What Is Poker?

My long-time friend and colleague at Global Poker, Mike Jones (no relation that we know of), just dropped me a note. He said he recalled me giving a lecture about the elemental components of poker. That is, how far can you strip poker to its bare bones and still have a game that is reasonably called “poker”? I told him I’d write down a few thoughts from that lecture – then it occurred to me that it might be worth bringing to these pages.

But before I go further, I need to bow in the direction of one of the Rushmore-level giants of the modern game – Mike Caro. Mike first asked (and answered) this question decades ago in some long-forgotten article. Only I didn’t forget it, and I hope I will do Mike’s original thesis justice.

Here’s what you need:
  1. Two or more players. Note that by this definition, video poker is not poker. Which shows that we’re on the right track.
  2. Each player must have a “hand” that is simply a marker along some monotonic scale. Meaning that if hand A beats hand B, and hand B beats hand C, then hand A must beat hand C. We are all familiar with “High card < One Pair < Two Pair < Three of a Kind < Straight…” Or “9-8 low < 9-7 low < 8-5 low < 8-4 low < 7-6 low…” But it could just as easily be “Vanilla < Chocolate < Chocolate Mocha Swirl < Cherry Garcia.”¹
  3. Part or all of your “hand” is unknown to your opponents. In draw poker, none of your cards are known by your opponents. In 5-card stud, all your cards but your single hole card are known by your opponents. Note that this “incomplete information” component of the game is vital, and helps explain why poker models the real world in so many ways.
  4. There is betting, and specifically a player must be able to fold. Because once you have folding, you have bluffing, and once you have bluffing, you have poker. The betting could be as simple as the first player having the option to bet a fixed amount, which the opponent may call, or fold.
  5. There must be something of value at stake (or at least agreed upon value). That is, it’s fine to play for matchsticks as long as all players agree that winning all the matchsticks is a worthwhile goal.
  6. Before any player acts at the beginning of a hand, there must be “chips” (whatever they are) in the pot. This gives players an incentive to risk her chips to win the ones in the pot. That’s why blinds and/or antes are necessary. If there weren’t any blinds or antes, it would be mathematically incorrect for any player to voluntarily put chips in the pot.

I hope you realized that I never mentioned a deck of cards. Let’s visualize a situation where Mike and I want to play poker. But we have no money and no playing cards. The good news: we have a giant bag of marbles and a giant bag of Tootsie Rolls. We are good to go.

Each of us assembles a pile of 50 Tootsie Rolls.

  • At the beginning of the hand, each of us antes one Tootsie Roll into the pot. Now there is something of value to contest.
  • To start the hand, we each draw a marble from the bag. The marbles have the following colors: Red, Orange, Yellow, Green, Blue, Indigo, and Violet. The closer the marble’s color to Violet, the higher value it has. So Yellow beats Red; Blue beats Orange; Violet beats everything (and ties with another Violet marble). We do not show the marble to our opponent.
  • The “under the gun” player may bet 2 Tootsie Rolls or fold. If he folds, the other player wins the pot. If the UTG player bets 2 Tooties Rolls, the other player may call the 2 Tootsie Rolls, raise 2 more Tootsie Rolls, or fold. If there is a raise, the first player may call the 2 Tootsie Roll raise or fold. That is the only betting permitted.
  • If the final bet or raise is called, there is a showdown. The marble nearest Violet wins the pot. Identical color marbles split the pot.

Because Mike and I are the competitors we are, we play until one of us has all the candy. We call this game, “Heads-up for Rolls.”²

This is a perfectly valid poker game. It’s not a very good poker game, because there is a simple game theory optimal (GTO) solution to it. That is, if both Mike and I know the correct strategy and play perfectly, we will be at “Nash Equilibrium,” and neither player will have an edge on the other. We will simply trade Tootsie Rolls back and forth, depending on how the marbles fall. For more on this, see Andrew Brokos’s excellent book Play Optimal Poker – it provides a rigorous analysis of how to solve games such as this.

Conversely, if one of us plays well and the other plays poorly, the superior player will end up with all the chocolate. That is, it is a skill game.

Now, there are ways that the game can be made more interesting:

  • Each player draws two marbles, but must reveal the first marble he draws to his opponent. The value of the hand is the average color of the two marbles.
  • Before the betting, each player has the option of swapping his marble for another one (or one of the marbles if he has two).
  • There can be a round of betting, then a marble swap, then another round of betting.

There is still a GTO solution to the game, but the additional complexities make determining and playing it more difficult.

But even without those additions, Mike and I are playing poker. And in fact, it would be interesting to see which (or both) of us would be able to discern GTO strategy. I’ll say this: I would never play Heads-up for Rolls against Andrew Brokos.

As you can see, part of the beauty of poker is the simplicity at its core. A couple of players, a “hand” of some sort, incomplete information, and betting.

So, the next time somebody asks you how poker works, don’t start with hand rankings. Try something like, “Visualize two piles of Tootsie Rolls, and a bag of marbles…”

And thank you a lot to Mike Caro for the inspiration, all those years ago.

¹ Of course, this last sequence is provably correct.

² Thank you. I’ll be here all month.


Lee Jones has been in the poker industry for over 30 years. He writes at the Global Poker blog, plays poker every chance he gets, and coaches poker. You can contact him at www.leejones.com.​