In this chart here, as of last Friday’s close, we see that during last week, price edged somewhat higher, and on Wednesday it spiked down and bounce up off S3, so the uptrend is still in progress.
The object lesson here now is to see what’s happening with TF2, the TopFinder. In last week’s chart, you see that TF2 was fit to the pullback of Oct. 4th, at the large arrow. Now in this week’s chart, we see there have been two more significant pullbacks, at Oct. 19th and Oct. 27th, marked by the 2nd and 3rd large arrows here. Those later two pullbacks don’t quite fit TF2, but are close. In accordance with the directions for using TopFinders that I’ve written in our forthcoming book, what I’ve done on this week’s chart is to adjust D, the duration of the TopFinder, so that it fits the three pullbacks as closely as possible. I found that raising D from the original 20.2 million shares to 22.4 million makes the Topfinder slightly undershoot the first pullback, slightly overshoot the second one, and exactly fit the third one. This is a very good readjustment for the TopFinder, and likely to work well to its end. The far right vertical purple line is at the horizontal location of the expected end of this TopFinder, which is several days later that the first fit we had.
This procedure of readjusting the TopFinder works well when the pullbacks are close enough to get one fit that works for all of them. Here, the first two miss the readjusted TF by only about 1.5 out of about 1150 on the S&P 500. However, if these later pullbacks were at such different levels that such a close compromise fit couldn’t be found, then we would have to put on a new TopFinder fitted to the later pullbacks, while keeping the old one on the chart, and watch how price evolves.

Mr Hawkins,
The graphs in Intros’ 3 & 4 seem to show sup/rest lines emanating from the mid-price, whereas, the above shows them starting from the low of a bar that does have the lowest mean. If you have the time to explain this apparent difference, I should be much obliged.
John G
Addendum,
Sorry, should have read: “does not have the lowest mean.”
John G
John,
In Paul Levine’s original articles on the Midas method, most of his charts were point charts, each point being the mean of that day’s price. So, he always calculated the Midas S/R curves using the mean prices. Nowadays, most of our charts are bar charts (or candles or equivolume boxes), so calculating the curves with mean prices will make support and resistance display at the middle of the price bars. Some people are content with doing that, but I am not. I prefer to calculate support curves with low prices so that when price does support at a curve, it will be the low end of the bar that touches and bounces up off the curve. Similarly, I use high prices to calculate resistance curves so that resistance displays by the tops of the bars bouncing down from the R curve. I go into this subject in detail in one chapter of our forthcoming book.
Mr Hawkins,
Many thanks for you helpful reply. Amazon still indicating not available yet!
John,
The book is in production now, which is a painfully slow process. They’re estimating availability in March or April.
Regarding software to run the Midas curves and TopFinder, if you are a position trader and not an intra-day trader, there is an excellent software package available from StockShare Publishing. They have a Midas plug-in for their main program which I advised them on how to program. See
http://stocksharepublishing.com/