Lee Jones and "The Clock": Revenge of the Short Stacks

Poker Stars' site-host/media-spokesman Lee Jones offered food for thought in his recent column in Card Player, "Folding Aces." The piece starts with Jones's recounting of a letter from a poker-playing friend, who offered a rare tourney anecdote, folding pocket aces pre-flop near the point of a Poker Stars tournament when the "bubble" was about to burst. In this case, it was easily the proper play, because the "cash" in this qualifier was the same for all who survived the super-satellite bubble: a seat to a higher tourney. The friend prevented herself from even a low risk to her middlish stack, and was justly rewarded when the bubble burst --- and she moved on --- a short time after.

What happened, though, was that Jones then twisted his piece from an examination of one thing (the interesting risk-vs-reward concept involved with this pre-flop dumping of rockets) to a diatribe against another, that bane of online tournaments --- the inevitable slow play that takes place before the bubble bursts.

Interesting indeed, because the two are absolutely unrelated. Dumping the aces in a situation like that presented in the column was the right play, independent of the "... everyone was taking the clock to the max and even timing out..." line in the friend's letter that seems to have raised Jones' hackles. And let's be honest: No one enjoys the tedious play just preceding the point where each site's software cries "Uncle!" and forces everyone to play hand-for-hand.

But there's a lot more here than meets the eye.

When Jones made the jump to the topic of "the clock," his piece jumped the tracks. It turns out that his thoughts and comments on his second topic are, at the least, open to debate. So take a moment to read the Jones diatribe, and then resume your readin' here.

Back so soon? Well, let's give credit where credit is due: Jones offers some excellent observations --- along with inadvertent comments on human behavior --- when he notes the following:

(a) It makes little net difference (on a table-by-table basis), because if eveyone is slow-playing until their time clocks elapse, there's no net gain in hands played for any table;

(b) So much time elapses during that period when everyone is stalling that the blinds rise out of all proportion to the preceding play, and what results is a crapshoot.

Jones takes a shot at all the "Einsteins" who do this, without considering that they really have no choice in the matter. If all the other players left at this point of a tourney are "taking the clock," then even the pure-as-freshly-fallen-snow among us have no real choice but to do the same thing. Or, as Einstein might have quipped (if he were a poker player), Extremespeed = Minus Chip Count. Figure that one out. In 30 seconds.

Jones continues by citing all the e-mails that he and Poker Stars receive. "'Give us more play! More blind levels! Deeper stacks!' And within reason," he adds, "we do that." And then Jones reaches the edge of reason, stating that "Stalling like that in a tournament is cheating. It's against the spirit of the game and tournament directors in brick-and-mortar events don't even tolerate it."

Wait-t-t, hold on there. I've never played in a big brick-and-mortar tournament (just a few teensy ones), but every single account I've read of a large tournament talks about how the hand-to-hand action slows down as the bubble approaches. Folks dither, dawdle... and dither a little bit more. Brick-and-mortar tournaments do have "the clock," and yes, by all accounts, tournament directors do step in when someone gets a little too obvious in their dawdling. But pretending that no one plays more deliberately in a "real life" tourney when the bubble looms is every bit as dishonest as labeling all online players who "take the clock" as "cheaters."

Not convinced? Well, then look at it this way: There is no difference between a poker player who "takes the clock," and a winning football team, in the last minutes of a game, that lets the play clock wind down to "1" before snapping the ball. Both are playing within the constructs and rules of the game, at a time when the situation demands strategies markedly different than during most other periods of play. That the play is tedious and maddening is a sign that the structure and rules of the game itself are imperfect, not that the participants are "cheaters."

We'll cut Jones a little bit of slack, though, and assume it's just frustration from reading the same e-mail five thousand times. And let's take a look at a couple of things that Jones skipped over in his need to vent:

1) The higher the buy-in, the more obvious the stalling as the bubble looms. If you play in a tourney with a $5 or $10 buy-in, you really won't see much change in the speed of the play until the bubble gets tight. You also tend to see the true stall only from the very shortest of stacks. Conversely, the behavior Jones mentions occurs most often in the big-dollar events, where even the refund of the buy-in that surviving the bubble represents represents a significant return on the time spent to that point. If I've spent three hours chasing what looks to be $10... so what? But if I've spent three hours chasing what looks to be $300... then yes, there's heightened incentive. And what makes it worse, from the viewpoint of the sites involved, is that they can measure exactly how long each hand takes. The spike in time and resources used is so damned easy for them to see, much easier to quantify than for their brick-and-mortar counterparts.

2) The blind structures within any given site and tournament are vital to the topic, far more so than Jones allows for in his piece. The flatter the blind structure (when all components are considered), the less of an effect that stalling has. In his piece, Jones created an example that illustrates exactly how --- relative to the number of hands that are being played --- the blinds might be climbing three times as fast as at earlier stages of the same tournament. And he's... ummm... the man; he's in a position to know.

At some point I'll provide a table to our readers that compares the relative values of blinds and antes for several online sites' multi-table-tournaments; such a comparative table points out a few things you might not otherwise realize. But that's a topic for another post. For now, let's stay on track and address those couple of things that Jones got wrong --- or ignored --- in the interests of pursuing the diatribe.

First, each online site has the option of adjusting its programming for exactly when a tournament near the bubble converts to hand-for-hand play. The most recent tournament I played in at Stars was over 1,000 players, but it still didn't going to hand-for-hand until exactly one player was left to be eliminated before the bubble was cracked. That's fine in a 100-player tournament, but given the number of tourney participants Stars pulls in (usually large, like this one), it's downright stoooopid.

"But wait," comes the chorus, "that's how they do it in real life."

We need a short digression to explain the wrongness of this one. The classic science-fiction writer Robert A. Heinlein popularized an overly fancy word, verisimilitude, which means: the quality of appearing to be true or real. In practice, it means trying to make the not-real as real-seeming as possible, or, as it applies here, to make the bits and bytes of an online casino as similar as possible to their real-life, brick-and-mortar counterparts.

But a funny thing happened on the way to the ISP: we've learned that online and casino poker are related, but different games. That's why mirroring what takes place in a live tournament at bubble time is simply the wrong approach. Online poker is more dependent on "the clock." It's not a matter of why or how or because; it simply is. The truth is that an online poker site can come closer to mimicking the feel of a live tournament at bubble time by doing things less like a live tournament, not more.

What should Stars consider? The straightforward fix is to change the onset of hand-for-hand play, so that it takes place when a certain percentage of players need to be eliminated to burst the bubble, rather than a single one or two. As an example, a 2000-player tournament might pay out to the top 180 players (9% of the entrants). Rather than going to hand-for-hand when the field reaches bubble-number-plus-1, there's no reason it can't go hand-for-hand when it reaches bubble-number-plus-percentage-of-entrants. In this example, that percentage might be 0.5%. 0.5% of 2000 = 10 players, which means that the software would kick everyone to hand-for-hand play when the field gets down to 190... the 180 of the actual cashing number plus the 10 calculated by the percentage. It wouldn't eliminate the slow play, but it would damn well take the worst of the edge off the practice.

However, there's still another avenue to explore. Stars is one of the few sites that uses a two-part clock: the standard amount of time that the player gets to make their normal decision, plus a time bank of 60 seconds for use in crucial situations. (And yes, short-stack players use the remainder of their bank time at the bubble, which makes the "clock" problem even worse.) But what if instead of this static, two-part approach, Stars and other sites went to a dynamic, regenerating time bank?

Imagine, if you will, a setup where you received 60 seconds of time to play any consecutive ten hands. As the oldest hand dropped out of memory, the time you spent during that hand would be added back into your revolving time bank. 60 seconds doesn't sound like a lot, but let's be honest: most people would be folding most of their hands anyway, and by doing it quickly, they preserve their time bank for those three or four hands out of the current ten when they actually might have decisions to make.

A setup like this would all but eliminate the stall-stall-stall-fold game that defines online bubble play. The argument against it would be that it has no real-life counterpart.

And that, friends, is exactly the point.