New Indicator... Comments wanted

richtrader · Oct 23, 2002

I have been thinking a lot recently about how best to graphically represent a day's price action in a very small space. I have come up with an indicator which I think will give me a lot of information about a given stock, and I wanted opinions on it before I do the programming to make it happen.

A bit of background. To find my universe of stocks each day, I narrow down by average daily range (ADR) and average volume (ADV). Based on an ADR of about 1.1 and ADV of about 475,000, I end up with a subset of about 400-425 symbols, depending on the day.

I keep this list of symbols in a grid on my screen and get realtime data against this grid to keep the prices updated. The grid displays standard info like last trade, time, change, volume, etc. I have functionality to go through each symbol and make notes and assessments based on somewhat standard chart-reading techniques as to whether the stock looks good to trade or not. I give each symbol a score from 5 to 95 based on what I think of it. I also calculate the expected daily volume based on the current volume extrapolated for an entire day / ADV. I tend to target symbols which have a ratio above, say, 2 for volume based breakouts.

I have a second grid containing all of the symbols I deem to be "hot". This is maybe 10-20 at any given time. I view realtime 1 minute charts of these guys along with tick and trade data and make trades off this info.

Anyway, within the grid of the 400 symbols, I would like to place a graphical indicator which would tell me at a glance what the stock is doing that day. This indicator has to be small enough to fit within a cell of the grid. My idea is to create a candle of the day's action and overlay it against the ADR and the movement from the previous day.

Here are some examples I mocked up. They are at double size of what they will actually be upon implementation.

The gray background represents the ADR. If the candle is larger than the gray area, it means todays range is larger to the exact ratio the candle outsizes the gray area. If the candle is smaller than the gray area, it means the daily range is larger. In this case, the daily range takes up the entire cell and the candle is proportioned within it.

The gray area is always centered within the cell. The candles are placed according to the High + Low + Close / 3 of the previous day. This value is the midpoint horizontally of the cell. If the candle is up from that value, it will be placed to the right of this midpoint to the exact degree it is up (scaled to fit within the cell). If the candle is down from that value, it will be placed to the left of this midpoint using the same ratio. If the candle is the same, it will be centered.

Can anyone think of anything further I can do? I very much appreciate any input.

richtrader · Oct 25, 2002

okay, i agree now judging by the number of responses that my one bar indicator was a dumb idea which wouldn't give me enough info.

so i am going to implement a mini RT 4-day chart with a 5 day sma (gray line in the example)

again, i'd welcome any input or ideas on any more info i could display.

Thanks,

Rich

Free Thinker · Oct 25, 2002

erlanger quote can do this in a quote sheet.

limitdown · Oct 28, 2002

Originally posted by richtrader
okay, i agree now judging by the number of responses that my one bar indicator was a dumb idea which wouldn't give me enough info.

so i am going to implement a mini RT 4-day chart with a 5 day sma (gray line in the example)

again, i'd welcome any input or ideas on any more info i could display.

Thanks,

Rich

curious to know what your baseline software is that allows customized generation of signals, indicators and such?

incidently, when creating an indicator, one has to decide whether or not:
1) you will be true to the ideal, and support the theorum or
2) you will be true to its purpose, namely to provide trading detail

heavy words, meant for those who understand...

KymarFye · Oct 28, 2002

Never noticed the original post - kind of an interesting experiment. I think the key, and arguably the only thing that really matters, is how exactly the indicator is supposed to affect - set up, filter out, or help to reconfigure - a trade or trades, or at least influence a bias.

richtrader · Oct 28, 2002

Hi limitdown,

Thanks for your reply and words of wisdom.

I write my own software and have a realtime data feed so I can pretty much do whatever I want.

My idea was to, and this goes to KymarFye's point as well, be able to tell at a glance if any of the stocks in my list are making a tradeable move. I ended up programming and implementing a 5 day candle chart for each symbol over this weekend which appears in every row of my symbol list. I rotated it 90 degrees as shown so I could fit more range on the bars. As quotes and trade ticks come in, the chart updates.

If I recognize a pattern that I like (like Margo's 3 inside day pattern for example), I can then get all of the other technical information (intraday and history charts, etc) about the stock by clicking on it. From there I can decide whether or not to take a position.

inandlong · Oct 28, 2002

Never judge the quality of your ideas and thoughts by the lack of response or the nature of the responses you receive.

limitdown · Oct 28, 2002

I agree with the other post, and I salute you.

writing your own software, then we need to speak offline.

also, having customized your own indicators is quite admirable. at least you don't have to go far for "tech support"

now, with all this, remember to take your "mindset" out of the programming world, and ownership of your product/effort and replace it with the mindset of a trader.

based on a five day pattern, inside days, outside days all reflect "patterns of traders committments". In essence you are measuring what is called money flow.

With money flow, there are fundamentals that preceed the committment of money to these positions. Take significant note of the CNBC and CnnFn commentators who keep expressing "on the floor commentators", who express the "lack of commitment" to the trading patterns, and note the violent swings from +126 +94 to -300 -128 (4 trading days in that order).

essentially until the ecconomic news stabilizes to a better degree, then measuring volume or money flow will have far less of a predictor of future movement than it would normally have.

cheers, and PM me