Does Scouting Video Poker Machine Payouts Make Sense?

21 June 2004

By Alan Krigman

Video poker veterans often discuss jacks-or-better, nothing-wild machines in terms of the returns paid on full houses and flushes. A "nine-six" game, for instance, returns 9-for-1 on a full house and 6-for-1 on a flush. Likewise, an "eight-five" machine gives back eight times the bet on a full house and five on a flush.

With an otherwise "standard" set of payouts and no bonuses, a nine-six game has a 99.5 percent payback for "expert" play. This is about the best anyone can anticipate nowadays. Accounting for cash back, comps, kitsch, and other booty the bosses bestow upon frequent patrons, it can yield a positive net expectation. But to the dismay of the hopeful, nine-six machines are rare.

At the eight-five level, payback is 97.3 percent; this'll beat almost any reel-type slot on the floor but approaches a point that peeves proficient punters. Machines with the other common full house and flush returns, eight and six, pay back an intermediate 98.3 percent -- better than some table games.

For those who play video poker regularly -- maybe a few hours a day, several days a week, or more -- small extra statistical percentages of every bet kept by the casinos have an erosive effect that takes its toll. What about the vast majority of players who indulge for a few hours a month or less? Should the distinctions among these games matter much to them?

The answer to this question depends on what's important to who's doing the asking. Individuals visit casinos for a variety of reasons. Contrary to the view held by many civilians and mass media reporters, it's not always to get rich quick without working hard. Still, some factors seem pervasive. Heading the list are the chance of enjoying a long enough session to consider the gambling experience worth the money at risk if not lost, and the probability of reaching some earnings goal.

Regardless of how particular video poker players set time or profit objectives, most will be satisfied with a royal flush jackpot. If not, they're obviously dropping their dough into the wrong device. So estimates of a game's potential for session duration and earnings would be of most interest to solid citizens who realize that, notwithstanding the day's horoscope, they're not apt to get this lucky. Prospects would accordingly discount the impact of the jackpot, and use a return percentage based only on the probabilities and payouts for the remaining hands.

The accompanying table compares the three typical versions of video poker on this basis for a competent, fast player with a $200 bankroll, betting five coins per spin at a $0.25 machine.

Expected player performance at 9-6, 8-5, and 8-5 jacks-or-better
video poker, assuming $1.25 bet per round on a $200 bankroll

game	return				chance of at least
	with JP	without JP	2 hrs	3 hrs	$100 profit	$200 profit
9-6	99.5%	97.50%	91%	80%	31%	10%
8-6	98.3%	96.30%	88%	72%	18%	4%
8-5	97.3%	95.30%	84%	63%	11%	1%

If you're planning to feed a portrait of Andy Jackson into the bill acceptor of a $0.25 machine and go for broke, differences between the three games are minor. However, the data show that the implications are meaningful on a single-session basis for serious players. And don't kid yourself into believing that your bankroll is chump change you can afford to lose. A stake equal to 160 times what you're comfortable betting makes you serious.

To illustrate the effect of the return differences, note from the table that your chance of staying in action for three hours under the assumed conditions falls from 80 to 63 percent going from nine-six to eight-five. Also, while doubling your money is a longshot without a jackpot even at nine-six, the 10 percent on that game far exceeds the 4 or 1 percent on the alternatives.

So, before you decide either that it's all just a matter of luck, or that you know a secret the casino bosses have been trying to keep under wraps, mutter to yourself this memorable metrical mantra by the muse of the machines, Sumner A Ingmark:

'Bout gambling and its surety, I've thought a lot,
Low edge improves security, though not a lot.