Richard Munchkin and I regularly get “how do you do it” questions which we periodically answer on our podcast. Sometimes I have to give a briefer answer than I want on the air because a more complete answer requires that people see things written down. Plus, my blog is read primarily by players who understand the basics of video poker. The podcast is geared more towards players who play a variety of games with an advantage.

So today we’ll look at two recent questions I received at [email protected].

1. I was practicing NSU Deuces on the Video Poker for Winners software and I ran into this hand:  W W 4? 6?K?.

I held the W W 4? 6?.  Video Poker for Winners said I made an error and I should have only held the deuces.

However, your strategy card for NSU Deuces says that I hold a Straight Flush 4 with one gap (SF4 -1) and that is what I had chosen.

Unless I am mis-reading your strategy card, which is possible, which strategy should I follow?  I am following your Level 4 Advanced NSU strategy on the card.

Understanding someone else’s notation includes a learning curve. This is a situation where you are misunderstanding SF4-1. In the two-deuce section of the strategy card, it refers to hands with one inside, which is not synonymous with one gap. While all gaps are insides, not all insides are gaps. Specifically, if the combination is close enough to the deuce that the deuce may be included in the same straight or straight flush, then the combination includes an additional inside. This makes WW46 a SF4-2, rather than a SF4-1.

Why is this? Well compare the number of cards that can fill in a straight flush starting from WW57 (a true SF4-1) with WW46. In the former, you’ll have a suited 9, 8, 6, 4, and 3 along with the other two deuces, for a total of seven cards. In the latter, you’ll find 8, 7, 5, and 3, along with the other two deuces, for a total of six cards.

Another way to look at it is to use Video Poker for Winners. If you enter the hand in question, you’ll find WW46 is worth 15.21 coins. If you change the straight flush combination to WW57, you’ll find that worth 16.28 coins. If you were playing for dollars, five coins at a time, the difference in the value of the combinations is $1.07. The value of two deuces by themselves is somewhere between those two numbers.

In the basic strategy on the same card, SF4-1 is spelled out more (WW45-WW56, WW57-WW97). Our assumption when we constructed the cards was that players would master the basic strategy before they went to the advanced strategy. We don’t include the ranges on the advanced cards because we need room to describe the various penalty card situations. We also include an insert with the strategy card, defining all the terms. We understand that many players ignore the insert, but the information is there if they look for it.

This looks like a case where a player went directly to the advanced strategy, didn’t read the insert, and made assumptions about what SF4-1 really means. Unfortunately, the definition was a bit more complicated than he first realized.

2. If I know of a 100.1+% video poker game that I could play “perfectly” (less than 1 error per 100k hands), is it reasonable to intentionally make small, inexpensive errors to disguise the quality of the play? I’ve found a few spots where +EV can be maintained.

A 100.1% edge is very tiny. (Yes, I know you said 100.1%+, but I’m not sure how much that plus sign adds, so I’m treating it as if it is negligible.) If you’re playing for dollars, at a moderate speed of 800 hands per hour, that means you have an advantage of $4 per hour. And you want to give up part of that for purposes of disguise? For me, the game is not even playable.

Video poker has ups and downs. It’s possible that you could play this game for several years, perfectly, and still be negative. 

For casinos evaluating your play, they probably look at 100% accuracy and 99.8% accuracy as being identical. One has to be really knowledgeable to play a game 99.8% accurately. And yet on this game, with 99.8% accuracy you’re playing a losing game.

Insofar as your claim of one error per one hundred thousand hands goes, count me a skeptic that you could achieve that level of near perfection. Yes, that’s possible, but how would you measure this? Did you actually play a million hands and only have 10 errors? I doubt that. Most players do not play as well as they think they do. I am 100% positive that I know 9/6 Jacks or Better at the 100% accuracy level. But that doesn’t mean I never make a mistake due to playing too fast, letting my mind wander, or just becoming distracted for an instant. These types of errors probably sometimes happen to you as well, even if you know the game extremely well.

Your game may be minimally playable during multiple point days or during some other promotion. But as it is, it’s not worth your time.