vpFREE2 Forums

Bankroll swings

#1

In a message dated 10/27/2003 8:38:46 AM Pacific Standard Time,
ugwf@hal-pc.org writes:
If you play a machine with a variance of 30, a payback of 101% and no CB,
will you see bigger swings in the bankroll than if you play a machine with a
variance of 30, a payback of 99% and 2% CB? My intuition says yes, but the TomSki
index for each would (I think) be the same.
Thanks
Your intuition is correct. Let's take it to an extreme, for illustration
purposes. Say you play a machine with a 1% total return, and 100% cash back (based
upon coin-in). It's pretty obvious that your bankroll "swings" will be small,
actually nonexistent since you will never drop below zero! So, the more of
the [total return + cash back] is cash back, the smaller the bankroll swings
will be. Unfortunately, for "normal" cash back percentages, the reduction in
swings would be quite small, perhaps comparable to the actual cash back
percentage.

Brian

[Non-text portions of this message have been removed]

#2

Gene F. wrote:

If you play a machine with a variance of 30, a payback of 101% and
no CB, will you see bigger swings in the bankroll than if you play a
machine with a variance of 30, a payback of 99% and 2% CB? My
intuition says yes ...

To which BJayGold replied:

Your intuition is correct. Let's take it to an extreme, for
illustration purposes. Say you play a machine with a 1% total
return, and 100% cash back (based upon coin-in). It's pretty obvious
that your bankroll "swings" will be small, actually nonexistent
since you will never drop below zero! So, the more of the [total
return + cash back] is cash back, the smaller the bankroll swings
will be.

Brian ... think this through a bit more. If we're comparing apples
with apples here than in your 1% machine return example that machine
also has a variance of 30.

However, any vp machine with a minimal 1% return will have extremely
low variance ... the outcome of every hand would certainly be a
complete loss of your wager in almost every case.

The only 1% ER machine that could have a variance of 30 would be one
in which the amount you're risking with each wager becomes varied and
potentially you could lose much more than a standard 5 credit wager.
But at this point we're on the fringe of fantasy.

Bottom line, Brian, the example fails on the variance count. As I
discussed earlier, both machines present the same bankroll risk.

- Harry