The Brew Matrix Experiment #2 - Mash Fines in the Boil x Kettle Trub in the Fermenter

[Bobby_M]

I don't think it helps to pick apart the statistics when Bobby said up front that this was a casual experiment/test. It discourages people from putting in the effort to do things like this. Seriously. Corncob, I am not trying to tear you down, what you said is probably true but I think a little bit of a filter might be a better approach. If you do not like the data, either put the effort in and make your own video & test or just let it pass. What this test showed is that there is a difference to people in the tasting and by Bobby himself. How much weight that holds for each of us is up to the reader.

If you were in person I do not think you would have given the statistical analysis. I am on a crusade for some more respect and manners on the internet. It takes a lot of time and effort and willingness to offer something to do a multi-week test like this. I get the impression that true statistics are kind of a mythical absolute that is almost never reached. It is a homebrew test and just don't squeeze the bag. Sorry and please carry on with your insightful posts...

I definitely don't want this thread to death spiral. I can read some criticisms about the approach and mildly defend and possibly disregard some things that are over my head. I know that true data nerds have the best intentions and don't want anyone to misunderstand reality.

 

[Bobby_M]

If anyone wants to play with the numbers..

 

[tdf]

Thanks for sharing the raw data. I’ve been enjoying this thread and I’m glad you put in the effort.

 

[pers]

Out of curiosity, was there any correlation between being able to correctly identify the 5th sample and the taster level of training?

 

[Bobby_M]

Out of curiosity, was there any correlation between being able to correctly identify the 5th sample and the taster level of training?

A taster accuracy of 3 means they scored the E cup the exact same as its duplicate sample. If they were one apart, that accuracy score went to 2, and so on.
I wasn't particularly impressed here. There were five non-BJCP judges that matched perfectly.

I think judges think too hard. I did hear one of them suggesting that maybe each one was dosed with Siebel calibration compounds. I suppose after tasting the diacetyl in A, it was a primed assumption.

Perhaps one thing that is interesting is that no matter how I applied weighting, whether it be extra weight for judge rank or E-cup accuracy, or BOTH, none of that changed the final ranking of the beers. That's why I left most of that out of the video.