So I'm going to continue my statistical analysis of my profile, maybe it will shed some light as to why I'm still single or how I should continue in the future. The first thing I decided to do was create word clouds. They're pretty hot right now in the whole blogging/web 2.0 scene and I thought they might prove useful. In order to find a use for them I started looking through my messages.
Right away I noticed two distinct populations (or datasets). Messages which were unsuccessful at obtaining a response from the solicited girl (Unanswered Messages), and messages that were successful at obtaining a response (Answered Messages). It was quite clear that I've become more successfull with time, and the majority of unanswered messages I've sent were at the beginning of my tenure on OkCupid, so I wondered if my message writing had improved and evolved over time. The first thing I did was plot my success rate over time (in this case months). Graph after the jump.
It's obvious right away that my speculation was true. May was by far my worst month, coming in at 1 reply for 11 messages sent. July was spent seeing The Separatist so I sent 0 messages (and received 0 replies to those messages). My success rate went up with each successive month, therefore I sent fewer messages in each successive month (9, 5, 3 and 2 for Sept - Nov so far). Makes sense, if my messages are successful then I don't need to send as many to get dates.
Going even further, if I assume that I'll get at least one date between Das German and Prostate Girl (see Scorecard), then I will have been on 12 dates from the website. 6 of my dates have come from unsolicited messages I've received, and 6 from messages that I've sent. 14 of my unsolicited messages were answered, therefore I'm hitting .429 when I get a reply to my original message. If I calculate my batting average for all the messages I've sent (which is 33) I'm hitting .182. Past my horrid opening month I'm hitting a much more respectable .273. Here's a percentage breakdown of my conversations (a conversation equals any message or series of messages between me and another unique profile).
Another thing I noticed was that I've been receiving more messages in each successive month as well. I'm not sure if this is because I've been more active, or if my profile has improved over that time, though there's no real way to test this (I have no access to past versions of my profile or activity history).
Obviously this month (3 messages received) isn't finished yet, but it's got a long way to catch up to September (9 messages), let alone October (16!). It might have something to do with the fact that my profile was deactivated/not very active for the majority of it so far.
To bring this post back to it's inspiration, OkTrends classifies successful men and women in different ways. For women, it's obvious - they just use the total number of messages they receive (the more messages, the more desirable the women). For men, since we don't receive near as many messages (men send 10X the number that women do), they've taken to a calculation involving the reply rate, as well as the number of messages received:
(# of Replies to Messages) + (Unsolicited Messages)
_________________________________
Number of Unsolicited Messages Sent
The calculation of my "Success Number":
14 + 44 / 33 = 1.76
The only thing I can really say about this is that a higher number is obviously better. Mine is bolstered a lot by the large number of unsolicited messages I receive, and has steadily increased as time has passed (meaning I've gotten more desirable, or more successful, or both).
Obviously October 2011 was a high water mark in my dating life.
From what I've read, most guys have a number less then 1 (indicating their reply rate is very low, or that they get few, to no, unsolicited messages). I believe OkTrends quoted a number of ~1.3 to be a fairly successful guy. Kudos to me?
Anyway, back to my original idea, word clouds! What I decided to do was analyze the messages I've sent for their most common words to see if there were any standouts between success and failure. I created two word clouds, one for messages that went unanswered, and one for successful messages.
Unanswered Messages:
Answered Messages:
Obviously you can probably discount "message", "Montreal" and "good", as they appear with roughly the same frequency between the two groups. One striking difference is the use of "like" trumping "good" in unanswered messages. I'm fairly sure it wasn't used in the colloquial sense in either group, but interesting nonetheless. Also, I started using opening lines in more recent attempts, including the phrase "OkCupid thinks I should send you a lovely message, and I agree", most of which appear at high rates in successful messages.
Words like "school", "want", "pretty" and "going" are negatively associated. Apparently going to school is a bad thing ("grad" school even more so). Whereas asking about their weekend or hoping, thinking and imagining are all positively associated. Another thing that caught my attention was "Tuesday" showing up in the later. I've opened, or ended quite a few messages with "How's your Tuesday going?"/"Hope you have a good Tuesday" - most of which were successful.
That's all she wrote for now. Let me know if you have any comments and suggestions. I'm not sure what else I can do with my profile, but this has definitely been informative to at least myself.
Have a good night, and a pleasant tomorrow.
Happy hunting as always,
This is hilarious. FINALLY Excel is useful for something.
ReplyDelete