+If there are multiple non-mutually-exclusive reasons why transitioning might seem like a good idea to someone, then the decision of whether to transition could take the form of a liability–threshold model: males transition if the _sum_ of their levels of femininity, autogynephilia, and culture-related-trans-disposition exceed some threshold (given some sensible scheme for quantifying and adding (!) these traits).
+
+You might ask: okay, but then where do the two types come from? This graph is just illustrating (conjectured) cause-and-effect relationships, but if we were actually to flesh it out as a complete Bayesian network, there would be additional data that quantitatively specifies what (probability distribution over) values each node takes conditional on the values of its parents. When I claim that Blanchard–Bailey–Lawrence's two-type taxonomy is a useful approximation for this causal model, I'm conjecturing that the distribution represented by this Bayesian network (if we had the complete network) could also be approximated a two-cluster model: _most_ people high in the "femininity" factor will be low in the "autogynephilia" factor and _vice versa_, such that you can buy decent predictive accuracy by casually speaking as if there were two discrete "types".
+
+
+[TODO—
+[The sexual orientation node increases femininity and decreases AGP, so those pathways are anti-correlated; however, the fact that straight AGP men also vary somewhat in their degree of femininity; some informal accounts (link Sailer) have emphasized how masculine (even hypermasculine) AGPs are, but this seems wrong]
+[briefly mention ETLE]
+[Berkson's paradox is also a thing]
+[People who don't quite seem to fit the coarse taxonomy might still be explained by the graph and a threshold model]
+]
+
+
+You might ask: okay, but hhy do I believe this? Anyone can name some variables and sketch a directed graph between them. Why should you believe this particular graph is _true_?
+
+Ultimately, the reader cannot abdicate responsibility to think it through and decide for herself ... but it seems to _me_ that all six arrows in the graph are things that we separately have a pretty large weight of evidence for, either in published scientific studies, or just informally looking at the world.
+
+The femininity→transition arrow is obvious. The sexual orientation→femininity arrow (representing the fact that gay men are more feminine than straight men), besides being stereotypical folk knowledge, has also been extensively documented, for example by [Lippa](/papers/lippa-gender-related_traits_in_gays.pdf) and by [Bailey and Zucker](/papers/bailey-zucker-childhood_sex-typed_behavior_and_sexual_orientation.pdf).
+
+The v-structure between
+
+[ETLE sexual orientation AGP v-structure, and effect of AGP on transition documented by Lawrence]
+
+[I don't have a good formal citation on cultural factors, but it seems really obvious if you've been paying attention for the last decade]
+]
+
+[quantifying the two-type effect:
+Lippa 2000 "Gender-Related Traits in [...]"
+2.70 effect of femininity for gay vs. not-day and 1.07 for "any" vs. "no" attraction to men
+mean GD score for non-lesbian women as 0.31; mean score for gay men was 0.30!
+—oh, maybe I want to be using Study 2, which had a better sample of gays
+GD occupations in study 2
+gay men are at .48 (.14); straight women at .36 (.13); straight men at .68 (.12)
+that's d=–1.61 between gay and straight men
+a gay man only needs to be 1 standard deviation (.48-.36 = 0.12) more feminine than average to be as feminine as a straight women
+whereas a straight man needs to be (.68-.36 = 0.32) 0.32/0.12=2.67 more feminine than average to be as feminine as a straight woman—that's rarer, but not impossible
+
+In percentile terms, 1-norm.cdf(1) = 0.15 of gay men are as feminine as a woman
+whereas 1-norm.cdf(2.67) = 0.003 of straight men are
+that's a likelihood ratio of 50 ... but the prior is not that far from 50:1 in the other direction! They cancel out!!
+
+For concreteness: what does the Bayes net spit out if 3% of men are gay, and 5% are AGP, and whatever other assumptions I need to make this work?
+Suppose gays transition if they're 2-sigma feminine ...
+
+]
+
+[further implications: as cultural factors increase, the late-onset type becomes more of a "NOS" rather than AGP type]