EDIT: Oh, and I found that "better" chart that I was talking about earlier relating to homicides in Britain. Here you go:
Firstly, "Homicides sans large anomalies unrelated to guns"? Well, that's vague. Do I have to hunt around for the double-dagger reference in all the rest of your 'support' evidence, or are can you give us something a bit more informative. (Also I use 'sans' a lot, but rarely in situations where I'm not being
ad hoc with the language, and knowingly highlit with italics. It may be a matter of writing style, but "w/out" (or "without" in full) would have been much better in that trend label, IMO. Consider me picky. But it
does sound like a weasel phrase for "stats rejected because they don't support the argument". Very careless.)
Secondly, this graph... Is it
actually the case that outside of the range 1997-2006 that Homicides exactly equates the "Homicides sans [
sic]..." value? That looks selective to me. The dotted line's movement on the graph outside of this range (and even in the eight years that can be infered, it is the
exact same value as the solid line for four of them!) might or might not show how extraordinary this limited range is, pattern-wise, but it looks like we don't get the opportunity to know and compare. Again, only stats that 'support' the evidence.
Thirdly, it doesn't even seem to support your case. If the solid line is (as it looks like trend title is claiming) the "guns and
only guns" trend rate, then the peak indicated by the undiluted value is a blip of non-gun homicides above the norm[1] i.e. a few people are
not using guns, rather than others using them
significantly more. (Meanwhile, if I'm wrong about the selectiveness in the second point,
every single homicide ('
sans' those that don't count) outside of that noted range was caused by guns? Wow. I don't know how the British press missed that trend.) So now we have stats that actually
contradict the implied state of affairs.
Fancy graph. "Better"? I don't think so. So many problems. Throw it away, it proves nothing (less than nothing, may even be contrary to your purpose) even if it could be trusted. (Assuming I'm not reading it upside down or something... No, fairly sure I'm not.)
[1] The norm being a general upward trend, which (without evidence to the contrary) one might as well assume to have started well before the handily-placed
first firearm control laws and be just a general upward trend dictated to by more population density, falling values 'immigrant problems', Hollywood, increased solar activity, recent alien parasite infestation... Well, YGTI. If anything you
might be able to say that, over the following five years from the gun control, the trend of (relevant) murders stopped going up like it had been (and, on a similar time-scale and period, the aftermath from the 1968 event seems to be suppressed significantly from the path it then continued to take, before people find out that they
can murder people after all (the other problems with the graph excepted). But it's just looks so broken that I don't think we could say a thing either way.