Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

ESPN/USPC (rating)

3 views
Skip to first unread message

RazzO

unread,
Feb 18, 1997, 3:00:00 AM2/18/97
to

hi, it's razzo,
Advertised on my site for their TV coverage.
In my opinion, there was way too many advertisments. There should have
been more play shown. Although, they made an attempt, and a good one(jim
showing pplz how the flop works) to educate how the game is played, the
advertisers prevailed, dominated, actually! I count egual time to
commercials as to the same time for showing us hands played.
BOTTOM LINE? Its television!

see ya's,
razzo
POKERwwwORLD
http://pokerwwworld.com

Jeff Woods

unread,
Feb 19, 1997, 3:00:00 AM2/19/97
to

RazzO wrote:
>
> In my opinion, there was way too many advertisments. There should have
> been more play shown. Although, they made an attempt, and a good one
> (jim showing pplz how the flop works) to educate how the game is
> played, the advertisers prevailed, dominated, actually! I count egual
> time to commercials as to the same time for showing us hands played.

I wsa also frustrated that the first 30 minutes showed virtually no
"action" at all -- four of the five busts came in the last 10 minutes of
the program. The first 45 minutes was fluff and guff.

And wasn't Phil the most GRACIOUS loser you ever saw? ;-)


fi...@datasync.com

unread,
Feb 19, 1997, 3:00:00 AM2/19/97
to

> Jeff Woods <je...@delta.com> writes:
> And wasn't Phil the most GRACIOUS loser you ever saw? ;-)

I absolutely loved the fact they showed his rant(s).

"How could you call a reraise with seven high?"
After he lost a big hand of course.

Heh. He R00LZ.

Keith Fichtemaier | For an alarmingly large segment of our society,
| the internet provides that quality of existence
fi...@datasync.com | that used to be referred to as having a life...
This post is satirized for your protection.


Doug Murphy

unread,
Feb 19, 1997, 3:00:00 AM2/19/97
to

In article <330B04...@delta.com>, Jeff Woods <je...@delta.com> wrote:
>
>I was also frustrated that the first 30 minutes showed virtually no

>"action" at all -- four of the five busts came in the last 10 minutes of
>the program. The first 45 minutes was fluff and guff.

What frustrated me more was that there was no recognition of the smaller,
strategic moves that the various players put on the others leading to the
big breaks shown on the show (i.e. outbursts and bust-outs). For example,
there must have been some precedent from earlier behind Phil's outburst
about the 7-high. What happened on the preceding hands to provoke him?

Also, how did Ken build up his stack when he was so short stacked? They
never showed the critical (i.e. short-stack) portion of the comeback.

However, I must point out that I'd rather see this type of show to attract
new players/fish to the game than not see any... Now if we could get a
decent timeslot...

Doug Murphy
Omaha (the city, not the game)

Guy Berentsen

unread,
Feb 19, 1997, 3:00:00 AM2/19/97
to

I would have liked to see more hands,
maybe they can release a 2 hour video
version with some hands edited back in ;-).

I also would have liked to
know how much time had assed & what the blinds/antes were each hand.

In article <19970220020...@ladder02.news.aol.com>,
Nergd <ne...@aol.com> wrote:
>
>But my main question is: HOW THE HELL COULD PHIL HELLMUTH CALL WITH K-10
>HOPING FOR A GUT SHOT AGAINST A (RE)RAISED POT WITH AN ACE ON THE BOARD?
>(Maybe he was showing the world that he isn't so perfect after all, to
>entice calls from all us rotten players who watched! ;-))

That move made me suspect that a $ deal had already been made.

Anyone know if/when a deal was made?

Tom Weideman

unread,
Feb 19, 1997, 3:00:00 AM2/19/97
to

Nergd wrote:

> But my main question is: HOW THE HELL COULD PHIL HELLMUTH CALL WITH K-10
> HOPING FOR A GUT SHOT AGAINST A (RE)RAISED POT WITH AN ACE ON THE BOARD?
> (Maybe he was showing the world that he isn't so perfect after all, to
> entice calls from all us rotten players who watched! ;-))

Unless I am mistaken (or you are talking about a different hand that I
don't remember), PH bet the flop, "Skyhawk" (heh) raised a bit, and then
Phil jammed, which was then called. He didn't call with the gut shot. Oh,
and I think PH had KJ and the board was AT6, with Ken holding A6. I know I
should review the tape before posting, but I'll go out on a limb with my
admittedly poor short-term memory. Lastly, I am not defending his play,
but it is not as bad as you indicate.

Tom Weideman

Nergd

unread,
Feb 20, 1997, 3:00:00 AM2/20/97
to

You make good points. I agree there were too many ads, but time-shifting
with a VCR took care of that.

The production qualities could have been better. There was a lot of
"background noise," but perhaps this was unavoidable, due to the desire to
pick up the conversations/remarks of the players, which was very good.

I agree there should have been more hands shown, as I think there was in
the ESPN coverage of the WSOP a few years ago.

We tend to forget, however, that Poker is an inherently boring game to
"spectate" IMHO, and other items have to be "added" to capture
(manufacture?) the drama. I did like the "commentary" and "color" of the
"announcers."

JOEL STARZECKI

unread,
Feb 20, 1997, 3:00:00 AM2/20/97
to

One thing I found interesting was the tournament was BROUGHT TO YOU by Pepsi
and Taco Bell, thus, the advertising during the hour featured many of those
sponsors products. To the best of my knowledge, when the WSOP was shown on
ESPN Binions was the sole sponsor.

Maybe this could bring new $$$ into smaller local tournament prize pools. Say
the event was co-sponsored by a local Pepsi bottler or area Taco Bell
restraunts. All room managers need to do is go out and attempt to sell the
product of poker to these businesses. I know that wont be easy, but it would
worht a try.


Scr...@ix.netcom.comanti-spam

unread,
Feb 20, 1997, 3:00:00 AM2/20/97
to

This is a tangent but doesn't Pepsi own Taco Bell? And has Budweiser
sponsored anything? I'd think this would be right up their alley.

SSRTINC

unread,
Feb 20, 1997, 3:00:00 AM2/20/97
to

Regarding whether or not they made a deal at the final table:

I heard that the last 3 made a deal due primarily to the fact that
Surinder Sunar, being
a resident of the UK, would have a huge tax obligation. By making a deal
they each
made out better than they would have otherwise.

That's just a rumor but it makes perfect sense. If you noticed, Helmuth
wasn't even
upset when Skyhawk called.

Dan

Jazbo Burns

unread,
Feb 20, 1997, 3:00:00 AM2/20/97
to

ssr...@aol.com (SSRTINC) wrote:
>I heard that the last 3 made a deal due primarily to the fact that
>Surinder Sunar, being a resident of the UK, would have a huge tax
>obligation. By making a deal they each made out better than they
>would have otherwise.

There may very well have been a deal, but you've got the reasoning
backwards. UK residents have the delightful situation of being able
to take their poker winnings tax free here (by treaty) and at home.

--jazbo
--
- - - - -
Video poker strategy cards for sale:
http://www.monmouth.com/~jburns/vidpoker.html

Shirley Woo & King Yao

unread,
Feb 20, 1997, 3:00:00 AM2/20/97
to

I had heard that splitting a tournament was not allowed by the AC
Commission, yet I've heard from many sources that the 3 finalists took a
break to go to the bathroom in order to discuss the split (they weren't
allowed to do it in the room since it was not allowed by the
commission).

What I don't understand is how did they mechanically do the split?
If the commission did not allow splits, they would have to pay the
winner the whole 500,000, was there so much trust among the three that
they knew they would get there share?

Scott Byron

unread,
Feb 21, 1997, 3:00:00 AM2/21/97
to

The reason the first half of the program was so damned dull is that the first two
hours of the final table were so dull. It's hard to tell from the broadcast, but
they played six-handed for almost three hours before anyone was eliminated.

What's ironic is that the final table at the WSOP last year was action-packed, and
it won't be on ESPN.

Oh, well.


Scott


James Morgan

unread,
Feb 21, 1997, 3:00:00 AM2/21/97
to

SSRTINC <ssr...@aol.com> wrote:
: Regarding whether or not they made a deal at the final table:

: I heard that the last 3 made a deal due primarily to the fact that


: Surinder Sunar, being
: a resident of the UK, would have a huge tax obligation. By making a deal
: they each
: made out better than they would have otherwise.

: That's just a rumor but it makes perfect sense. If you noticed, Helmuth


: wasn't even
: upset when Skyhawk called.


That would explain the re-raise with KT and a board of AJx.

I for one, thought that was a poor play. The bet was fine, but
calling the raise and/or going over the top seemed virtually guaranteed
to lose (Although i guess it could get a jack to fold)


Jim Morgan


: Dan

Jeff Woods

unread,
Feb 21, 1997, 3:00:00 AM2/21/97
to

Jazbo Burns wrote:
>
> >I heard that the last 3 made a deal due primarily to the fact that
> >Surinder Sunar, being a resident of the UK, would have a huge tax
> >obligation. By making a deal they each made out better than they
> >would have otherwise.
>
> There may very well have been a deal, but you've got the reasoning
> backwards. UK residents have the delightful situation of being able
> to take their poker winnings tax free here (by treaty) and at home.

If that's the case, then I doubt there was a deal. Seems to me that it
would have been FAR to the advantage of the others to let Surindar *win*
the thing, and then get paid the amount of the deal under the table, if
he got his tax free.


Chris G. Straghalis

unread,
Feb 21, 1997, 3:00:00 AM2/21/97
to
Pepsi indeeds does own Taco Bell, although they are looking to spin off
their restaurant division.

pEaCe,

Chris

Nergd

unread,
Feb 22, 1997, 3:00:00 AM2/22/97
to

Anyone know if/when a deal was made?

>"Deals" are illegal in N.J., (A regulation of which I approve) but I
don't think that answers your question.<

Nergd

unread,
Feb 22, 1997, 3:00:00 AM2/22/97
to

Nergd wrote:

> But my main question is: HOW THE HELL COULD PHIL HELLMUTH CALL WITH K-10
> HOPING FOR A GUT SHOT AGAINST A (RE)RAISED POT WITH AN ACE ON THE BOARD?
> (Maybe he was showing the world that he isn't so perfect after all, to
> entice calls from all us rotten players who watched! ;-))

Unless I am mistaken (or you are talking about a different hand that I

don't remember), PH bet the flop, "Skyhawk" (heh) raised a bit, and then
Phil jammed, which was then called. He didn't call with the gut shot.
Oh,
and I think PH had KJ and the board was AT6, with Ken holding A6. I know
I
should review the tape before posting, but I'll go out on a limb with my
admittedly poor short-term memory. Lastly, I am not defending his play,
but it is not as bad as you indicate.

Tom Weideman

>I should review, too. I thought PH called the re-raise.

The outrageousness that I'm trying to illustrate here might be a bit
subtle. It just seems to me that the Ace on the board is of particular
note. It also seems to me that it would be reasonable for PH to expect his
opponent to have a pair of Aces, otherwise his opponent would muck,
fearing PH had the Ace. Assuming that's true (that PH's opponent had a
pair of Aces) why go all in on a gutshot?

Is that a better way of asking the question?<


russell rosenblum

unread,
Feb 24, 1997, 3:00:00 AM2/24/97
to

I agree. There are only 2 ways that Phil makes that play. One is if there
was a deal. And 2 is if he is on MONSTER tilt. To just say that it was an
aggressive play does not really make sense. He basically limped pre-flop.
He could have tried to steal the blind. Once the ace flopped I can't see a
need for him to make that play. If he were Sunar I could see it. If he
were short stacked and needed to double through it might make some sense.
But with his stack size he could have agressively blinded his opponants to
death.

I think a deal was made, and they were still playing for "the trophy".
Phil probably had enough of that, and took the first opportunity he could
to just jam and get the hell out of there. Remember all three of these
guys are pros, and the $$$ difference between first and second (and god
forbid third!), was massive. It was such a huge difference that it would
almost pay for the chip leader to take a three way split rather than risk
getting a $50,000 payday. The question is when they made the deal and how
much ph got compared to sky. I say phil took $300,000 Sy took $250,000 and
sunar took about $200,00 (whatever was left). They had to play out the
tourney, and I am sure they did not "fix" who would win. I just think PH
got bored.

By the way the "no deal" rule is really poor, and it really hurts the
regular tourney players.


> Nergd wrote:
>
> > But my main question is: HOW THE HELL COULD PHIL HELLMUTH CALL WITH K-10
> > HOPING FOR A GUT SHOT AGAINST A (RE)RAISED POT WITH AN ACE ON THE BOARD?
> > (Maybe he was showing the world that he isn't so perfect after all, to
> > entice calls from all us rotten players who watched! ;-))
>

Randall Flagg

unread,
Feb 24, 1997, 3:00:00 AM2/24/97
to

russell rosenblum wrote:
>
> I agree. There are only 2 ways that Phil makes that play. One is if there
> was a deal. And 2 is if he is on MONSTER tilt. To just say that it was an
> aggressive play does not really make sense. He basically limped pre-flop.
> He could have tried to steal the blind. Once the ace flopped I can't see a
> need for him to make that play. If he were Sunar I could see it. If he
> were short stacked and needed to double through it might make some sense.
> But with his stack size he could have agressively blinded his opponants to
> death.
>
> I think a deal was made, and they were still playing for "the trophy".
> Phil probably had enough of that, and took the first opportunity he could
> to just jam and get the hell out of there. Remember all three of these
> guys are pros, and the $$$ difference between first and second (and god
> forbid third!), was massive. It was such a huge difference that it would
> almost pay for the chip leader to take a three way split rather than risk
> getting a $50,000 payday. The question is when they made the deal and how
> much ph got compared to sky. I say phil took $300,000 Sy took $250,000 and
> sunar took about $200,00 (whatever was left). They had to play out the
> tourney, and I am sure they did not "fix" who would win. I just think PH
> got bored.

A deal had been made at that point, and the deal was precisely as
another poster indicated: each player got $200k, and they played for
the remaining $65k. The deal was made shortly after the hand in which
Sunar doubled up with his 76s versus Phil's pocket 4's. That was the
first time Phil ever considered making a deal, since he had been chip
leader the entire time. At that point, he told Sunar and Flaton that
he'd be willing to discuss a deal, but they had better be prepared to
give him the best of it. At the time of the deal, he actually had the
smallest number of chips, but all three were fairly close, so he
actually did get the best of the deal.

Phil did not get bored. Even if the deal had distributed all of the
prize money, Phil would probably be too competitive to ever 'get bored'
at that final table. He made the play that he felt was best at the
time. He didn't 'call' with a gutshot; he moved all-in with a
gutshot. He obviously felt that he could get Flaton off of many of the
possible hands that Flaton could have, as well as the added equity of
having outs. Phil didn't seem to think about it for very long at all,
and it might not have been too good of a play, but it's not nearly as
bad as many of us have indicated. If you were in Flaton's position, and
you had a hand such as A5, would you have been able to withstand the
heat? We weren't able to see enough of the play to determine if Flaton
would have known to call with such a weak hand, but Phil obviously
thought that he would let it go. Flaton happened to have a huge hand,
and it looked like a bad move on Phil's part, but none of us were able
to see the many hands in which Phil's aggression paid off.

And yes, there is a decent possibility that Phil was tilted out of his
skull on that hand. As I said, he seemed to act extremely quickly when
Flaton raised him. For three days, Phil had expected that he would be
winning that $500k, and the previous hour had caused him to lose the
chip lead and make a deal with a $300k difference in earnings. Sure, he
might have been tilted, since he is a fairly emotional player.

Phil plays a lot differently than most players, and people are always
quick to point out when he gets tripped with his over-aggression. The
fact remains that his aggression seems to pay off in tournaments, and I
doubt that there is a better horse to bet on in a hold'em tournament.


RF.

Lee Jones

unread,
Feb 25, 1997, 3:00:00 AM2/25/97
to

In article <russell-2402...@russell.worldweb.net>,

russell rosenblum <rus...@worldweb.net> wrote:
>I agree. There are only 2 ways that Phil makes that play. One is if there
>was a deal. And 2 is if he is on MONSTER tilt.

Either is credible...

>By the way the "no deal" rule is really poor, and it really hurts the
>regular tourney players.

[re-igniting the tourney deal-making wars]

I simply do not understand this. In other sports, folks are expected to play
until there's a winner. One winner.

What galls me is the complete and total hypocrisy in the prize structures
that are claimed in the big tournaments. They have this *massive* drop-off
(like 40/20/10/peanuts) in the pay-offs, but it's clear that most participants
would much rather reduce the variance, so they make deals.

I don't know the details, but I understand that at the recent 4Q event, they
substantially reduced the slope of the pay-offs, true? That seems like a
reasonable thing. It seems that in *every* tournament I see, one or more
deals are made among the last few players. I claim that that's prima faciae
proof that the players don't like the payoffs as they're structured.

The result, as we see it, is just when it should be *really* exciting (we're
down to 2-4 players, and the difference in the payoffs is humongous), we
see crazed behavior that basically tells us without a doubt that a deal was
made. Folks that aren't so familiar with the whole situation have got to
be a little confused. And once again, it lends an air of backroom dealing
and shadiness to poker.

Look, this is *not* hard. Gather the top 90-100 tournament players, and
ask 'em (via a poll, survey, whatever) what they would like to see for
the prize payoff. If they say 40/20/10, fine, *they* want to be hypocrits.
But they just might tell you the truth, which would be a lot flatter
distribution.

Then we could have tournaments that ended with a bang instead of a whimper.

Regards, Lee
--
Lee Jones | "And as I listened, your voice seemed so clear
le...@sgi.com | so calmly you were calling your god"
415-933-3356 | -Loreena McKennitt

Erik Reuter

unread,
Feb 25, 1997, 3:00:00 AM2/25/97
to

In article <5etqhi$r...@fido.asd.sgi.com>, le...@diver.asd.sgi.com (Lee

Jones) wrote:
> What galls me is the complete and total hypocrisy in the prize structures
> that are claimed in the big tournaments. They have this *massive* drop-off
> (like 40/20/10/peanuts) in the pay-offs, but it's clear that most participants
> would much rather reduce the variance, so they make deals.

I'd like to see a tournament where the difference in payout between every
paid place is approximately equal to a buy-in.

Choose the number of paid places as x = IntegerPart( sqrt(2N) - 0.5 )
where N is the number of tournament entries. The top prize will be
200%/(x+1) and the lowest prize (= prize increment) will be 200%/( x(x+1)
).

For example, for a 200 person tournament with $100 buy-in, x = 19 places
paid, and the payout would be

$2000
$1894
$1789
$1684
$1578
$1473
$1368
$1263
$1158
$1052
$947
$842
$736
$631
$526
$421
$315
$210
$105

--
Erik Reuter, e-re...@uiuc.edu

Jeff Woods

unread,
Feb 25, 1997, 3:00:00 AM2/25/97
to

russell rosenblum wrote:
>
> I agree. There are only 2 ways that Phil makes that play. One is if
> there was a deal. And 2 is if he is on MONSTER tilt. To just say that
> it was an aggressive play does not really make sense.

Ok, now, question. Which was it? If it was because of a deal (as
someone says to have heard from Ken Flaton first hand today), then were
his antics during his interview ("No, third place is not acceptable,
I've never finished third in my life") just *advertising* -- an act for
players who might see him and play against him later? If his third
place finish was planned as part of a deal, why be on such a raging tilt
after the fact?


Dan Goldman

unread,
Feb 25, 1997, 3:00:00 AM2/25/97
to

PMFJI, but...

I would NEVER play in a tournament with such a structure. I understand
your intent here, but think of it strictly in money odds terms. 200
players, and if I last until the very end I get 20:1 on my money? The
fact that 20 places are paid and there is some equity is meaningless.

Not meaning to be difficult here - but egalitarian arguments like this
ignore the fact that tournaments are a test of SKILL. The highest
placing player SHOULD get a significantly larger share of the prize
pool. This is the way it is in all major money tournaments with
individual competition (PGA, tennis, etc.)

This is a good opportunity to bring up another point, though. To me,
the real problem with tournament structures is that when you reach the
final table, the limits are frequently so high that the outcome is
largely dictated by chance. I can't tell you how often I have reached
the final table a big leader and taken 2, 3 or 4 place because a
shortchipped player got a single monster hand that put him in the lead -
and I have done the reverse, as well. This is rarely (but occasionally)
the result of skilled play. I don't know what the right way is to deal
with this without making tournaments inordinately long.

Suggestions?

...dan

Steve Brecher

unread,
Feb 25, 1997, 3:00:00 AM2/25/97
to

dgol...@smalltalk.com wrote about Erik Reuter's suggestion for a
tournament payout structure that is a much flatter function of finish
position than those currently in use, i.e., with a slope of one buy-in per
position:

> PMFJI, but...
>
> I would NEVER play in a tournament with such a structure. I understand
> your intent here, but think of it strictly in money odds terms. 200
> players, and if I last until the very end I get 20:1 on my money? The
> fact that 20 places are paid and there is some equity is meaningless.
>
> Not meaning to be difficult here - but egalitarian arguments like this
> ignore the fact that tournaments are a test of SKILL. The highest
> placing player SHOULD get a significantly larger share of the prize
> pool. This is the way it is in all major money tournaments with
> individual competition (PGA, tennis, etc.)
>
> This is a good opportunity to bring up another point, though. To me,
> the real problem with tournament structures is that when you reach the
> final table, the limits are frequently so high that the outcome is
> largely dictated by chance. I can't tell you how often I have reached
> the final table a big leader and taken 2, 3 or 4 place because a
> shortchipped player got a single monster hand that put him in the lead -
> and I have done the reverse, as well. This is rarely (but occasionally)
> the result of skilled play. I don't know what the right way is to deal
> with this without making tournaments inordinately long.
>
> Suggestions?

It's not another point -- it's *the* point. Unlike golf and tennis, luck
is a significant factor in every poker tournament -- and not just at the
final table. While the relative importance of luck increases with the
blinds to average stack ratio, luck is significant throughout.

I can't see a way to derive an objectively best payout structure. There
will always be some grumbling. Erik's proposal is too flat for my taste,
but I do like what Steve Morrow did for the Four Queens Poker Classic last
month. That is, I like taking some off 1st and 2nd, giving 4th-9th a
couple of percent more each, paying at least the buy-in to all paid
places, and in some cases paying an additional table. Actually, the "at
least the buy-in" part I don't care about so much except for what I guess
to be a second-order effect of being appealing to other people and
increasing tournament participation.

--
st...@brecher.reno.nv.us (Steve Brecher)

Erik Reuter

unread,
Feb 25, 1997, 3:00:00 AM2/25/97
to

> I would NEVER play in a tournament with such a structure

....


> I can't tell you how often I have reached
> the final table a big leader and taken 2, 3 or 4 place because a
> shortchipped player got a single monster hand that put him in the lead -
> and I have done the reverse, as well. This is rarely (but occasionally)
> the result of skilled play. I don't know what the right way is to deal
> with this without making tournaments inordinately long.
>
> Suggestions?

Hmmm. I suggest that you reconsider your first statement.

--
Erik Reuter, e-re...@uiuc.edu

Erik Reuter

unread,
Feb 25, 1997, 3:00:00 AM2/25/97
to

In article <steve-25029...@brecher.reno.nv.us>,

st...@brecher.reno.nv.us (Steve Brecher) wrote:
> Erik's proposal is too flat for my taste,

I'm not sure if "flat" translates as "top prize is too low and the prize
increment is too low" or if it means that the distribution of money paid
as a function of place has no super-linear terms.

If the former, note that the number of places paid and top prize can
easily be adjusted up or down. For instance, for an increment of about 4
buy-ins it would be

x = IntegerPart( sqrt(N/2) - 0.5 )

with the other calculations unchanged. For the 200 player example, this
would give x = 9 places paid and a top prize of 20% (40:1) [the 9 places
would pay 20%/17.8%/15.6%/13.3%/11.1%/8.9%/6.7%/4.4%/2.2%]

For the people who don't like the fact that one hand can often make a
difference between 1st and 2nd (or 5th and 6th, or whatever) which would
be a large money difference in a conventional tournament, the structure
above would help to smooth out that variance.

On the other hand, if you are skillful enough to come in 3rd place instead
of 8th place, you have increased your prize by almost a factor of 4, so
there is a reasonable payoff for skillful play.

Also, if the above formula were applied to a $10K buy-in 300 player
tournament, x = 11 places would pay, and the top prize would be $500K [the
payout would be $500K/$454K/$409K/$363K/$318K/$272K/$227K/$181K/$136K/$45K
]

--
Erik Reuter, e-re...@uiuc.edu

russell rosenblum

unread,
Feb 25, 1997, 3:00:00 AM2/25/97
to

> PMFJI, but...
>
> I would NEVER play in a tournament with such a structure. I understand
> your intent here, but think of it strictly in money odds terms. 200
> players, and if I last until the very end I get 20:1 on my money? The
> fact that 20 places are paid and there is some equity is meaningless.
>
> Not meaning to be difficult here - but egalitarian arguments like this
> ignore the fact that tournaments are a test of SKILL. The highest
> placing player SHOULD get a significantly larger share of the prize
> pool. This is the way it is in all major money tournaments with
> individual competition (PGA, tennis, etc.)
>
> This is a good opportunity to bring up another point, though. To me,
> the real problem with tournament structures is that when you reach the
> final table, the limits are frequently so high that the outcome is

> largely dictated by chance. I can't tell you how often I have reached


> the final table a big leader and taken 2, 3 or 4 place because a
> shortchipped player got a single monster hand that put him in the lead -
> and I have done the reverse, as well. This is rarely (but occasionally)
> the result of skilled play. I don't know what the right way is to deal
> with this without making tournaments inordinately long.


I agree on both counts (although they seem contradictory). I think you
could have a breakdown of 33% of the prize money to 1st. 25% to second.
18%to third. 12% to fourth. And divide the remaining as sit. I am not
locked into this exact structure by any means (I put very little thought
into it). I just think that in order to compensate for the arguably small
gap between first and second (third etc..) in terms of "skill" there
should be a little "fairer" distribution of prizes.

How about this: (without exact numbers)

If the tourney has a default prize fund (say as is).

The tourney has an alternate prize fund--That only applies when there are
(for example) five or fewer players remaining. This prize fund has a
closer distribution, and must be agreed upon unanimously by live players
or play continues.

This would protect both sides.
A person who has an ENOURMOUS chip lead (or percieved enourmous skill
advantage, say PH against me at no limit holdem, I may have areasonable
chip lead, and may be willing to take less. PH may still want to play
because he believes he will beat me anyway.) hcan veto the alternate prize
fund and play for the HUGE first prize.

However in a relatively close match the players can vote to stop play and
accept a closer distribution.

(Example if normal prize for last three is 40%,20%,10%, Alternate pool
might be 34%,22%,14%. When there are two players left it could go to (only
60% left) 33%,27% or just play on).

Players could even agree to the alternate pool, and then continue to play
to see who finishes where. They could even be required to continue playing
for slots even after an agreement was reached. I think this would deter
tourneys from being cut short because of deals, but still protect
reasonable players from a massive varience in a close match.

Obvoiusly the numbers would need to be worked on, and could be based on
"historicaly typical deals" Just a thought what do you all think?

russell rosenblum

unread,
Feb 25, 1997, 3:00:00 AM2/25/97
to

Actually thank you. I had not yet read the other posters description of a
deal. I was merely hypothosizing. I do basically agree with you re: PH
very agressive style. I also agree that making the deal may very well have
helped put him on tilt. When I said "got bored" with the tourney I did not
really use the right term. I should have said, PH had just about had it
with the tourney. In other words he looked like he just wanted to finish
it up. I don't think he would have made that same play if they did not
have a deal worked out (obviously I am just guessing). I think the
remaining $65 was peanuts compared to the frustration he was feeling, and
he just wanted to "put an end" to the tourney NOW, instead of grinding out
a win.

By the way I do think the deal as it was structured was fair. And Ken
still deserves a hell of a big congratulations for his victory. He really
made a hell of a comeback just getting to a point where PH would make a
deal at all.

russell rosenblum

unread,
Feb 25, 1997, 3:00:00 AM2/25/97
to

From what I read your comments on the 4-queens is accurate. By the way I
do agree with you. I would rather see a closer distribution of the prize
money, than a bunch of deal-making "when the tourney should be exiting."
One of the comments made by some of the pros against changing the prize
distribution was that "negotiating skills are a part of your tourney
skills." In other words, one more chance to take advantage of the
inexperienced players.

If you had 3 pros at a table they may always agree to a more reasonable
breakdown anyway. Deal making allows them to "run over" the inexperianced
tourney player come deal time. I would MUCH rather see a more equal
distribution amongst X finalists. If not then it is crazy not to allow
deals.

Look at the taj tourney first place was $500,000. What was second about
$100,000? I would make a deal to.

As far as comparing it to other sports, In the super bowl the difference
between the winners share and the losers share is not Five times. In
Tennis (say wimbledon) I don't think it is that large. In boxing winning
or losing rarely matters in terms of $$$. Imagine if all boxing matches
were say "winner takes all," I am sure the boxers managers would be mighty
tempted to "get together" between rounds and "work something out" in a
close match.

Imagine if you went to work everyday and were told that your work was
excellent, but it was the "second best" so you were only going to make 20%
of what the next best person (even marginally better) made for the same
job.

I understand that some of that is the nature of being a professional
gambler. Do you think that negotiating skill should be part of the tourney
skills? I am mixed. I would probably prefer a more just distribution and
less deal making, although I don't consider deal making to be cheating in
any way, but then I also think you should be allowed to turn your cards
face up on the table, judge an opponents expression and decide whether or
not to call.

Oh yea, and by your post (lee) I assume you do believe that PH made a
deal, and did not go on major tilt. (I agree)

Michael Zimmers

unread,
Feb 26, 1997, 3:00:00 AM2/26/97
to

In article <5etqhi$r...@fido.asd.sgi.com>, le...@diver.asd.sgi.com (Lee
Jones) wrote:

> What galls me is the complete and total hypocrisy in the prize structures
> that are claimed in the big tournaments. They have this *massive* drop-off
> (like 40/20/10/peanuts) in the pay-offs, but it's clear that most participants
> would much rather reduce the variance, so they make deals.

> [...]


> Look, this is *not* hard. Gather the top 90-100 tournament players, and
> ask 'em (via a poll, survey, whatever) what they would like to see for
> the prize payoff. If they say 40/20/10, fine, *they* want to be hypocrits.
> But they just might tell you the truth, which would be a lot flatter
> distribution.

Another possibility is that the major tournament players like the current
structure, where they can manipulate the payout to their advantage by
negotiating (or denying) a deal. This goes beyond the mere getting-the-
best-of-it in a particular prize pool cutup, too. Imagine, for example, a
final table containing several veteran superstars, and a couple of streakers
who are clearly out of their depth. With an altered prize payout like Lee
suggests, the luckyboys will get a bigger share of the prize money. If,
however, the current payout structure is in place, the vets can wait for
a chance to knock out one or more of the luckyboys, *then* make a deal
once they feel everyone's stack size more or less accurately reflects
their expected equity in the prize money.

Food for thought...

--
Michael Zimmers
SoftHelp -- Solutions for Open Systems
Voice: 408 996 1965
Fax: 408 996 1974

Michael Zimmers

unread,
Feb 26, 1997, 3:00:00 AM2/26/97
to

Lee Jones

unread,
Feb 26, 1997, 3:00:00 AM2/26/97
to

In article <mzimmers-250...@mzimmers.vip.best.com>,
Michael Zimmers <mzim...@best.com> wrote:

>Imagine, for example, a
>final table containing several veteran superstars, and a couple of streakers
>who are clearly out of their depth. With an altered prize payout like Lee
>suggests, the luckyboys will get a bigger share of the prize money. If,
>however, the current payout structure is in place, the vets can wait for
>a chance to knock out one or more of the luckyboys, *then* make a deal
>once they feel everyone's stack size more or less accurately reflects
>their expected equity in the prize money.

This is actually the most plausible explanation I have heard yet.
Interestingly enough, I doubt many of the veteran superstars could express
it so eloquently, but I think that's what they're thinking. "Lee Jones -
shee-it, we can eat him up like yesterday's lasagna, and *then* split up the
serious prize money."

This, by the way, quickly moves you to thinking about implicit (or explicit)
agreements among the tourney vets to crush any streakers that somehow sneak
into the final table. Knowing, as we do, that many of those folks are playing
on each other's bankrolls, have pieces of each other, have made saves between
themselves, etc., you start to wonder if they don't have a significant
motivation to squeeze out anybody who isn't part of their financial network.

More food for thought.

Regards, Lee
--
Lee Jones | "Look out now - oh look out!
le...@sgi.com | She came in through the bathroom window..."
415-933-3356 | -Lennon/McCartney

Jazbo Burns

unread,
Feb 26, 1997, 3:00:00 AM2/26/97
to

Interesting thread you've started, Lee. I too would prefer a flatter
prize structure since I think it would more accurately reflect the
skills of the players (especially near the end when the blinds get out
of proportion). However, I doubt we will see any significant changes,
both for the reasons that Michael Zimmers expressed so well about
implicit collusion (or maybe not so implicit) by the "regulars" at the
end of the big tournaments and another that I think may have even
greater weight.

In the big tournaments, the casinos get a lot of play by advertising
the size of the top prize. Remember the WSOP had a $1M payout long
before any reasonable payout schedule would justify it. Even in
smaller tournaments (like the Gold Coast, or is it Orleans now), there
are many reasons the casinos want to be able to advertise a big
windfall for the overall winner. One of main ones is that it attracts
players, especially players that have a lottery mentality. One could
argue that this is one of the main reasons for the tremendous growth
of tournaments over the last ten years. I don't think casinos are
likely to change a winning formula.

Stephen H. Landrum

unread,
Feb 27, 1997, 3:00:00 AM2/27/97
to

Jazbo Burns wrote:
>
> Interesting thread you've started, Lee. I too would prefer a flatter
> prize structure since I think it would more accurately reflect the
> skills of the players (especially near the end when the blinds get out
> of proportion).

Perhaps. I had thought this way myself for a while.

But, there are significant changes in the tournament that are not
directly obvious when the payout structure is changed. In a
winner-takes-all tournament, every chip in your stack is worth exactly
its percentage of all of the chips in play in the tournament.
When multiple places are paid, the value of the first chip you hold
increases, and the value of each subsequent chip is less.

A tournament that pays out many places and with a fairly flat payout
structure will probably after analysis have some very non-intuitive
strategic elements. Current tournaments already have this, but I
think that by taking more money from first place and distributing it
down the line you increase the deviation from "normal" poker
strategies.

For instance, at the extreme a tournament of N players that pays N-1
places equally or almost equally, the only strategy that should be used
is survival. It is likely that every hand should be folded, including
pocket rockets, in the hopes that someone else makes the mistake of
playing a hand.

> In the big tournaments, the casinos get a lot of play by advertising
> the size of the top prize. Remember the WSOP had a $1M payout long
> before any reasonable payout schedule would justify it. Even in
> smaller tournaments (like the Gold Coast, or is it Orleans now), there
> are many reasons the casinos want to be able to advertise a big
> windfall for the overall winner. One of main ones is that it attracts
> players, especially players that have a lottery mentality. One could
> argue that this is one of the main reasons for the tremendous growth
> of tournaments over the last ten years. I don't think casinos are
> likely to change a winning formula.

I'm also not likely to play in a tournament where the payouts of the
first places are not many multiples of the buy-in. If I'm interested
in flat payout structures, I can play in ring games.
--
"Stephen H. Landrum" <slan...@pacbell.net>

Dan Goldman

unread,
Feb 27, 1997, 3:00:00 AM2/27/97
to

Erik Reuter wrote:
>
> In article <33132B...@smalltalk.com>, dgol...@smalltalk.com wrote:
>
> > I would NEVER play in a tournament with such a structure
> ....

> > I can't tell you how often I have reached
> > the final table a big leader and taken 2, 3 or 4 place because a
> > shortchipped player got a single monster hand that put him in the lead -
> > and I have done the reverse, as well. This is rarely (but occasionally)
> > the result of skilled play. I don't know what the right way is to deal
> > with this without making tournaments inordinately long.
> >
> > Suggestions?
>
> Hmmm. I suggest that you reconsider your first statement.
>
> --
> Erik Reuter, e-re...@uiuc.edu

Erik:

Pardon me, but what the hell are you talking about?

You excised an important part of my message, which deals with FINAL
TABLE PLAY. Your structure simply doesn't provide the money odds to
make a tournament interesting...I would take the same money and play in
a live game.

My point was that the combination of a wide variance among place money
and rapid escalation of blinds makes luck a major factor at the final
table. It has little to do with early play. The fact that the
recognized names regularly come into the money tells me that luck is a
factor only in the short run. But the fact that recognized names don't
take a larger share of TOP money indicates that, while they can use
their skills in the first nn hours of a tournament to reach the money,
luck takes over at the final table. Tom McEvoy says something to the
effect that he considers the best players to be only a 2% or so favorite
in a given tournament. This says a lot about luck in the short run.

...dan

Dan Goldman

unread,
Feb 27, 1997, 3:00:00 AM2/27/97
to

One thing that we have done a number of times at the Hollywood Park
tournaments, which bears some similarity to what you're described:

Let's say the pool is $5000, and what is left at 5 players is
distributed like this:

1 2000
2 1000
3 500
4 400
5 300

...and everyone is reasonably close; that is, there isn't a runaway
chip leader. There is $4200 remaining in the prize pool. We will often
agree to take $500 each and play for the remaining $1700 on a 50/30/20
split. This assures everyone of 3rd place money and softens the luck
factor a bit. This would make the new distribution as follows:

1 500 + 850 1350
2 500 + 510 1010
3 500 + 340 840
4 500
5 500

much more equitable.

...dan

Dan Goldman

unread,
Feb 27, 1997, 3:00:00 AM2/27/97
to

Michael Zimmers wrote:
>
> Another possibility is that the major tournament players like the current
> structure, where they can manipulate the payout to their advantage by
> negotiating (or denying) a deal. This goes beyond the mere getting-the-
> best-of-it in a particular prize pool cutup, too. Imagine, for example, a

> final table containing several veteran superstars, and a couple of streakers
> who are clearly out of their depth. With an altered prize payout like Lee
> suggests, the luckyboys will get a bigger share of the prize money. If,
> however, the current payout structure is in place, the vets can wait for
> a chance to knock out one or more of the luckyboys, *then* make a deal
> once they feel everyone's stack size more or less accurately reflects
> their expected equity in the prize money.
>

This happened to me in the first tournament in which I placed in the
money. It was not a huge tournament; the pool was about $6000. We
were down to 5 players, and I was clearly the least experienced of the
group. One player had the lead, I was second, another was pretty close
behind me and the other two were distant. We took a 10 minute break.
When we returned, I noticed that players 1 and 3 were NEVER in the same
hand. We got down to three players, and we were very close to even. I
offered a 3 way split and was immediately turned down. It then became
clear to me that the other two had agreed on a split and were making
sure I finished third. Sure enough, I did.

Fortunately, this came back on the guy in first place almost
immediately. Two weeks later I found myself at the final table with a
small lead and four players. This guy offered a deal, the others were
interested but I refused. We eliminated the other two and took a break
before playing heads-up. The pool for this one was about $5000, so
first was $2000 and second was $1000. I had about 60% of the chips. He
offered me a reasonable deal; $1600/$1400 split. I knew I was the
better heads-up player, although this advantage was marginal. I refused
a deal and pressed him every hand, finally winning with 6-6 to his A-Q
suited. It was a classic final hand - flop came A-Q-6 with the 6 of
spades. He went all in, of course. The turn was a spade, giving him 9
more outs...but the river was the fourth 6.

All of life should be like this :D

...dan

Lee Jones

unread,
Feb 27, 1997, 3:00:00 AM2/27/97
to

In article <3315E9...@smalltalk.com>,
Dan Goldman <dgol...@smalltalk.com> wrote:

>This happened to me in the first tournament in which I placed in the
>money. It was not a huge tournament; the pool was about $6000. We
>were down to 5 players, and I was clearly the least experienced of the
>group. One player had the lead, I was second, another was pretty close
>behind me and the other two were distant. We took a 10 minute break.
>When we returned, I noticed that players 1 and 3 were NEVER in the same
>hand. We got down to three players, and we were very close to even. I
>offered a 3 way split and was immediately turned down. It then became
>clear to me that the other two had agreed on a split and were making
>sure I finished third. Sure enough, I did.

This is *exactly* the sort of thing that makes me want to play in ring
games. I suspect that this kind of collusion takes place on a regular
basis, sometimes less overt, sometimes more overt. I once heard a floorman
at Bay 101 *bragging* about he and another guy did that to a third person
who had refused a split.

Perhaps we deserve the smoky back-room under-the-table deal-making reputation
that we have.

Regards, Lee
--
Lee Jones | "Sometimes life is a bitter pill -
le...@sgi.com | I love you now, do you love me still?"
415-933-3356 | -Chaka Kahn/Bruce Hornsby

Erik Reuter

unread,
Feb 27, 1997, 3:00:00 AM2/27/97
to

> Pardon me, but what the hell are you talking about?

No pardon necessary. The hell that I am talking about is that we can't
have it both ways. You asked for suggestions on what to do about the fact
that luck makes a big difference between finishing 1st and 2nd (or 2nd and
3rd, etc.), and I suggested (somewhat wryly) that the way to deal with
this is to have a payout structure similar to what I proposed where the
prize money difference between 1st and 2nd isn't as great as in
conventional tournaments. If we use such a structure but keep the same
number of places paid, that results in a lower top prize, which you state
you dislike.

> Your structure simply doesn't provide the money odds to
> make a tournament interesting...I would take the same money and play in
> a live game.

A perfectly reasonable opinion, and one I expect others share. However,
note that a variation on the structure I suggested could still give as
high a payout for first as desired, it would simply pay fewer places.

For instance, we could pay the top 4 places in a 200 player tourney 80/60/40/20.
Or the top 5 places in a 300 player tourney 100/80/60/40/20. The main
point of my proposed payout structure, which I may not have adequately
communicated earlier, is that the difference between payouts is linear.
The formula for top prize (in units of buy-ins) is 2N/(x+1) where N is the
number of entries and x is the number of places paid. The prize increment
is just the top prize divided by x.

> My point was that the combination of a wide variance among place money
> and rapid escalation of blinds makes luck a major factor at the final
> table. It has little to do with early play. The fact that the
> recognized names regularly come into the money tells me that luck is a
> factor only in the short run. But the fact that recognized names don't
> take a larger share of TOP money indicates that, while they can use
> their skills in the first nn hours of a tournament to reach the money,
> luck takes over at the final table. Tom McEvoy says something to the
> effect that he considers the best players to be only a 2% or so favorite
> in a given tournament. This says a lot about luck in the short run.

In other words, a great tournament player may make it to one of the top x
places with high probability (compared to typical players), but
considering only the set of players who make it to the top x places, the
probability of the great tournament player finishing 1st isn't much
greater than any typical player finishing 1st.

This would seem to support my proposed linear payout structure, given that
an x can be chosen such that the above paragraph is true.

--
Erik Reuter, e-re...@uiuc.edu

russell rosenblum

unread,
Feb 27, 1997, 3:00:00 AM2/27/97
to

Oh i don't think the deal would include a "who finishes where". Just
dollars. i think Phil was genuinly upset that he did not take down the
$500,000 and had to make a deal. What would have been funny as hell
though, is if after the deal was made, phil made the the exact same play
at the same time, BUT hit the inside straight. If he then wins the
tourney, how the hell does he then explain being really pissed about a
deal that was "never actually made."

Dan Goldman

unread,
Feb 28, 1997, 3:00:00 AM2/28/97
to

Erik Reuter wrote:
>
> In article <3315E7...@smalltalk.com>, dgol...@smalltalk.com wrote:
>

> For instance, we could pay the top 4 places in a 200 player tourney 80/60/40/20.
> Or the top 5 places in a 300 player tourney 100/80/60/40/20. The main
> point of my proposed payout structure, which I may not have adequately
> communicated earlier, is that the difference between payouts is linear.

OK, I think I understand now, and I now disagree for a different reason
:D

Your solution does deal with the issue of late-tourney luck in a
reasonable way. However, I suspect it would permanently kill
tournaments, because not enough average players would have a chance to
finish in the money. Someone made an analogy to tennis and golf
earlier, and there is something we should remember. NONE of us would
enter the Masters or the US Open. Poker tournaments are sort of the
ultimate egalitarian competition in that anyone willing to risk the
money can step up and take a chance. Fact is, this is how *I* got
started in tournaments. I heard about one at the Commerce about 5 years
ago and decided to give it a shot. I got lucky and got into the money,
and I have been doing it ever since.

But if players like me are routinely clobbered in tournaments, we will
give up. If out of 100 players only 4 come in the money, too many of
the others won't try again.

> > My point was that the combination of a wide variance among place money
> > and rapid escalation of blinds makes luck a major factor at the final
> > table.
>

> In other words, a great tournament player may make it to one of the top x
> places with high probability (compared to typical players), but
> considering only the set of players who make it to the top x places, the
> probability of the great tournament player finishing 1st isn't much
> greater than any typical player finishing 1st.
>
> This would seem to support my proposed linear payout structure, given that
> an x can be chosen such that the above paragraph is true.
>

Right. This would seem to support a linear structure for FINAL TABLE
finishes, I think...but I would like to think on it a bit more. :)

...dan

Dan Goldman

unread,
Feb 28, 1997, 3:00:00 AM2/28/97
to

Steve:

>> When multiple places are paid, the value of the first chip you hold
increases, and the value of each subsequent chip is less. <<

Right...except it's actually the LAST chip that's the most valuable :D

...dan

Stephen H. Landrum

unread,
Feb 28, 1997, 3:00:00 AM2/28/97
to

I think of my stack of chips as a LIFO stack, so the first chip is the
most important because it's the last chip I'll lose. The last chip
in my stack is the first one to be tossed into the next pot I
participate in, so it's the least important one.

My programmer backrground slips through again :)

Dan Goldman

unread,
Mar 3, 1997, 3:00:00 AM3/3/97
to

Stephen:

>
> My programmer backrground slips through again :)
> --

You geek :D

(I can say that; I'm a geek too.)

...dan

0 new messages