Below is a post I made on May 15 when this forum was temporarily available during the server change. I see the post was subsequently lost.

NOTE: I've continued to test the use of RSI trendlines, searching for rules that will consistently generate profits. So far, it's been a mixed bag. One can still be whipsawed, and there are times when there can be a significant pullback before a signal ultimately yields a profit (or not). Indeed, sometimes a pullback will quickly turn into a major move in the other direction! Bottom line: I'm still fascinated by the potential of using RSI trendlines for identifying potentially profitable market moves, but paper trading has shown me that this doesn't work well in all markets.

One thing I've noticed that does seem to happen often enough to be a reliable trait is the tendency for RSI to bounce off of extended trendlines. For example, July Silver's RSI DID subsequently bounce off of the R4 trendline (same day as it dropped through S2, after which it has returned to the S2 uptrend).

......... (Below is my May 15 post.)

Ok, Jay, here's another chart. Participants at this forum will probably recognize it as yet another of my RSI trendline charts. Intrigued as I am by RSI trendlines, I make no apology for this. However, I should also point out that I'm no expert, but merely offering a few personal observations. I'm trying to identify some consistent rules and, as such, welcome viewers' comments or suggestions.

What most intrigues me about these trendlines is how an upward RSI breakthrough of a downtrend line often signals a price rally, and a downward RSI breakthrough of an uptrend line often signals a price decline. Another intriguing thing I've noticed is that RSI often bounces off of these lines.

Here is a chart of July Silver through Thursday. I've labeled the uptrend lines as 'S' (support) lines and the downtrend lines as 'R' (resistance) lines. On this chart, I've labeled four buy and sell points: A, B, C, and D. I've also shown two problem areas: ?1 and ?2. Also, in the price part of the chart, I've shown three horizontal lines of support and resistance.

One of my current rules is that for a breakthrough to be valid, corresponding trendlines must already have formed (identified by at least two points). Also, I prefer that those points be at least a week apart. Thus, we see a sell signal, A, occurring in January after S1 and R1 had formed. Unfortunately, that didn't work out well, as price soon stalled and RSI retreated back to above S1. I've shown the failure as ?1. One might have been warned, however, by the lack of confirmation of price dropping below the support line shown on the price chart.

Next, we see a buy signal, B, as RSI broke through R1, and as price also confirmed by rising above the resistance line shown on the price chart. This signal certainly worked out well !

End of February, the rally was over, and there was a sell signal, C, as RSI broke through S1, although the corresponding downtrend line, R2, had only weakly formed at that point. Also, note how RSI's decline was soon after obstructed by the extension of the R1 downtrend line. This activity then morphed into a two-month span, ?2, of nebulousness, where R2 had to be redrawn a couple of times before a support line, S2, could be identified. (I've labeled two of these lines R2 and R3.) The sell signal yielded a short-term gain, but the lack of price trend during that interval might have been vexing.

Finally, a clear buy signal, D, occurred, beginning of May, albeit before the breakthrough of the resistance line shown on the price chart had occurred. That line might now provide support, although the price peak that occurred in February should provide resistance. If price soon breaks to the downside, I'll be looking for an RSI break through S2 to be slowed by a bounce off of R4.