Bloggers / Matthew Pitt / 13 January 2009 / Leave a comment

If you take away all the bells and whistles from poker, it becomes apparent that it is essentially a game of mathematics and as a result, players are subjected to what is called variance.

Poker forums the world over, contain literally hundreds of posts stating the author is running badly or is suffering from a bad spell of variance. Every poker player worth their salt has "run bad" at some point in their career, but is there a way to actually reduce the mathematical variance involved in poker?

The actual game you play dramatically alters the amount of variance you will encounter at the tables. Someone who plays Texas Hold'em will, as a general rule, suffer less swings than a Pot-limit Omaha player. This is mainly due to the fact that a Hold'em player can be a huge favourite pre and post flop, where as it is very rare to be better than a 60/40 favourite in Omaha at any point. As an Omaha player will essentially taking more coin flip situations, they should expect a higher degree of variance.

Betting structures also have a major effect on proceedings. It may come as a surprise that no-limit games have less variance than fixed-limit varieties. Why is this the case when it is possible to lose or win an entire stack in one hand in no-limit betting?

The thinking behind this principle is that in fixed-limit you almost always have to have the best hand to win as there is less room for bluffing and a lot more hands go to showdown, where in no-limit games, pots are won more regularly on the flop or turn and you do not have to have the best hand to win.

Another consideration is the number of tables you play. It is an accepted concept that the more tables you play, the lower your earn rate becomes and a lower earn rate leads to larger swings in your bankroll. By concentrating on one table, your skills and earn rate should increase and therefore your variance reduced. However, playing multiple tables does allow you to log more volume, which will help get through baron spells at a faster rate than playing a solitary table.

Finally, the type of game can contribute to higher variance than normal. For instance, cash games naturally have lower variance than sit and go's which in turn, have a lower variance than multi-table tournaments. Again this boils down to the number of showdowns during a session and how often you have to commit your chips as a slight favourite.

It is quite alarming that if you go all-in twice as a 70% favourite, you are actually only 49% to win both encounters! Imagine how many times you commit yourself in each of the recognised poker disciplines and it is easy to see why some have a higher exposure to the deadly variance!

Related Readings

Volume and variance - Marcus Bateman
Bankroll management, variance and bluffing - Pud's Poker Blog
A Slave To Variance - Pud's Poker Blog
Becoming a better player - Annette Obrestad
Monkey grinding and some MTTs - John Tabatabai
Ten Sick Poker hands - Poker Web Watch