Thompson Stickle -- Almost a Breakthrough?

Last week (after a two month wait!) I received my Y-DNA test results from FamilyTree DNA.  Although a newcomer to this particular tool for integrating genetic research into my family history work, I've been working working with autosomal DNA for the past couple of years.  Through that experience I've developed a basic understanding of the underlying science, and have had some success integrating results of genetic matches into my research (even though poor response rates on emails sent to potential "cousins" are a source of unending frustration).
After doing what I hoped was appropriate due diligence I decided to take the 37-marker variety of the Y test.  That choice seemed to provide the most bang for the buck, and appeared well-suited to accomplishing my goal of tying together some of the many STICKLE lines that I've identified through my one name study research.  I had no real expectation that the test results would provide any insights or clues related to my work on my own direct paternal line.
To my great surprise my report from FTDNA highlighted 36 matches, with at least 7 showing a genetic distance of 2 or less.  The list included a wide range of surnames, with a couple Dentons, a couple Barker/Bakers, an Edwards, and a Stevenson, to name just a few.  I was intrigued to count at least 7 entries with the surname Howell (or some variation), 5 of which indicated a genetic distance of 1 or 2 -- i.e. reflecting a very high probability of a paternal line match within 6 generations.  The proliferation of Howells struck me as odd: it is not a surname that has been on my genealogy radar screen in any of my research.  
At the very top of my list of likely matches was a PERDUE/PERDEW entry that showed a genetic distance of 0 and a strong likelihood of a match within 4 or 5 generations.  Fascinating.Strangely, though, there was a surname absent from the FTDNA list: not a single entry bore a name that was even remotely close to STICKLE.  A bit surprising? Or not? Read on....
I have a well-documented 'paper tree' -- the product of 30 years' work.  But on my direct paternal line the trail still ends where it has ended since I first got hooked: with Thompson Stickle, my 3rd great grandfather on my direct paternal line,  My most persistent 'brick wall' ancestor, Thompson was born somewhere in east central Ohio (likely Muskingum county) in 1818 to parents who I believe came from Frederick county, Virginia. (Autosomal tests completed by myself and several 4th cousins all help to confirm our shared connection to Thompson.)
A family tradition, strong circumstantial evidence, and at least one published source from the 19th century suggest that Thompson was orphaned or abandoned at a very young age, and raised by another family in the community.  In fact, Thompson named his oldest son Adin Slaughter Stickle, my 2nd great grandfather, in honor of the man who had raised him.
Over the years I've developed many theories and hunches about Thompson's origins, but found no real proof (even though dozens of Ancestry junk trees purport to connect him with a variety of fathers, often preposterously).
In my research I've scoured the records of all of the counties in east central Ohio, looking for Stickle families and attempting to connect Thompson with at least one of them.  While I've located several Stickle families, I've uncovered no convincing evidence directly connecting Thompson to any of them (even though there are numerous unsourced Ancestry trees that link Thompson to various parents, often in preposterous combinations).
Conditions in east central Ohio during the first two decades of the 19th century compound the challenges of the search.  The area had only recently been officially opened to white settlement.  The institutional framework of society was immature, and social conditions were very much in flux. Many families were squatters, and many were simply passing through -- settling for only a few years before pulling up stakes and heading further west.  Records are spotty, and many families had a strong interest in avoiding the notice of official record keepers.
As I reviewed my results from FTM I began to realize that the tradition of Thompson having been orphaned might have been a euphemism for some other turn of events that separated him from his biological family at a young age.
As mentioned above, one of the matches furnished by FTDNA suggests a genetic distance of 0, and posits a 93% probability that the match has occurred within the last 4 to 6 generations.  This match carried the surname PERDUE/PERDEW.  Again, not a name that has shown up in any of my Stickle-related work.  But a match that's tantalizingly close -- both chronologically and geographically.   Needless to say, I promptly dashed off an email to the lady who was identified as owning the PERDUE/PERDEW match. To my relief she responded the next day.  It was clear that she was a serious researcher, and we actually spoke last weekend. It turns out that the kit I match to was actually her father's and that their own  brick wall also exists with a 3rd great grandfather, Abner Perdew, who was born in southern Muskingum county . . .  in 1815. Just three years before Thompson Stickle.
I'm now leaning toward the hypothesis that Thompson might  have been the illegitimate son of one of the daughters in one of the STICKLE households in the area. That is certainly nowhere near a conclusion at this point, but I do know that illegitimacy rates were fairly high on the western frontier in the first 2 decades of the 19th century.  And the surname had to come from somewhere.                   For what I will call sentimental reasons I'm actually hoping that Thompson's mother was in fact a Stickle.  I've had this name all my life and I'd like to think that my connection to it goes back a little further than a mere 200 years!  On top of that, I have invested a great deal of thought and time into my one name study!!                                                                                                                         Meanwhile, the research goes on.  My new Purdue cousin (and her brother and a paternal aunt) have all volunteered to do autosomal tests,  The generational distance we are looking at is on the outer borders of the usefulness of the autosomal test.  But with several Perdues to match against my existing collection of known descendents of Thompson we might be able to find something.This week I'll review tax duplicates and census records and also look more closely at any printed histories or other sources that describe the early history of the area.  My past work with these sources has been narrowly focused on locating families with the Stickle surname -- now I want to look for Perdews and Howells as well! Finally, it appears that I'll need to invest another $108 to upgrade to FTM's Y-67 test, as some solid genetic genealogists make a good case that the Y-37 test has a propensity for generating false positives.  But after all these years we have what I believe is a good lead on Thompson.  That's exciting.                                                                                                                                               More to come.  I hope!