Please understand that Gordon brews exclusively with RO or distilled water since his tap water in Dayton is fairly hard and alkaline. While that tap water could actually be suited for stout and porter brewing, its not well suited for other beers. So Gordon apparently follows AJ's advice to always start with RO.
Gordon was smart enough to figure out that this is a good idea on his own.
Worthwhile advice that works in many brewing situations, but not all.
It works in all cases provided the synthesized water meets the requirements. The brewer can synthesize pretty much anything he likes so if he needs alkalinity (which he rarely does) he adds alkalinity. The fact that he is starting with RO does not preclude his ability to do that nor does it relieve him of the responsibility of doing that if it is necessary.
The recommendation to reserve roast grains from the main mash is an important need when brewing with low alkalinity water.
IMO that's nonsense for the reasons I have set out in a previous post in this thread. If you are getting 'overnight in the coffee pot' harshness you are either using too much roast material or mishandling it or the maltster mishandled it. It's only in the mash for an hour plus sparge time which wouldn't seem to allow for 'overnight in the coffee pot' reactions to take place. I will, however, acknowledge that this is my opinion and also note that Guiness does (or did) extract the dark grain separately at their St James Gate facility but not at Park Royal. This latter contradicts the statement that this practice is 'an important need' with low akalinity water as London water is (per Lewis) lower in alkalinity (4.10 mVal) than Dublin (5.45). But note that both those values are much higher than 9% roast barley could neutralize even with a pretty stiff mash so both breweries must decarbonate.
Failing to either neutralize the roast acidity with mashing water alkalinity or reserve that roast from the main mash will diminish the resulting beer flavor and perception.
This has certainly not been my experience.
The beer will end up a little more tart and acidic if there isn't enough alkalinity and I find that those roast flavors are harsher.
Beer pH is set mostly by the yeast. I won't say mash pH has nothing to do with it - no feedback regulation system is perfect but the effect is minor.
Stout pH is at the low end of typical beer pH values at 4.21 ± 0.17 (one standard deviation) based on 21 commercial stouts measured by Michael Lewis and 2 of mine for which I collected complete data. The distribution is peaked about the mean but is otherwise pretty uniform. There is no correlation between pH and, for example, OG. Correlating the above data sets of pH and OG values gives a correlation coefficient of 0.09 which means that 99% of the variation in pH is caused by things other than the OG of the beer. Put another way there is a 34% chance that completely uncorrelated data could give us a coefficient as high as 0.09 (we say we are confident in the 0.09 value at the 34% level - 5% or less is considered good.)
Many brewers have found that roast flavors do become smoother as the kettle wort pH is pushed up into the 5.6 range.
Given that the recommendations in most texts are 5.0 - 5.2, that I brew stouts with kettle pH at 5.2 and they have none of the qualities attributed to insufficient alkalinity I'll have to call BS on that one.
The one outlier to that pH recommendation is for Irish Dry Stout which should have a crisp acidic bite to compliment the roast barley and raw barley flavors that are also signatures of that style.
Except that dry stouts don't have exceptionally low pH. They fall right around the average at 4.2. Other, richer stouts have pH lower than that, all the way down to 3.89 (two Samuel Smith beers). I believe that a lot of people think that tartness is mainly a function of pH and that may indeed be the case with inorganic acids but with these carboxylic organic acids that we are concerned with in brewing the cations have lots of flavor and I believe it is these anions that are responsible for the tartness of dry stouts.
When I wrote that I thought it seemed reasonable but I wasn't sure so I made solutions of HCl, HTart, HLac, and HMal of sufficient strength to bring pH to 3.0. With the hydrochloric acid the sourness was barely perceptible. Perhaps this is because of the sweetening effect of the chloride ion. There was certainly nothing you could call a flavor. The tartaric acid solution didn't taste very sour either but was definitely more sour than the HCl and had a flavor of sorts which was different from pure sour for sure and seemed to have a drying (astringent) component to it. The lactic was slightly more sour tasting than the tartaric and the drying feeling was there too. The flavor (the quality of the sour?) was definitely stronger. The malic, as all the wine makers here will have already guessed, had the strongest impression of sourness and the strongest flavor which one might imagine could get to 'bite' level if strengthened but at this level the solution would at best be called a bit sour. I hope some readers will try this experiment and I hope we can put to bed this notion that if the beer pH creeps low by 0.1 it will take on harsh sourness. It isn't, apparently, the hydrogen ion at all that is responsible for the sensation at food pH levels but rather the attending cation(s).
For most dark beer brewing, including enough alkalinity in the mashing water tends to produce better flavored beers.
I can only say, then, that my dark beers are not among 'most' beers but I have not experimented intentionally with beers mashed at high pH to see if they get noticeably worse. But putting my taste test together with the Lewis data I am beginning to suspect that these arguments are like the ones advanced in earlier days that CDs could never be as good as vinyl (I think there are some people who still believe that).
I was going to put some numbers RE the stout I have in the fermenter now as they may add some insight into what people are thinking about here but I'll do that in a separate post.