Conversation Analytics in Text Messages


Once upon a time I met somebody. We texted all. damn. day. It was great. Apart from the impending flexibility limits in my thumbs, life was looking up.

This is a companion discussion topic for the original entry at


This was a dope read. Really impressed by the stats, made me think back to all of the times I was talking to a girl and all of the things I try to use to get the vibe. I would be interested in hearing how her energy seemed to you in messages, it seemed based on the stats like early on she was messaging you a lot and writing a lot and really into you but you consistently had short messages so maybe that led to her lowering her eagerness. I don’t think you’re washed though, I hope you use this again in your next relationship situation and see if it holds true.


They called me the love doctor back in my hay-day, so I’m naturally curious about this post! I think we have a wonderful opportunity here to cross two insanely powerful human intellectual curiousities: love-psychology and computer science. As they say, failures are great learning opportunities. Text activity between you all was pretty high and consistent compared to situtations in which numbers are exchanged, interest isn’t initially reciprocapted, and someone doesn’t text back in the first place. Those quick-drops are a lot less devastating. But in your scenario, you had them for so long! Big ups! I wonder what you’d find if you further dove into why they stayed devoted to your conversations for so long, why did the decline start to happen. I’m speculating, since I don’t have all the details, but maybe plausibility of things developing further weren’t there from their perspective. Maybe progress was slowing and so interest did as well? In my sad, sad experience often times the other person has lost interest when the “chase” became more of a “light jog”. Consistency was the measure of a successfull pursuit, and that means consistent progress.

I’m sending you positive energy and best wishes in all your future endeavors with future love interests. We all gotta reach the top of that rollercoaster sooner or later.


lolll, see id only want the stats after the fact cuz otherwise im definitely overthinking everything. but the one interesting point is like baselining. cuz everyone prolly got different habits/norms so its like in order to generalize you really kinda gotta train some type of model maybe and then use that to compare estimated behavior to actual. and look @ residuals or something along that line. well thats the first thing that comes to mind, but then its like okay how much time do you need to collect enough data to get a baseline that encompasses all the different modes of conversational operation? its also possible that people don’t operate like machines and conversational dynamics are continually changing in non-predictable ways, in which case you might as well just ask said person how they feeling :rofl:


@KrispOnTheBeat not nan person ever called you the love doctor in any dimension of any universe, but anyways! ah yea, definitely had all of these thoughts at one point in time, you know how life goes…

things change, people change, feelings change too