The same guy that claimed a 9:20 scream that people here accept to fit their theory also said the sex was consensual and the police at first suspected that it was..
I do think that this need addressing and confronting (and not for the first time....).
You are implying here (I believe) that one should either believe Guede or disbelieve him - and that to "choose" to believe some of what he says, while at the same time "choosing" to disbelieve other parts of his version of events, is intellectually dishonest.
But that's simply not what is happening in this instance. What some - including me - have tried to do is examine Guede's
motivation for constructing the various elements of his version of events. By attempting to do so, it becomes more possible (though obviously not infallible) to hypothesise when Guede is telling the truth (i.e. where it suits him best to do so) and where he is lying (again, where it suits him best to do so).
So let's take the two elements of Guede's version that are mentioned here: the "consensual sex" element and the "scream at 9.20-9.30" element. I would argue strongly that the first of these is likely to be a lie: it should be abundantly clear to any normal-thinking person that Guede has to claim the sex was consensual (given that he cannot deny the presence of his DNA in the victim's vagina), for to do otherwise is to admit to a criminal offence there and then. And taken together with what we think we know of the victim's character, and the other evidence and testimony in the case, it's highly likely that this element was a lie that was told for Guede's own self-interest.
Now, on to the "scream at 9.20-9.30" element. As I and others have discussed here many times before (perhaps you missed it?), it's reasonable (in my opinion) to suppose that Guede was concerned at the time when he made this statement (while he was on the run in Germany) that earwitnesses outside the cottage might have heard the victim screaming, and might therefore be able to pinpoint the time of the scream(s). He might reasonably therefore have decided that his version of events
had to place the scream at the correct time, in order to avoid the possibility of flat contradiction by reliable witnesses further down the line.
For example, suppose Guede's version of events had been that he had had sex with Meredith, and that it was only some time after the event (say 10pm+) that he'd gone to the toilet and the mystical intruders had swept in and killed Meredith. If one or more witnesses then came forward to testify with certainty that they had heard a loud terrified woman's scream at 9.20pm, Guede would immediately find his version of events being gravely challenged. He therefore (in my opinion) knew that had to concoct a curious version where consensual sexual interaction started, but where he (Guede) had to terminate the sexual activity abruptly in order to go to the toilet, whereupon the intruders burst in etc etc. That was (in my opinion) the only way in which Guede could fit both the proven sexual interaction and the c9.20pm scream into the very narrow time frame.
Of course, my opinion is that the
truth is that none of this curious dance ever happened. The truth is that Guede confronted Meredith shortly after she arrived home at 9pm, and that he attacked and stabbed her by 9.20, in the course of a serious sexual assault. But of course Guede doesn't want to admit to any of this. And yet he knows he has to correlate the scream time in order to match any potential witnesses.
And
that is how it's reasonable to suppose that some parts of what Guede said are accurate, while other parts are not. The crucial factor is what one thinks Guede
needs to say in his own best interest.