I've developed--and uploaded--an initial version of a spreadsheet to
record VP activity. Primary rationale for uploading is to show what
I think ought to be captured, and get some feedback. I have
concentrated upon the process of capturing the data; once this has
been done, various and sundry "processing" functions can be
developed later.
Intent of this is to be usable for at least the most-common positive
VP games. In addition to raw win-loss values, I also propose to
capture slot-club points, and the number of low-frequency hands
(Royal Flushes, various four of a kind, straight flushes, 5 of a
kind, etc).
Intent of the latter is to produce an "adjusted return" statistic;
the percentage return which would have been obtained had zero low-
frequency hands been received. For purposes of discussion, assume
that, in Downtown Deuces Wild, total expected return is 101%, of
which 2% results from Natural Royals and 4% from Four Deuces. Thus,
perfect play and the "expected" percentages of all lower-paying
hands--but zero Royals and zero Four Deuces--should result in a
return of 95%; loss of $5 per $100 wagered. If one DOES receive one
or more high-paying hands, and subtracts their payouts from total
return, the remaining percentage should still be around 95%.
Also "backing out" the medium-frequency hands (Straight Flushes, 5
of a kind, wild Royals, natural four of various kinds) would
improve the reliability of the "adjusted return", but it would
definitely require that more hands be recorded. I haven't done any
detailed analysis, but I "rather suspect" that hands which return
less than 25-for-1 tend to be fairly-evenly distributed; recording
of same would tend to be unnecessary.
Columns, in my spreadsheet.
A -- Casino. In a prior post, I proposed a scheme of Casino
Identifiers. Two-character state, one-character "region", three-
character casino name. (Last comprised--except as necessary to
avoid duplications--of the first three consonants in Casino Name).
Nevada, Laughlin, Edgewater
Nevada, Las Vegas, El Cortez
Nevada, Las Vegas, Hilton
B-D Start, stop, elapsed times. Last = stop minus start.
E -- Type of game. A restricted list of names. For actual usage,
there'd be another worksheet, containing (at least) full name of
game, expected percentage return, and pay schedule.
F -- Won/lost. Dollars won or lost.
G -- Slot Club points.
H -- Factor. Points per dollar of coin-in. (Actual factor at time
of play; likely to vary due to promotions, etc).
I -- Coin in, or dollars wagered. Points times Factor.
J -- Unadjusted Return Percentage. Coin in, plus or minus
won/lost, divided by coin in.
K-S Number of various low-frequency hands; Natural Royals, etc.
T Adjusted percentage. Return which would have been obtained,
in current session, had zero low-frequency hands been received.
U -- Expected percentage. Expected return when zero low-frequency
hands are received (assuming best-strategy play).
V -- Efficiency. Adjusted percentage, divided by expected
percentage. Assuming perfect play, long-term average of this should
be 100%.
If efficiency is significantly below 100%, one would be well-advised
to practice, with a computer "simulator", to determine whether one
is in fact following correct strategy.
As noted, I'm very much interested in "feedback" . .