Two Plus Two Older Archives

Two Plus Two Older Archives (http://archives2.twoplustwo.com/index.php)
-   Poker Theory (http://archives2.twoplustwo.com/forumdisplay.php?f=13)
-   -   Technical analysis of poker charts -> any value? (http://archives2.twoplustwo.com/showthread.php?t=403781)

12-22-2005 11:57 PM

Technical analysis of poker charts -> any value?
 
I know many players that chart their poker winnings, generally with Time on the X-axis and BB/100 or $$ on the Y-axis.

After reading Mason's "Gambling Theory and Other Topics", I was intrigued with the idea of non-self weighting
strategies. Buy the book if you are unaware of this concept [img]/images/graemlins/smile.gif[/img]

If trends can be identified in a chart of your poker play, you may choose to pursue a non-self weighting strategy and alter your play based on your chart.

For example, you may elect to limit play when your poker chart is negative ("bearish"), and increase play when your chart is positive ("bullish"), just as a technical analyst may choose to increase his investment in a stock with a bullish stock chart.

A bullish chart may be one that is in a strong established uptrend, or has powered through previous resistance levels.

A bearish chart may be in a strongly negative downtrend, or one that crashed through previous support levels.

Many other definitions of bullish/bearish charts can be defined.

I am curious to ask those who chart your poker play: Does your poker play follow a "Random Walk" model, or a trending model?

A. Random Walk model

Random walk theory gained popularity in 1973 when Burton Malkiel wrote "A Random Walk Down Wall Street", a book that is now regarded as an investment classic. Random walk is a stock market theory that states that the past movement or direction of the price of a stock or overall market cannot be used to predict its future movement. Originally examined by Maurice Kendall in 1953, the theory states that stock price fluctuations are independent of each other and have the same probability distribution, but that over a period of time, prices maintain an upward trend.

In short, random walk says that stocks take a random and unpredictable path. The chance of a stock's future price going up is the same as it going down. A follower of random walk believes it is impossible to outperform the market without assuming additional risk. In his book, Malkiel preaches that both technical analysis and fundamental analysis are largely a waste of time and are still unproven in outperforming the markets.

More info: http://www.investopedia.com/universi.../concepts5.asp


B. Trending model

Personally, I'm a believer that stock prices do not trade in a random manner. Reason: Human emotion. Buyers were greedy
in 1999/early 2000 sending the NASDAQ to new highs. Fear took hold after 9/11 as stocks crashed to new lows, then recovered as buyers became less fearful.

Extreme confidence and fear should also be observable when examining a chart of one's poker play.

I suspect a poker playing bot would likely produce charts that show much more random fluctuations, where a human player's chart would show trends based on his state of mind. Trends should be found in a bot's chart but much stronger uptrends and downtrends should be found in a human's chart. Tilt, agressiveness, passivity, and tired play may cause trends to continue longer than what may be considered typical by standard variance.

The question I am pondering if you can alter your winrate by following a non-self weighting strategy based on your poker chart.

Anyway, check out your poker playing charts. Anyone care to discuss:

1. Are there defined levels of support and resistance in a chart of one's poker play? A stock chart may indicate strong resistence at say $50, bouncing off that level several times, falling back each time as sellers sell at that limit. In poker, it's possible similar resistance points may be
visible. For example, it is possible a player may have long term trouble pushing their bankroll past $1000, or $10000, as a "mental barrier" may exist at that level.

2. How strong are the uptrends and downtrends? How long can they last, and what happens when the trends end?

3. Are classic technical formations (triangles, wedges, head and shoulders, pennants, etc) visible in charts of poker play? Do these patterns have any predictive abilities?

4. Are the above items seen in the same frequency when charts of a poker playing bot are examined?

damaniac 12-23-2005 02:27 AM

Re: Technical analysis of poker charts -> any value?
 
First, let me say that I only have a vague understanding of the ideas you are mentioning.

That said, those graphs are mostly for fun. You get keep on eye on how you're running, show off you're OMG 200BB DOWNSWING, and look at a large upward spike 6 months later and fondly recall the 1 hour session where you won 75BB or whatever.

The problem (I think) with trying to use those charts for anything is that the longrun is simply way too long. Players will, presumably without major changes in ability or that of the opposition, run at very different winrates over different 50k hand periods. It'd doubtful that a confident/not confident mindset is going to run throughout a period that long: so much occurs that you're mental state is either going to remain pretty level (if you're good) or be all over the place (if you're not), but I'd be surprised if any one mood (or other factor) is likely to dominate over that many hands.

12-24-2005 12:34 PM

Re: Technical analysis of poker charts -> any value?
 
I had a similar thought the other day. Basically you could analyze your data and test for correlations between your winrate at any point and how well you were running just before that.

And yeah, if you find there is probably a correlation, it makes sense to try out some things you mentioned (such as play lower when running bad and play higher when running good)

AlanBostick 12-24-2005 02:14 PM

Re: Technical analysis of poker charts -> any value?
 
IMHO, technical analysis of stock charts is mumbo-jumbo on a par with analyzing patterns of flocks of birds or examining the entrails of a goat.

The reason for such things as "resistance" and "support" levels are structural: traders tend to put limit orders at nice round numbers. Nobody says "sell JUNK when the price reaches 28 5/8"; it's always "sell at 30". Resistance to price increases and support for price declines occur because of clusters of limit orders tend to get executed together at those levels.

No such underlying structure exists in poker results, and so technical analysis cannot reveal its effects.

As long as you are interested in such things, would you like to buy a lucky rabbit's foot? Or, how about a lighting a candle to St. Dismas?

12-25-2005 01:06 AM

Re: Technical analysis of poker charts -> any value?
 
[ QUOTE ]
No such underlying structure exists in poker results, and so technical analysis cannot reveal its effects.

[/ QUOTE ]

dude, you've never tilted before?

some times after taking a few beats I tilt...


All times are GMT -4. The time now is 09:53 AM.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, vBulletin Solutions Inc.