DeGaulle wrote:You said you are dosing 17 grams. Does that fit well into the stock filter basket or did you get some other basket that allows higher dosing ? I don't know about the head space in your machine's group head, but if the puck is jammed against the screen even before you start the pre-infusion, it can be disturbed and channeling can occur. In that case you might want to dose a tad less.
To me, WDT as described is always a good idea if your grinder produces clumps or disperses into the basket unevenly. At first it seemed like OCD to me as well, but I have included into my routine and has done a lot for me in terms of getting consistency.

17g fits perfectly into my basket... I found that any less than that leaves too much headroom between my coffee and shower, leaving me with super soppy puck.

in regards to the earlier comments-
i've tried doing WDT, but I still see the same effect happening. could it be my basket or something? i've considered picking up a 15G VST basket... mine isn't clogged or anything; i can hold it up to light and see through all the holes. just not sure why some small spots seem to be slightly blocked. maybe i'll try tamping lighter tonight...

i would try to reduce the dose, grind much finer to compensate flow rate.
VST baskets need fine grind and good edge sealing.

I also suggest you to experiment on your stock basket first.

I finally fixed my laptop screen (stupid toshiba u920t has an exposed lcd cable in the back and bends over itself every time it closes....had to order one from Australia...)

But, now that I can see it properly I see what everyone is talking about, and looks like it is happening at 12oclock, 4oclock and 8 o clock. Kind of a bold conclusion to assume it is a consistent problem from one shot, but assuming it is:

My first thought is that maybe you are doing a tripple tamp and are applying pressure unevenly each time or rotating something? nutating tamp done incorrectly?

other thoughts

You mentioned your grinder clumping, so I would think it could also be caused by rotating the basket while grinding a few times, but allowing it to clump in a few locations more than others. Then you are doing the wdt but I agree with the others that something in your process of wdt-tamp is not fully distributing and leaving those three areas more dense than the neighboring areas. It must be pretty darn close though since the initial dispersion is quite even across the screen, but once at pressure it is finding easier paths. Maybe a short video showing your grind-tamp procedure would help?