# Variations on Guaranteed BJ

Dec 17, 2007 11:17 PM

Last week, I described one of two ways a player can play Guaranteed Play Blackjack. In both cases, the player pays \$20 and receives 30 special tokens and "borrows" a stack of \$5 chips. The amount of chips is irrelevant as the player must pay back however many he borrows.

In the first method of play, the player must place one token with each \$5 chip and whether the player wins, loses or pushes, the token is taken by the dealer. If the player double-downs or splits, he must place an additional token as well.

In this alternative version of the game, the player must place two tokens with each \$5 wager, but he only loses the tokens if he loses the hand to the dealer. Again, if the player double-downs or splits, he must place an additional two tokens down as well.

As I did last week, I created a computer simulation that played Guaranteed Blackjack using this second variation. To save time, I took a couple of shortcuts but I don’t think this affected the larger results. If the player needed to double-down or split but had ran out of tokens, I allowed the player to go beyond the 15 sets (30 divided by two). Also, I am not sure of the rule with Surrender, so I basically made it a freebie. Both of these rule changes should mean that my simulation was a smidge more generous to the player than the actual game.

In the end, the numbers for this variation of the game wind up very similar to the first variation. If the player were to simply show up at a blackjack table and decide that when he has lost 15 total hands (where each split hand counts as one and a double down as two), he would find that he is a net loser about 52% of the time. He’ll win about 42% of the time and the rest will be pushes.

Obviously, the most he could lose would be 15 units (in my simulation, I had a few cases where he lost more, which was because I allowed the player to go beyond 15 to finish a split or double down). More than 85% of the losing sessions had a loss of between 1 and 8 units. Just over half of the losing sessions were a loss of 1-4 units.

On the winning side, most of the winning sessions were small. However, there were a significant number that were larger than 15 units; roughly 5% of the winning sessions. I’ll get back to the significance of this in a moment.

Of course, none of this takes into account that in order to play Guaranteed Play Blackjack you have to put up \$20 (4 units in this case) to buy the tokens. On the positive side, this means that the most you can lose is \$20.

For the roughly 25% of the sessions that you would have lost more than 4 units, this can be nice. The problem is that for another 23% of the sessions that you would have pushed or lost 1-3 units, you have actually increased your loss.

As in the first variation, this pattern continues on the winning side as well. All sessions in which the player won 1-3 units now become losses. This is 33% of the winning sessions; 6% of the sessions remain pushes. The remaining 21-22% of the sessions will be winners. Basically, you cut your chances of winning by half by using Guaranteed Play Blackjack. This is a significant cost in order to limit your loss to \$40.

Relative to the first variation, this variation offers one significant advantage. In the first version of the game, you’re limited to 30 hands — win, lose or push. As a result, even if you were to win all 30 hands (extremely unlikely), the most you can win is 30 units (okay, a bit more with blackjacks). In this second variation, if you get hot, you can keep on playing. So, in essence, you are wagering \$20 and hoping for a hot streak that will allow you to win far more than you normally could on a \$20 bankroll.

As I mentioned last week, my simulations assume perfect play. Most players don’t play this way. This variation of the game will be less forgiving then that first variation to the novice player as the novice is likely to lose more hands than an expert, thus shortening his session considerably. This may reduce the likelihood of a winning session to a very small percentage.

In the end, Guaranteed Play Blackjack offers exactly what its name implies — guaranteed "play." No one can guarantee winning play, but I still think you’d be better off applying your \$20 to a good blackjack book or training software. I’ve never bothered with a blackjack book, but we do sell Masque Multi-Play Video Poker software that includes Blackjack as well. For \$12, you’ll still have money left over from your \$20. If you’d like to order it, you can send a check or money order to Compu-Flyers, P.O. Box 132, Bogota, NJ 07603.