The Price of Fish

Everyone knows – or ought to by now – that game selection is one of the most important decisions a poker player can make. It's neither dramatic nor tension-filled, like waiting to see if your opponent makes his flush when you're sitting there with a made hand. And it certainly lacks the breath-holding, waiting-to-exhale drama of watching your opponent while he decides whether to call your bluff. But game selection is a lot more important than that. After all, just one pot is riding on your bluff; picking a good game can result in bigger wins and smaller losses as long as the game is good. Conversely, it can be disastrous if you choose a bad game.

Next issue you'll see side-by-side companion columns by Nolan Dalla and me as we jointly examine the impact of good and bad players in a game. But this issue I'll delve into the price of fish – or more precisely, the effect of adding one or two fishy players to your game, or one or two very good ones. After all, in that real-world game you're sitting in – as long as it's in a public cardroom or casino and not a home game, where the same coterie of regulars shows up weekly – the mix of players is always changing.

A game that begins with a lineup of tough, tight, aggressive players might change character, sometimes within as short a period as an hour, as good players leave and are replaced by poor ones. Astute players always monitor the lineup. That old poker mantra, "Be selective, be aggressive," doesn't deal solely with how you play a given hand; it is equally applicable when it comes to choosing a game. Selectivity means passing on tough games to play against a lineup you can beat; and being aggressive means sticking around as long as the game is good and you have the best of it.

So, what's a fish worth, anyway? Suppose you're in a game against relatively weak opponents, but one by one the weak players get tired of losing and drift away, only to be replaced by tough players who were sitting on the rail salivating as they awaited their chance to go up against that mostly fishy lineup.

To learn more about this, I ran a few simulations using Wilson's Turbo Texas Hold'em, which allows the user to insert player profiles – programs, really, that are designed to play with a variety of different characteristics – just like the guys at your table. To negate the effect of random card distribution from one simulation to the next, we used the repeatable deal feature, and the same hands were dealt in each of the simulations. Only the player mix changed. Each simulation was 600,000 hands. At 30 hands per hour, that's 10 years of poker. Each of the games was 10-handed, and the limit was $20-$40.

In the first simulation, Wilson's "Brett Maverick" profile occupied two of the seats; the other eight were taken by "Welcome Waldo." I'll leave it to you to figure out which profile was the fish – the names are a big help, trust me. At simulation's end, the Maverick profiles won an average of $44.50 per simulated hour, while each of the Waldos was stuck an average of slightly more than $11 per hour. That's not bad, and not unlike real life, either – although the simulation did not account for tokes or time collection, and over a 10-year period, the cost of the game would amount to more than $320,000 per player.

The second simulation was exactly like the first, except two more "Brett Mavericks" were added to the two already in the game, and the "Welcome Waldo" gene pool was reduced from eight to six. At the end of another simulated 10 years, Brett and his clones averaged $36.25 per hour, while the Waldos lost an average of $23.22 per hour.

The third time around, six "Brett Maverick" clones were up against four "Welcome Waldo" clones. Brett and the boys won $26.59 per hour, while the Waldo four lost $39.81 per hour.

It's easy to see what's going on here. Brett's average win dropped by 18 percent when the Waldo-to-Maverick ratio dropped from eight-to-two to six-to-four. That's about a 9 percent drop in winnings each time a "Welcome Waldo" was plucked from the game and replaced by a tougher "Brett Maverick." When the lineup went from six-to-four in favor of Waldo to six-to-four in favor of Maverick – now there were six solid players and only four fish – each Maverick lost about 26 percent from his win rate.

If different player profiles were used, these results would vary, just as they would if a different repeatable deal code were used and a different mix of 600,000 hands were dealt to each computerized participant in our simulation. After all, cards do make a difference, and more importantly, so does the relative difference in skill between the strong player profiles and the fishy ones.

All good players are not equally talented, and by the same token, some poor players are weak while others are positively atrocious. Moreover, as good a piece of software as Wilson's Turbo Texas Hold'em is – and in my opinion, it's in a league by itself – simulations are not real life. With all of these caveats firmly in mind, I'd still be willing to venture a guess that the price of fish is worth between 10 percent and 15 percent every time a strong player leaves your game and is replaced by a fish.

So, if you're a $20-$40 hold'em player who's winning about one big bet per hour, do whatever you can to convince your strong opponents to leave the game, particularly if there are weaker players waiting to be seated. Think of it this way: Each time a weak player replaces a strong one, it's just like he's paying you an additional $4 or $5 an hour to sit in your game. That's nice work if you can get it, isn't it?

Visit my website at www.loukrieger.com. My newest book, Poker for Dummies, is available at major bookstores everywhere.