Film was not designed (for the most part) to be exposed and processed. The assumption is that there will be a reasonable lag between exposure and development. This lag is on the order of a minute or so, not seconds. On the other end, the LIK is supposed to be stable for up to a year or so, and I have verified this myself. Hot weather does make it worse.
Stephen's data in the reference from Rudi shows pretty much what I have seen, only his effect is more pronounced at longer times (days). Of course I refer to Kodak film, and we always over designed Kodak film to exceed ANSI standards. I did not see a reference to the type of film used in the referenced test, but you will find that id varies quite a bit from product to product.
That said, I suspect it is LKI due to a short time in the first test and a long time in the second. You might try exposing and then waiting about 2 hours and then processing. Something like that. Or, you could make up a batch of test strips at the same time, keep them for 1 day, and then freeze all but one. Use that one as the base and then take one from the freezer and use it as your reference when you need one. Make sure you protect all film from moisture, freezer burn, and also make sure you thaw it properly before use.