Playing It Smart: Do all-slots casinos protect gamblers from themselves?

The Pennsylvania legislature has authorized "slot parlors" around the state. It and the governor have said they won't extend the franchise to table games. The rationale for slots and not tables involves safeguarding the solid citizens against high losses.

Contenders for the limited number of licenses were enthusiastic, to say the least. And the plans proposed were no plain-Jane grind joints but glitzatoria rivaling the plush pleasure palaces of Atlantic City, Las Vegas, and other major gambling venues.

These applicants are neither shoestring cowboy outfits nor babes from the betting backwaters. Yet, they're vying for the privilege of spending megabucks to build and operate world-class casinos just for slots. None of them appears to be a great philanthropic trust, so they're apparently confident they'll amortize their investments, cover their costs, and pocket profits. They know the source of the revenues for these achievements. Where on earth do the politicians think the cash will come from?

Some gamblers will win, regardless of whether they're at the slots or the tables. A few will do so in a big way, more will show modest gains. Of more interest, to patrons and politicians alike, are the rest. Those who lose. The fraction upon whom luck does not smile, and the amount of the hit they take.

To investigate this, in particular comparing slots and table games, consider an all-or-nothing scenario and a typical time frame. That is, posit punters beginning with bankrolls that represent their entertainment budgets and continue gambling until they go bust or run out of time. Working from the probabilities and payoffs of any game, and the sizes of bets made, the fraction of players who'll exhaust stakes of given sizes before completing sessions of various durations can be mathematically estimated.

For purposes of illustration, pick four games the traditional three-reel slots, 8-5 video poker, minibaccarat, and blackjack. Assume bets are $0.75 at the slots, $1.25 at video poker, $5 on Banker at minibaccarat, and $5 at blackjack. Use four hours as the desired length of time.

Say 1,000 people try their luck at each of the games. The accompanying table shows how many would lose all of various stakes within four hours, on the average.

For reference, the parameters used in calculating numbers of players per thousand shown in the table are as follows. Slot machine: house edge is 7.5 percent, decision rate is 10 rounds per minute or 600 rounds per hour; video poker: house edge is 5 percent, decision rate is five rounds per minute or 300 rounds per hour; minibaccarat: house edge is 1.06 percent, decision rate is 200 rounds per hour; blackjack: house edge is 0.5 percent, decision rate is 84 rounds per hour (four active spots at the table and the dealer shuffles by hand).

Average number of players per thousand who will exhaust the indicated bankrolls within four hours at the various games:

game		bankroll
	$50	$100	$200
3-reel slot machine	925	842	664
8-5 video poker	865	714	417
minibaccarat - Banker	785	558	202
blackjack	651	357	60

The differences between the average numbers of players expected to crash and burn in the cited games result primarily from two factors. They are house advantage or edge and decision rate. Both get less as you go from the slots to video poker to minibaccarat to blackjack. Although it's often ignored in evaluating games, decision rate is of particular importance because the edge nibbles away at a bankroll every time a new bet is made.

So, the politicians have it backwards. And the folks with the slot parlor licenses in Pennsylvania will be chortling with glee all the way to the bank. A conundrum captured in this clever couplet by that versifier of the vigorish, Sumner A Ingmark: