What
Kind of Records Should I Keep?
You should establish
two basic statistical barometers: Your win rate, expressed
as average amount of money won or lost per hour, and another
figure, called "standard deviation," to measure
short-term fluctuations. Your goal is to determine your
average hourly expectation. Here's what to do:
Whenever you
play online (or off):
|
Record
your buy-in. |
|
Record
games, limits, and site (e.g., $4-$8 Omaha/8;). |
|
Record
amount won or lost at the end of each hour. (Although
you can calculate average hourly winnings or losses
without recording data every sixty minutes, you'll
need this hourly information to calculate your standard
deviation, which we'll show you how to do shortly.) |
|
At
the end of the session, record the number of hours played
and the cash-out amount. |
You'll also want to
record this cumulative information:
|
Amount
won or lost for the month |
|
Amount
won or lost for the entire year |
|
|
Total
number of hours played during the year |
If you play at more
than one online site, we advise keeping separate hourly,
monthly, and yearly records for each one, then combining
them monthly or at least annually for overall results.
The calculations are
actually quite simple, even for the mathematically phobic.
Computing your win or loss rate is just a matter of dividing
amount of money won or lost by number of hours played. This
calculation yields the average amount of money won or lost
each hour. In statistics, we call that average figure the
mean.
Knowing how much you're
winning or losing on an hourly basis is important. But it's
also important to know if your average, or mean, is a realistic
barometer of your playing data.
That sounds confusing,
but it's really not. Here's a simple case to illustrate:
Let's say San Francisco and Kansas City each have an average
(mean) annual temperature of 65 degrees. But in San Francisco,
the temperature rarely gets extremely hot or cold, while
Kansas City is brutally hot in summer and bitterly cold
in winter. While mean annual temperature might be the same
for both cities, there's more of a spread between the highs
and lows in Kansas City than in San Francisco. Consequently,
the mean temperature of 65 degrees is more representative
of San Francisco's temperate climate than of Kansas City's
climatic extremes.
Now let's take the
same concept to poker: Two players might each win an average
of $15 per hour. One has big wins or big losses very rarely,
while the other goes through wild swings or fluctuations
(greater variance) to arrive at the same average win rate.
Who's better off? Clearly, it's the player who achieves
the same win rate while putting less of his bankroll at
risk.
To measure variance
- which speaks volumes about risk to your bankroll - we
need to do a bit more with those observed values (amounts
you won or lost each hour and recorded in your computer
file or Site site) that you used to calculate average wins
or losses. We need to know just how well the mean reflects
reality. That's where the standard deviation comes in.
