Video poker play - as is true for most casino games - is composed of streaks. There may be streaks of hands that win followed by streaks of hands that lose. Or there could be streaks of hands with high pairs, then streaks of hands with single high cards. Or there could be streaks where straights are predominant. In short, video poker is a random game.

While streaks are a normal part of random games, it is still very frustrating when you are in a long streak where none of the hands is a winner. I find one of the most frustrating streaks is one containing a low pair. Hands containing a single high card are mostly losers. Rightly so, these hands show up near the bottom of a strategy chart, so if they don't pan out, the results are expected. Low pairs, on the other hand, appear fairly high on strategy charts. In fact, for most of the non-wild card games, a low pair is positioned just slightly below a four-card open straight with three high cards (KQJT).

According to the Video Poker Strategy Master program, the expected return from holding a low pair is 0.8237 credits per credit played on a full pay (9/6) Jacks or Better game. In 9/6 Double Double Bonus Poker, a low pair consisting of 2's, 3's or 4's returns 0.8658 credits per credit played. A low pair consisting of 5's through 10's returns 0.7332 credits per credit played. These returns are quite high considering the return from holding a single high card is around half that amount.

So here is the conundrum. With such decent returns, why does it seem that a low pair almost never turns into a paying hand? How many times have you held a high card or two and then realized your hand contained a low pair (which is the proper hold)? It seems that every time I change my hold from something else to the low pair, I end up with nothing. It is almost as if the d**n game knows you made a mistake and is determined to punish you for making it. It seems that the only thing on the plus side when dealt a low pair is that if you do end up with a paying hand, it will pay better than a high pair. However, the ultimate paying hands from a low-pair hold seem really few and far between.

So, how uncommon are these elusive paying hands when dealt a low pair?

Let's take a look at what you can expect from a full pay (9 for 1 for a full house, 6 for 1 for a flush) Jacks or Better game. Note that other non-wild card games will give similar results. The information shown below is based on a hand with a low pair only and no possibility of a straight or flush as that might alter the proper hold away from a low pair. The actual return numbers are produced by the Winpoker program by Zamzone.

Consider a hand consisting of the 6 of clubs, 6 of diamonds, king of diamonds, 8 of hearts, and 2 of spades. Even though this hand is used as an example, any hand with a low pair and no shot at a flush or straight produces exactly the same results.

The highest return of any possible hold in the sample hand comes from holding the two 6's. That return is 4.1184 credits from the initial five credits bet. That is a better than 80 percent return. A low pair does have a fairly decent return. But, what makes up the return of 4.1184 credits? How is that return spread over how many different hands? Here is the breakdown.

There are a total of 16,215 possible hands that can be made when holding the two 6's. Of those 16,215 hands, 11,559 will return nothing. That is a very large number; over 70 percent of the possible hands will end up as losses. That means that only about three out of every 10 hands will be a winner of any kind. That is why it seems you never win anything when you hold a low pair. The win frequency is only about 30 percent.

As far as changing your hold to a low pair from something else, the human mind is sometimes a bit strange. You remember those hands when you had to switch to holding a low pair from holding something else. You especially remember losses - and seven out of 10 times you lose. If you keep good records of your actually results of several thousand hands containing just a low pair, you would find your results would be fairly close to the 70 percent losing hand rate.