High, double-suited, non-pair hands in O/8

[chaos]

While perusing Bill Boston's tables this weekend I noticed the following results (all hands double suited):

A9TJ +4182
A9TQ +3066
A9TK - 368
A9JQ +1138
A9JK +2774
A9QK -4219

A9QK scoops 11.3% of the pots but is a loser! While A9JK can make one more straight than A9QK that doesn't seem like enough to account for the difference between winning and losing. Am I missing something?

[Buzz]

Chaos - I seem to have the second edition of Bill's book.

For double suited hands, my edition shows all of the hands you listed with negative cash results, as follows:
<font color="red">A9TJ -2010
A9TQ -6111
A9TK -2238
A9JQ -8535
A9JK -4069
A9QK -4505</font color>

Hmmm. Guess he chose a different line-up for my edition than for yours. There seems to be the greatest difference between your and my listed result for A9JQd and the least difference between your and my listed result for A9QKd. I wonder why.

Buzz

[chaos]

A9TJ ds was the best in both editions. This may be because the second flush can only be Jack-high at best and the simulated player was programmed to not chase with a fourth-nut flush draw. So it loses less by avoiding better flushes.

The spread between the best and worst performer increased from $6525 in your edition to $8401 in mine. I am not sure what this implies.

Since I do not understand what is going on in the simulation, I am not inclined to give much weight to the data points.

I am never sure if his A9QK ds results apply to a hand with a suited Ace and a suited King, or a suited Ace and a suited Queen, or some mix of the two. Are his single suited hands with an Ace, suited to the Ace? He doesn't appear to say in my edition.

[beernutz]

Would one of you mind posting the parameters used by Bill Boston that were input to Wilson Turbo? I would like to try and redo his work. Thanks very much.

[Buzz]

"Would one of you mind posting the parameters used by Bill Boston that were input to Wilson Turbo?"

Beernutz - On page 5 of my edition, Bill writes, 'The player type that I selected was "tight players."'

Except for that statement, I can't help you. I don't know what Bill's parameters were, and from what Chaos has written in this thread, I don't think Chaos does either. The parameters were evidently different for the data in the edition I have than for the data in the edition Chaos has.

"I would like to try and redo his work."

It might be more useful if you ran certain selected hands against various selected line-ups. For example, how does TJQKn do against (1) 9 loose pre-flop players, (2) 8 loose pre-flop players + 1 tight pre-flop player, (3) 7 loose pre-flop players + 2 tight pre-flop players, (4) 6 loose pre-flop players + 3 tight pre-flop players, (5) 5 loose pre-flop players + 4 tight pre-flop players, (6) 4 loose pre-flop players + 5 tight pre-flop players, (7) 3 loose pre-flop players + 6 tight pre-flop players, (8) 2 loose pre-flop players + 7 tight pre-flop players, (9) 1 loose pre-flop player + 8 tight pre-flop players, (10) 9 tight pre-flop players. Bill shows TJQKn as losing $4859 against tight players in 10000 simulations. It seems like TJQKn might be playable in some situations, but how many loose players at the table would it take for the hand to start showing a profit?

It also would be very helpful, for a given line-up of tight and loose players, to know specifically which cards are suited, and to see a comparison between Ad2dQdKd, Ad2dQdKh, Ad2dQhKd, Ad2hQdKd, Ah2dQdKd, Ad2dQcKh, Ad2cQdKh, Ad2cQhKd, Ah2dQdKc, Ah2dQcKd, Ah2cQdKd, and Ah2cQsKd. The first eleven of those twelve hands are single suited in diamonds, but you might expect each of them to play somewhat differently. And it would be interesting to make a comparison with the non-suited hand. Bill only lists A2QKs. By A2QKs, I think he means the ace is suited to the deuce and the other two cards are different suits (as in Ad2dQcKh). But what about the other suited hands?

Those kinds of comparisons might lead you to reasonable generalizations for use in actual play. If you do run any simulations, I think several of us here would be very interested in seeing the results, and might add some insight.

Buzz

[chaos]

Not much more information in my edition than Buzz's. 9-handed $10-20 game with "tight players."

He also states, "The results are based on games in which players have high starting-hand requirements and are aggressive with their good hands after the flop."

[beernutz]

For some reason, I seem to remember reading somewhere that Boston used chans as the lineup which is a table full of the C. Chan player who is described as 'Aggressive in TIGHT spots - ADVISES accordingly'.

If I do run some sims I'll be happy to post them.

[DPCondit]

I doubt he used that lineup. It was supposed to be tight, not necessarily Chans tight.

Don