ELI J. FINKEL AND BENJAMIN R. KARNEY are wrong at least when it comes to POF

After reading yet another story about how online dating can not predict who will have a romantic relationship I thought I’d say something.      Its clear from reading the paper and these posts that Eli and his team do not know what POF  is actually doing when it comes to matching.   What we are doing is completely different than everyone else.

1. Similarity and other Psychological constructs  are not used in our matching system.   This is something a Psychologicist would do if they were responsible for making a matching system.   In my opinion   Psychological approaches to matchmaking are next to useless,   they offer relatively small improvements over chance from all the testing we have done.    We’ve hired the best Psychologicist’s to try and create a matching system and than validate it using our  couples data.   In the end Psychological evaluations and matching turned out to be completely useless.    So we agree with Eli and team there.

Back in 2007 I went on the today show and revealed that I created a behavioral matchmaking system that helped match people up.  Almost a year later Match.com hired a bunch of people to try and copy what i was doing and that seems to be their system today.   This is a great system for getting people to send messages and in match’s case optimize revenues  but it is a poor system  for predicting relationships.

Given that   behavioral matchmaking and recommendation systems were not powerful enough I started getting people who left the site to tell me who they entered into a relationship with.      Using MILLIONS of people who left the site in a relationship we could  very quickly determine what kinds of people got together in a relationship and with who.      After about 6 months 45% of the relationships split up giving us a very clear picture of what kinds of traits work in relationships and which ones don’t.   Keep in mind we make people answer hundreds of personality questions and we have all the data they entered on their profiles along with other stuff we collect.  Using simple predictive models we are able to tell  who you are most likely going to start dating with a very high degree of accuracy.   If you do enter into a relationship with someone and it is a unstable relationship we could probably tell you the week in which your relationship will end.

Given what i’ve stated above  there is no way that Eli  and his team can argue that  “a mathematical formula can not identify pairs of singles who are especially likely to have a successful romantic relationship”     Predictive modeling works in every other industry imaginable Online dating is no different.   In short sites Like Eharmony, PerfectMatch and Chemistry are designed by Psychologists who think the world should work in a certain way and they build a matching system that enforces that worldview.  At  POF  we just find successful couples and lets them be the basis for the matching system.

The following graphs the probability of a relationship breaking up.      In order to be better than chance at predicting long term relationships all we have to is make sure we do not generate matches that lead break ups.  Last year we could predict which relationships will fail with a greater than 80% accuracy.    The relationships shown here are not the result of a matching system but relationships that randomly happened on POF.   The sample size is around 21,000 and skewed to heavily to under 35.  The way to read the graph is  42% of relationships formed on the week of 2/20/2011 are still in tact on Feb 13th 2012.  That is to say neither the male or female came back to the site to signup for a new account.

27 Responses to “ELI J. FINKEL AND BENJAMIN R. KARNEY are wrong at least when it comes to POF”

  1. Online Dating « Make A Public Commitment Says:

    […] ELI J. FINKEL AND BENJAMIN R. KARNEY are wrong at least when it comes to POF […]

  2. james Says:

    That’s Interesting. It sounds like you are using a neural network for matching. It would seem to be hard to prevent over training without a feedback loop. I guess the result of over training is not enough matches and under training your network results in poor matches.

    How much compute power does this algorithm take? Cool Stuff!

  3. Dating Industry Under Fire For Dubious Matchmaking Claims | Online Dating Insider Says:

    […] I was reading it, Markus Frind, founder of Plentyoffish (recently rebranded as POF), sent me ELI J. FINKEL AND BENJAMIN R. KARNEY are wrong at least when it comes to POF. Markus was responding to a research report by Eli Finkel, Associate Professor of Social […]

  4. Kas Thomas Says:

    I still think more transparency is needed in how the various dating sites “match” people and what the science is behind their claims. Author K.C. Kimball does a good job of summarizing some of the science in his latest e-book (http://www.amazon.com/Alpha-Males-Online-Dating-ebook/dp/B0077T7UAA). This type of information needs to become better-known. Right now it’s hard to separate the voodoo science from the real science. At least Kimball cites references that anyone can go look up.

  5. Marc Lesnick Says:

    Thanks Markus, but I need to better understand that chart.

    42% of couples that met on Jan 20 did not login the following week, which suggests the relationship is still in tact?

    And the same group….21,000 people, by Dec 2011 – 90% of the relationships from January of the same year are still intact.

    If it is from week to week, then how does that help? Relationships end at random times.

  6. Markus Says:

    Its % of relationships started in that week are still in tact on Feb 13th 2012. The sum of all points on x access is 21,000 people.

    From what i’m hearing POF is the only dating site in the industry measuring outcomes of real couples.

    • alicia Says:

      i just left a comment this morning and its deleted again?
      im sorry markus but i will leave a comment on every blog you post until you answer me!



  7. Dating Industry Under Fire For Dubious Matchmaking Claims Says:

    […] I was reading it, Markus Frind, founder of Plentyoffish (recently rebranded as POF), sent me ELI J. FINKEL AND BENJAMIN R. KARNEY are wrong at least when it comes to POF. Markus was responding to a research report by Eli Finkel, Associate Professor of Social […]

  8. Andrea Fuentes Says:

    Interesting article; glad to see you are successful.

    I’m starting a social cooking website (not dating) and am glad to hear about the success of mobile.

    Hey, a friendly tip, you might want to double check spelling of psychologist; looks like a typo in a couple of places.

  9. Barb Elgin Says:

    Markus – Am I understanding you correctly when you say your method is to count those men or women who haven’t logged back in to POF as still in a relationship?

  10. Markus Says:

    No, they are users who have TOLD us they are in a relationship. Loggins doesn’t mean a thing.

    • alicia Says:

      why are my comments on your blogs being deleted and why am i paying for upgrading membership CHARGED but then deleted FOR NOTHING> i believe this is a form of fraud to be charged for services YOU CANNOT USE. i will file a REAL LAWSUIT AND CRIMINAL CHARGES AGAINST YOUR SITE IF I DO NOT GET ANSWERS/

  11. Pam Says:

    “The way to read the graph is 42% of relationships formed on the week of 2/20/2011 are still in tact on Feb 13th 2012. That is to say neither the male or female came back to the site to signup for a new account.”

    I think either your logic is way off or you are not expressing yourself very well, because you went on to answer a comment with this “No, they are users who have TOLD us they are in a relationship. Loggins doesn’t mean a thing.”

    For instance you quote 42% of relationships are still in tact…going on what exactly? since you then said because neither the male or female came back to the site for sign up for a new account…..

    well, errr that could mean that many of those 42% actually were no longer in a relationship AND decided not to sign up to POF, and perhaps used another dating site, or none at all.

    Then you say they TOLD you….really? how accurate is that information? because I simply cannot imagine that ALL 42% of those people would be willing to keep in touch with you, or answer your ongoing questions about their relationship success or failure.

  12. tommy Says:

    have any questions about fish? go to tommyjohnfish.webs.com “we answer all your questions”

  13. Randall Bell Says:

    “Psychologicists” don’t exist.

    Maybe before you publicly trash an entire body of research, you should learn some of the basics – like how to spell their field?

  14. himagain Says:

    In a country famed for arranged marriages, India, where it is an obsession to create a good match, I was disappointed many years ago to find how many marriages were failures. Seems like the international figures are similar even for the dating games and POF approach.
    I am entirely fascinated with the future of psychopolitics (created by the USSSR in the 60’s ) with the data piles of people like Markus, much less Google et al.

    As an old dog learning new tricks, I’ve been fascinated to learn that my assumption, as a psychologist, that virtually all people basically lie to either to gain an advantage or to avoid punishment (or ostracism) appears quite wrong.

  15. drontal for cats Says:

    drontal for cats…

    […]ELI J. FINKEL AND BENJAMIN R. KARNEY are wrong at least when it comes to POF « Plenty of fish blog[…]…

  16. alicia Says:


  17. Carmelia Ray Says:

    Very interesting read. Markus seems to have a very strong opinion about psychological profiling and it’s effect on matchmaking where it comes to “online dating”. I appreciate the viewpoint. This is completely contrary to the “traditional” matchmakers (offline) view point that can prove psychological matchmaking based on compatibilities are necessary and a MUST for long lasting relationship success.
    The test of course based on the time frame doesn’t really satisfy the population of people looking to “settle down”. POF is a great site for dating. I’m sure there are thousands of short and long term “dating” relationships formed.
    I’m curious to know the stats over a longer period of time..say 3-4 yrs? This may indicate the long-term stability of those pairings in 2012.

  18. Carmelia Ray Says:

    If POF can accurately predict to the closest week of WHEN a relationshp is going to break up… Plenty of Fish has the cyrstal ball of your relationship future. POF may also stand for: PREDICTION OF FAILURE. Love it!
    I’m sure people would PAY for that! Think about it guys. Build an app for it. *Free Idea for the day

  19. MuscularDevelopmentStore.com Releases Navy SEAL 9-Week Training Plan, Designed to Simulate Authentic SEAL Training | muscular development, how to build muscle, muscle building Says:

    […] Hybrid Fitness BlogFitness Bloggers and Enthusiasts…Are you fitfluential?FitnessFixation.comHealthy Body Healthy MindPlenty of fish blog […]

  20. gary Says:

    my sister is being stalked by one of ur so called fine dating people and u need to do somthing to track people in case somthing bad happens.if she gets hurt in any way ,the lawsuit will be ledgendary.u can contact me at gbreese67@gmail.com.your help would be greatly appreciated to stop this situation plz.

  21. Goa Packages Says:

    Very well! I am not sure where you’re getting your info, but good very curious topic. I needs to spend some time learning much more or understanding more. Thanks for fantastic info I was looking for this information for my work.

  22. Anthony Caudill Says:

    “The best psychologists”… this is comedy. I still don’t get matched to women I’m attracted to even when I rely on “ultra matches”. That’s why I no longer pay. I get better results with OKCupid’s system… at least there I can weigh the results towards people who agree with me on social issues and screen for people who don’t shower at a reasonable interval..

    And oh yeah, all the psychologists remain clueless about how to match people. They are all on level as far as that goes. How about this: I LIKE GIRLS WITH BIG EYES. CAN YOU POSSIBLY HIRE A BUNCH OF PEOPLE TO LOOK THROUGH PROFILES AND FIND ME BIG EYED GIRLS, AND SHOW MY PROFILE TO THESE SAME GIRLS???? I think I’d probably pay 20 or 30 bucks a month for that kind of service.

    Carmellia, there is a branch of psychology called socionics (developed in Ukraine) which is pretty good for predicting relationship failure, at least for young people. You might look into that.

  23. Leona Says:

    Keep any repair or replacement costs in mind when you weigh your interest in any item.
    When utilized in institutions and music universities,
    this can be the ideal bench mainly because it can face up to every day use, and
    wear; along with longer lasting than vinyl. Just in case you require a replacement,
    there are actually numerous bench colors and also leg styles
    available to select the one that matches your preferences.

  24. Poulin Says:

    The best way to predict relationships is based on several links of planets in astrology with date of birth. In fact it is very simple and confirmed by experience.

  25. Katie Says:

    I really like POF, but on March 27, 2014 it quit loading, I have tried several times today but it would not come through, I checked with my internet carrier but they said it was not their problem, check with the POF people. can you fix the web site please.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: