johnnyc
johnnyc
  • Threads: 5
  • Posts: 12
Joined: Jun 19, 2010
November 11th, 2010 at 4:52:51 AM permalink
When the Wizard discusses games like video poker, he compares games using Expected Return. For example, a video poker game may have a return of 98.98%

When the Wizard says a player's expected value for certain Blackjack rules is -0.51%, does the equate to a return of 99.49%?
Wizard
Administrator
Wizard
  • Threads: 1520
  • Posts: 27103
Joined: Oct 14, 2009
November 11th, 2010 at 6:35:35 AM permalink
Expected return = 1 + Expected value.

Usually you refer to expected return in machine games, and expected value in table games. It seems silly, but I'm just trying to be consistent with the way everybody else does it.
"For with much wisdom comes much sorrow." -- Ecclesiastes 1:18 (NIV)
ItsCalledSoccer
ItsCalledSoccer
  • Threads: 42
  • Posts: 735
Joined: Aug 30, 2010
November 11th, 2010 at 8:04:06 AM permalink
Quote: Wizard

Expected return = 1 + Expected value.

Usually you refer to expected return in machine games, and expected value in table games. It seems silly, but I'm just trying to be consistent with the way everybody else does it.



Not presuming I know more than the Wizard about calculating odds, but I'm used to a different definition of "return," more in the classic business sense, like in "investment return," "rate of return," etc.

In that sense, wouldn't return be (Q1 - Q0) / Q0? E.g., I invested in MGM at $3 in Spring 09, cashed out at $8 that summer, a return of (8 - 3) / 3 = 167% (forgetting about annualization, time value, etc.).

Or, if you walk in with $100 and walk out with $95, your "return" is (95 - 100) / 100 = minus-5%.

I guess statistical returns are a different animal than a financial return or something?
MathExtremist
MathExtremist
  • Threads: 88
  • Posts: 6526
Joined: Aug 31, 2010
November 11th, 2010 at 8:22:07 AM permalink
Quote: Wizard

Expected return = 1 + Expected value.

Usually you refer to expected return in machine games, and expected value in table games. It seems silly, but I'm just trying to be consistent with the way everybody else does it.



It has to do with the "N-to-1" vs. "N-for-1" issue. Slot games payouts are universally N-for-1 - the lowest award on any paytable is 0, which is a loss, so 1 represents a push (e.g. Jacks or Better). The "expected return" of a game, a.k.a. the payback percentage or the return-to-player (RTP) percentage, is the paytable times the probability distribution. That's going to be a number shy of 1.0, but can never be negative (since none of the pay values are negative).

Contrast with table game payouts. Those are usually N-to-1, so the lowest award on any paytable is -1 (a loss). When you take a table game paytable (even if it's just +1 and -1 for an even-money game like the passline) and multiply by the probability distribution, you get a number just shy of 0.0. So because the payout values are stated as -1 from how they'd be stated in slot games, so is the overall return percentage.
"In my own case, when it seemed to me after a long illness that death was close at hand, I found no little solace in playing constantly at dice." -- Girolamo Cardano, 1563
ItsCalledSoccer
ItsCalledSoccer
  • Threads: 42
  • Posts: 735
Joined: Aug 30, 2010
November 11th, 2010 at 8:28:43 AM permalink
I thought EV was just something to measure your session by. If a game has an EV of, oh, minus-1% and you return minus-2%, then your "luck" was a little worse than expected. However EV is measured - whether it's N-to1 or N-for-1 - didn't really make a difference?
johnnyc
johnnyc
  • Threads: 5
  • Posts: 12
Joined: Jun 19, 2010
November 11th, 2010 at 7:39:20 PM permalink
Thanks for the clarification!
  • Jump to: