sorry if i wasnt clear... yes, i am talking about intra day volume for the up issues on a given exchange vs the down volume for the given exchanges. i usually run one for the nyse, one for the nas, and one combined for a picture on the overall market. any service that can provide you with TRIN or TICK should be able to do so for both exchanges and also provide a/d volume for them too, as that is a component in calculating TRIN and TICK.
Since this topic covered the spectrum from TRIN being useful, not useful, to simple things being better, here is a purely mechanical system trading off of TRIN and S&P500 data. This is an extremely simple system -13 lines of code. A total of three inputs that has a smooth grouping of values, so I'm confident that it will hold up going forward. I'm not trading it. This system trades off the TRIN as well as price action. It is using S&P500 data, not ES data. 20 min timeframe $50 deduction for comm/slip always in the market/reverse no stops or profit targets avg trade length 2.0 2 years' 2/00 - 2/02 (the data I had for TRIN) So yes, TRIN is a valuable market tool to use.
I couldn't trade without them. In particular using the TRIN at the open is extremely helpful. For those of you who suggest these market internal indacators aren't useful and use simple backtesting as the argument: I think that flawed. There is as much art to trading as there is science. If you look at relative extremes compared to what the market is doing tick/trin work wonders. If you plug numbers into a formula it's not going to tell you much.
I find them a very useful tool on my day trading, i recomend you trying to apply to technical indicators to tick and trin (i personally use macd and fast stoch), they're very uiseful on finding reversals and trend strength on tick and trin... Let me know what your conclusions were
That's what I do too. Bollinger bands and stochastics, but I use it along with TICK which I find more relevant. Using TICK without TRIN can be dangerous.
The other thing you can use is a model that correlates the stregth that breaks "channel lines", such as boll. or dolch.. Normally i use a ponderated average of day volume with average volume on the past x days (the x varies with each stock) which i use very frequently on swing trading with options. the use of this method with futures is yet being tested....
I would caution against using the SP500 data to test ES systems that involve stop-losses. You can get different results, although I believe that you can get a good idea whether a system works or not using only the SP500 data which is more easy to get.