Tricky Outs counting??

[WarGawd]

Just when I think I get this perfectly str8 in my head, something comes up to make me question it yet again. I did a forum search, couldn't find anything specifically related, and it seems like this is the best place to pose the question. Here's the problem:

Every week I read the Cardplayer online tournament coverage. Last week I noticed the author had made a minor error in his coverage of the Stars Sunday Million. I wrote him a response, and he replied, but it seems I was only half right.

Player A had pushed all in preflop holding AsQd
Player B had called holding Qc5d
Flop was 7d 6d 4d

The author neglected to consider B's OESFD, and stated that B had "a 35% chance of survival with any non diamond 8 or 3"
I pointed out that the 8d or 3d would complete the str8 flush, plus any of the 3 remaining 5's would give B the win. Using the "Rule of 4" I approximated B's 11 outs would give him a 44% chance of survival. And here's the problem...running those cards through the Cardplayer and ITH odds calculators, I see the 35% is correct.

What gives? Is it the fact that B's 11 outs must be "discounted" to account for the fact that sometimes when B hits his 5, A also hits an ace or a diamond for the flush to win (not 8 or 3)? If this is why the "Rule of 4" breaks down, how does a player modify his outs counting on the fly? 35% translates to ~ 9 outs on the flop...is an apparent discount of 2 full outs warranted or expected?

Someone more experienced please offer your take on this
Thx
WarGawd

[swiggidy]

First off, the rule of 4 slightly over-estimates with more than 10 outs. 11 outs it actually 42%.

If you just consider the 5, which comes 13% of the time. An A will come 6% of the time (3 outs, one chance), plus if a non-diamond 3 or 8 comes you split the pot (about 1%).

As a rule of thumb you can't count the 5s as an out at the table. You may already be behind an overpair, if not the 5 could be counterfeited. Plus, would you really be happy investing more money with Q5 on a 4567 board? If you knew villain had missed overs I would count the 3 fives as 1/2 out each.

What you should take from this is identifying live outs. That's more important than worrying about the exact odds in a given specific matchup where you already know both hands (which never happens in practice).

[WarGawd]

If you just consider the 5, which comes 13% of the time. An A will come 6% of the time (3 outs, one chance), plus if a non-diamond 3 or 8 comes you split the pot (about 1%).

I've re-read that about 9 times, convinced you made an error somewhere...only Player B holds the str8 draw...a non-diamond 8 or 3 gives B the win, not a chop??? Not sure that affects any conclusions drawn below though.

As a rule of thumb you can't count the 5s as an out at the table. You may already be behind an overpair, if not the 5 could be counterfeited.

I agree...this was on FT of (actually UB 200k guarantee) tourney, so players could be pushing any 2. As such, I suppose B really shouldn't use the 5's as legitimate outs in estimating his odds to win, as A's range is likely very wide.

Plus, would you really be happy investing more money with Q5 on a 4567 board?

This was a specific question about this situation...A is already all-in, B has called, and I was wondering about the accuracy of the statement that B had a 35% chance of tourney survival...mentally I had arrived @ 44% using the approximation given by rule of 4. If this was a cash game, I highly doubt I'd be playing Q5 offsuit in the 1st place except as a bluff or a steal.

If you knew villain had missed overs I would count the 3 fives as 1/2 out each.).

And I think this is really what I was digging at...is it just experience that says you'd discount the 5's to 1/2 out each? Is there any more formal publicized guidance out there?

What you should take from this is identifying live outs. That's more important than worrying about the exact odds in a given specific matchup where you already know both hands (which never happens in practice).

Sure, of course For the sake of discussion just for the moment let's just step back to where B is contemplating calling A's all-in push. Should B discount the diamond outs (because A could have the higher flush draw) as well as the 5's to quickly guesstimate his win % as say 7 outs/~28%...or does it even matter since B is probably less than a 2:1 dog in 3-way play on the FT for a difference of probably $25k between 1st & 3rd...I think I call there regardless.

Thanx for your feedback
WarGawd

[swiggidy]

If you just consider the 5, which comes 13% of the time. An A will come 6% of the time (3 outs, one chance), plus if a non-diamond 3 or 8 comes you split the pot (about 1%).

I've re-read that about 9 times, convinced you made an error somewhere...only Player B holds the str8 draw...a non-diamond 8 or 3 gives B the win, not a chop??? Not sure that affects any conclusions drawn below though.

If the 5 comes on the turn then there are six cards that could come on the river which result in a split pot.

If you knew villain had missed overs I would count the 3 fives as 1/2 out each.).

And I think this is really what I was digging at...is it just experience that says you'd discount the 5's to 1/2 out each? Is there any more formal publicized guidance out there?

This is kind of what I was getting at above. If you analyze the situation you could determine the exact equity of a 5 by counting the hands you win vs the hands you lose.

As far as I know there is no formal guideline. This is probably because it doesn't matter that much. A five on the turn gives you 75% equity, so maybe it should be counted as 2.25 outs instead of 1.5. This extra fraction of an out increases your chances of winning by 1.5% (using the 2/4 rule). Very few (if any) decisions at the table are going to be swayed by 1.5%. If this pushes you over the boundary between folding and calling then the wrong decision is only a very small mistake.

The 2/4 rule was designed for situations where if you hit you're a sure win, if you miss you're a sure loss. When villain has redraws calculating percentages becomes more complicated. This guy has posted a lot of odds calculations, you should be able to figure it out if you want:
http://www.math.sfu.ca/~alspach/

For the sake of discussion just for the moment let's just step back to where B is contemplating calling A's all-in push. Should B discount the diamond outs (because A could have the higher flush draw)

Read dependent. If you're in a $3 tourney and villain is a donk I may count the flush outs (discounted more heavily, like 1/4). Here I would discount it based upon his range and how likely he is without a diamond. I've played some tourney's where people push TT or A7 with no diamond. If it's mid stakes I would probably not count the diamonds. If it was higher and person was likely making a move I might count them again (discounted).

Thanx for your feedback
WarGawd

no prob, and welcome to FTR

I don't want to discourage you from analyzing the game deeper. I did the same thing when I started. It may help your understanding which may indirectly help you at the table. But being able to do this exact calculation won't really help your game (which is why I'm reluctant to put more thought into it ).

[WarGawd]

[If the 5 comes on the turn then there are six cards that could come on the river which result in a split pot.

Ahhhhhh

A five on the turn gives you 75% equity, so maybe it should be counted as 2.25 outs instead of 1.5.

I assume you use some type of calculator for this? You don't do these in your head do you?

I don't want to discourage you from analyzing the game deeper. I did the same thing when I started. It may help your understanding which may indirectly help you at the table. But being able to do this exact calculation won't really help your game (which is why I'm reluctant to put more thought into it ).

I just think that the deeper true understanding I have, the easier it will be to decide for myself what's really important in any given situation.
Much appreciate the link and your time.

Cheers
WarGawd

[swiggidy]

I assume you use some type of calculator for this? You don't do these in your head do you?

I knew I forgot something. Pokerstove is free. Google and download it, you'll likey.