Thursday, April 12, 2007

Annals of Anecdotal Poker Perusal (v1.n1)

Concerning the vulnerability of strong hands post flop.


Abstract
We investigate the proposition that certain very strong hands post flop are, contrary to popular belief, in fact quite vulnerable. Results are presented and discussed.

Introduction
Other authors have previously observed that flopped straights, an apparently strong hand, are in fact quite vulnerable. This observation has been referred to variously as “Rule 45” or “Flopped straights always lose.” The intention of this study was to investigate whether or not this proposition also applied to stronger flopped hands.

A controlled experiment was carried out and all the relevant data recorded. Owing to various circumstances beyond the author's control (see Methods), only one hand met the criteria to be included in this study. This did however support the position that strong flopped hands (in this case, a king high flush) may in fact be more vulnerable than heretofore believed. While the statistical significance of these observations is problematic, the outcome is so surprising and of potential importance for the community of poker players and researchers, that we felt early publication was in order. Besides that, we have a promotion case coming up, and needed some new publications.

Methods
The investigators three-tabled at NLHE .10/.25 on the PokerStars site. The limits were chosen to ensure that the study remained within budget, although in fact the research resulted in a profit of $45. A total of N=317 hands were recorded, and for exactly Q=1 could it be determined that one of the players had flopped a flush or better. This hand is discussed in detail below. Pressures to publish, and threatened onset of insanity from playing at these limits on this site precluded further data collection.

Results
Eight handed, we were on the button with 5♥5♣.UTG limped, UTG+1 min raised, and all but one MP player and the blinds saw the flop of Q♦5♦7♦. A MP player bet $0.50 into a pot of $2.85, raised to $1 by the cutoff. With our set, we reraised to $2. MP folded and the cutoff called. The turn was the 5♣. The cutoff bet $2 of his remaining $7.50 into the $7.35 pot. We simply called, and then reraised the $2.50 river bet to put him all in. He had flopped a flush with K♦T♦.

Discussion and conclusions
As noted above, our data are somewhat limited, so it would be somewhat premature to draw any strong conclusions. In particular, another author has proposed that 55, also called Presto, is a hand of special signficance in that it wins with improbable frequency against apparently stronger hands. A check of our database (which we could not report in the body of the article, because the data was not gathered under controlled conditions) shows that on one other occasion 55 overcame a flopped top full house (JJ on a J55 flop), but it is unclear whether or not this would fit the parameters of the study in any case, as the JJ hand was already behind on the flop. Arguably, to support our tentative conclusions, a further J would have had to arrive on the turn or river.

Nonetheless, it seems that a trend is emerging, in which very strong flopped hands frequently lose. This was first observed concerning straights (and perhaps also by the previously mentioned author and ourselves concerning top two pair, though again this situation was discussed in the context of being behind already on the flop). If indeed it is the case that stronger and stronger hands are being run down, to use the vernacular parlance, then this may be an effect analogous to that by which industrial contaminants work their way upwards through the food chain, and is certainly of great significance.

Labels: ,