifestation of this is that a woman will sleep with a man she perceives as Alpha much sooner and with much less investment than she would with a man whom she classifies as beta. The case of the Alpha is one of genuine desire, the case of the beta is one of negotiated desire .
A woman’s inclination in either direction (Alpha Fucks/Beta Bucks) is dependent on her stage in the sexual market place , where the “Party Years” are characterized by a focus largely on the Alpha Fucks side of the dynamic, whereas the subsequent phases are characterized by an increased focus on securing her long-term security and thus by a focus on Beta bucks. This is largely driven by the realization that she is no longer capable of competing with younger women in the sexual market place. It is important to mention that the strategic choice of a woman between alpha fucks and beta bucks is not a binary one, meaning that she selects one or the other, but rather exists on a scale between the two. Furthermore, it can be exacerbated by what stage of her menstrual cycle she is currently in  and by other psychological factors.
There is a fairly large literature base of materials relating to “Alpha tells” and “Beta tells” meaning behaviors that demonstrate either a woman’s responses, behaviors and actions around a man she perceives as alpha, or conversely a man’s behavior that signals himself as beta . Chateau Heartiste has a long running series dedicated to spotting such tells in pictures . For simplicity, the reductionist approach when designing the pictures and details in the profile focused on long-term provisioning traits (financial status, professional success, outward trappings of wealth, educational status) vs. appearance/protector traits.
The foundation of the theoretical framework for the experiment is based on the sexual market graph , which divides the life-cycle of men and women in the sexual market place. The graph lists 18 – 23 as a woman’s rise to her sexual market peak (22 – 23) and, her then decline as she nears the epiphany phase (24 – 30) . The epiphany phase usually takes place between age 28 and age 32, and this is where a woman’s focus shifts from securing short-term mating to long-term provisioning. This is also where most first marriages take place , and where the sexual market value of the declining woman intersects with the rising sexual market value of a man at a comparable age.
The exploratory hypothesis is that a profile that signals “long-term provisioning” such as listing a professional occupation, college degree, trappings of wealth (clothing), higher intelligence and higher degree of trustworthiness will elicit more matches from women who are in the epiphany phase or nearing “the wall”. Conversely that a profile listing very few traits that would indicate long-term provisioning capabilities, will elicit fewer matches from women who are in the epiphany phase or nearing “the wall”.
Furthermore, a secondary hypothesis is that the latter profile will elicit a higher number of matches from women who are in within 2-3 years of their sexual market peak.
The experiment was designed as exploratory research conducted on a small sample and with a limited scope.
The variables that can be observed are as follows:
- Number of profile likes (right swipes)
- Demographic data of those who liked the profile
Important variables that cannot be observed:
- Number of profile displays (number of people who saw the profile)
- Demographic data of those who did not like the profile
The experiment is designed to determine the effect of profile photos on the number of likes and the effect of the demographics of those women who liked the profile. For this purpose, a total of 20 pictures were taken, 10 using a plain background, and 10 using a background outdoors. These pictures were classified into 2 unique photo sets and 2 control sets, with no picture featuring in multiple sets. Control pictures were used to establish a baseline for each location. A total of 4 tests were ran for each location, 2 control tests and 2 regular tests.
Each photo set contains 3 types of picture:
A) Headshot with smile and eye contact taken against a white background
B) Full body picture in clothing taken outdoors
C) Half-body picture taken in clothing utilizing props (Taken in office setting wearing a suit for provisioning profile, bar with leather jacket for non-provisioning profile)
For the purpose of establishing a baseline, 2 further tests were conducted featuring only a single photograph taken against the same background:
A) Headshot without smile and eye contact (Control 1)
B) Headshot with smile and eye contact (Control 2)
It was decided to not employ a written bio for any of the tests to avoid this variable affecting the experiment.
All the pictures were tested on photofeeler running tests of 80 votes with the demographic range of women ages 19 – 34. All the selected pictures had an attractiveness rating above 70%.
2 profiles were created to conduct the experiment:
In this profile the pictures used were those that scored above 70% for attractiveness, but lower for trustworthiness and intelligence. This is based on the premise that intelligence and trustworthiness are secondary factors for short-term mating in women. Body language adopted for the pictures was more aggressive and the camera angles took care to shoot either straight on subjective or downwards and up. In addition a minimum of 25% of the raters had to comment that the person in the pictures appeared arrogant or aggressive. The clothing wore in the pictures took care not to display trappings of wealth or success, and consisted of jeans, sweaters, leather jackets, and plain white t-shirts.
Furthermore, the alpha fucks profile listed no alma matter, education, and no job title.
In this profile, the pictures used were those that scored above 70% for attractiveness, but also had scores of above 70% for both trustworthiness and intelligence. This is based on the premise that intelligence and trustworthiness are important factors in long-term mate choice in women. In addition less aggressive camera angles and posture was adopted for the beta bucks profile pictures, including shooting at a downwards angle towards the subject, adopting less aggressive and confident body language. The clothing wore in the pictures consisted of polo shirts, suit, tie and corporate appropriate button up shirts.
Furthermore, the profile listed a well-known University as the alma matter and a professional job title towards the higher end of the corporate payscale in order to further elevate the perception of “Provider”.
The experiment was conducted as follows. 2 consecutive boosts were ran in Tinder’s “golden hour” from 8:30 PM to 9:30 PM on Sunday evening  the consensus on when to use a boost is generally between 7 PM and 9 PM on Sunday, however some argue that Thursday during the same time interval is better. For the purpose of maintaining control over this variable, for 4 consecutive Sundays, 2 boosts were ran, the first one starting at 8:30 PM and running until 9 PM, the second running from 9 PM to 9:30 PM.
Two locations were selected for the boost, one in Eastern Europe one in the United States of America. Both cities have populations above 2 million but not above 5 million. This was a choice criteria to ensure that Tinder would have a reasonably high volume of users in each city. Furthermore, it was done to gain some insight into whether cultural or social norms had an effect on the experimental outcomes. The demographics of each city was checked to ensure that they had roughly the same distribution of age ranges among women.
The experimental results are based on a very small sample over a very short time period (1 hour total run time of Tinder Boost), thus while the results are interesting and appear to confirm the theory, little certainty can be attributed to the results without a larger sample size, longer running time, and ideally with insights into how many times each profile was viewed. Thus, the results must be viewed as unreliable.
The total number of matches between each of the two profiles was fairly minimal in both locations, however the two profiles with more than 1 picture performed significantly better than the controls in both locations. The non-provider profile outperformed the provider profile in both locations, however due to the small difference and the fact that Tinder is reputed to be an app for short-term liaisons more than long term relationships this may be the cause of the difference rather a female preference that can be extrapolated. One of the more surprising results is that it appears that Smiling vs. Non-smiling and Eye contact vs. No eye contact has a small effect, but that this is at best negligible.
A larger difference was seen in the average age of the matches, where the average age of the women who liked the Beta Bucks profile were on average 5 years older than the women who liked the Alpha fucks profile in the US, and 3 years older in Eastern Europe. The control profiles showed that not smiling and not having direct eye contact appeared to result in an average lower age than a profile photo with eye contact and smiling.
The occupations of the matches also differed quite a bit based on the profile. It appears that professional women, especially those in the older demographic are more likely to like a profile that indicates a higher degree of provisioning than younger women. This does appear to be a continuum rather than binary, but does indicate that a profile that demonstrates more provision characteristics does attract an older and more professionally established demographic.
Keeping in mind that the following experiment was exploratory in nature and thus offers little in terms of data that can be generalized, the results are interesting, however it remains to be seen if they hold up in general or if they were specific to the one subject of this experiment. Without access to total number of profile views, it is difficult to say anything about total effect, especially as the difference between total number of matches differed very little. A small difference in number of active women on Tinder during the running time for the test could easily explain the difference in matches. However, the demographic data is more difficult to explain.
Furthermore, smiling vs. not smiling and eye contact vs. no eye contact appeared to have little effect on the number of total profile likes. This is roughly as one would expect based on the contradictory data from OKcupid and Photofeeler . One must keep in mind that while all the pictures that were used had high attractiveness scores when tested in the demographic range 19 – 34, there is some uncertainty about how women rate male attractiveness, furthermore, one cannot control for important variables in female mate choice. Previous essays have explored how women rate male appearance, and an assumption has to be made that the same pattern follows on all rating services.
The author’s major reaction to the results of this test is that it does appear to correlate with what Red Pill theory would predict, however at very low sample sizes, over a limited time and with limited testing, it is hard to make any definite conclusions from this data. It was however an interesting experiment to conduct. The results also correlate with previous data analysis of the dating marketing conducted on Black Label Logic, including Book Value and Macro and Micro.
Based on cursory observation, the author does not that the prop use in the different profiles does appear to influence the demographics somewhat, thus one would be inclined to argue that selecting the context of the pictures can have a useful effect. Overall, I would recommend using a service to have your pictures rated, as the results of the tests vs. the author’s expectation were often in stark contrast to one-another.