Quote from Arnold Snyder in Blackbelt in Blackjack Book/2005 Edition, "You don't bet big simply because the count is high, you bet big because the count should come down... Ideally, you make small bets while the count is rising, and large bets while it is coming back down"(pages 200-201).
In my experience, the BEST Counters WAIT until AFTER a High Count PLATEAU, AS the Count BEGINS to DESCEND, BEFORE they INCREASE their Bets.
Interesting quote from the late Lawrence Revere in his book "Playing Blackjack As A Business"(1977 Edition),
"I did not, for many years, think that anyone could beat a Blacjack game. Then I began to notice certain winners. They were not counters, but I saw as I studied their actions that they played in a different way. The primary difference was that those who hit the most had the greatest success"(page 43).
"Lot's of rookie card counters think when a strong count has them betting it up, winning is a 'done deal'. This is far from true... Sometimes you'll count through half a dozen shoes waiting to find an opportunity to put some real money out there, and when it finally comes you'll lose five or six big bets in a row".
As you know, Revere was talking about a time when he did not count, and noticed that others that hit more seemed to do better. Then he hired Braun (IBM) to study the game, and figured out that low cards favored the house, and big cards favored the player. Thus the impetus for the book.
Considering the time-frame of that exerpt from the book, (mid-late 1950's) maybe a hand-full of 21 players on the planet knew how to count.
Nickels- I wonder about the system Braun used to do his blackjack analysis. In the fifties and to some extent in the early sixties we are talking about tubes (gas filled), drum memory and a hard disk(RAMAC) as big as a large frig. Not many companies had those things, but the Chicago area probably had a few where he could get time. Programming was very primitive and I bet that part of the task was more difficult than the math.