Study Abstract:
http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=640052&dopt=Abstract
The administration of tamoxifen, 20 mg/day for 10 days, to normal males produced a moderate increase in luteinizing hormone (LH), follicle-stimulating hormone (FSH), testosterone, and estradiol levels, comparable to the effect of 150 mg of clomiphene citrate (Clomid).
What this means is that Nolvadex increases LH, FSH, testosterone and estradiol in much the same way as Clomid. You argued otherwise.
However, whereas Clomid produced a decrease in the LH response to LH-releasing hormone (LHRH), no such effect was seen after the administration of tamoxifen. In fact, prolonged treatment (6 weeks) with tamoxifen significantly increased the LH response to LHRL (sic)
It seems that you read the above statement to mean that LH output decreased with Clomid, but that isn't what it said. It says that LH response to LHRH (luteinizing hormone releasing hormone; perhaps more often called gonadotropin releasing hormone: GnRH) is decreased with clomid. But the thing that you have to realize is that GnRH (LHRH) output is significantly increased with use of clomid or nolvadex. In fact, this apparent desensitization _may_ (I don't know) be caused because GnRH output is excessively elevated. Ultimately, there is a net gain in LH output.
Also of note, it is important to realize that a LOT of studies are flawed and so it is important to read the entire study to determine if the study is even valid. Unfortunately, full study text is not usually available to everyone for free. In the case of an old study like this, the study might not be available online anywhere. So sometimes you just have to take a study summary abstract at face value, but you shouldn't put too much emphasis on it. In the case of clomid, I really don't care what the study data says because the anecdoctal evidence is so overwelming that it increases serum testosterone. It is like doing a study to show that Benadryl makes many people sleepy; we already know that it does and if the study says otherwise, then the study is just plain wrong.