Here is an idea from long ago… using a sliding-window technique to compromise between responsive charts and summary-level information.

For scalping on @ES, 1597 share bars are about the fastest candles I can comfortably watch. Yet, when turns are fast, my bead-reader measure of the trend direction is too slow to turn around. So, I created a chart out of 377 share bar data, but created synthetic bars out of a rolling 5-bar window. These synthetic bars show approximately the same level of information as the 1597 bars did, so I don’t have to put up with any extra price noise.

These two example pictures should make it clear what I mean… you can see the 1597 share bars on the left, and my synthetic 1597-level info on the right, made out of ultra-fast 377-share bar data.

Interesting to note that the good PattyB (eotpro) signals on the 377 chart generally have corresponding PattyB signals on the 1597 chart. What a great indicator! It’s just that the rest of the “bead” indicators don’t usually line up in time to validate the faster turns. The 377 chart is fast enough, all around, though.

Also note how generally disorganized the beads are in that first pic, when the synthetic price chops through that blue moving average. Two indications of choppy action. (This works on normal charts as well, but the MA isn’t usually quite so eager to get inside the action.)

I’ll look through a bunch of historical data before I try to replace my 1597 chart with this synthetic one, but the idea makes enough sense…

Aside: Note that while the drawing part would be a ton harder, it should be possible to create synthetic renko/range/change/etc bars manually and draw them. The problem is, you wouldn’t be able to apply indicators to those charts, since there’s not a 1-to-1 correspondence between the underlying chart data and the synthetic bars. Oh well… I would like to play with those change bars, so maybe I’ll write the code regardless.