PT Misc Stats: High Card & One Pair

[arkady]

I think I may have found another fundamental difference between marginal winners and big winners. This in my opinion is directly related to high showdown value, because to have the smallest loss possible in these two categories one requires a higher SD.

I don't see a meaningful %, but I am going to go w/sample size and if you guys could share yours, I might be able to confirm my theory.

This is for 10/20 only. 23k hands. VPIP: 20.71

High Card: Went to SD: 176/1.5k - Money lost: <font color="red"> -26.6k </font> - hands won $: 273 - won $ at sd: 22.73%

One Pair: Went to SD: 767/2.6k - Money lost <font color="red"> -20.9K </font> - hands won $: 912 - won $ at sd: 47.59%

To make sense of some these numbers and what I think they should be versus what they are right now.

I think my went to SD is too low. I suspect I am folding way too many winners with middle or even low pair. Not snapping off bluffs w/A or even K high.

Hands won $ is probably too high. I am not actually even sure what this number means, but I have a feeling its way too high especially for High Card.

Won at SD%. As usual in my case, this number is just too high. Seems like if I am going to insist on going to SD ONLY with winners, I will simply be folding the winner too often. This is a reflection of my game as a whole, but it is profoundly apparent in these two categories.

Overall obtaining a high SD seems to be the most challenging hump in my poker game to date. ____1____ made a post recently that his SD is too low overall. Which indicates to me that this is definitely a difficult aspect to deal with. Because it trascends something one can teach and requires a poker sense that maybe some simply do not have. Or at the very least requires a different perspective to obtain.

Getting back to the original point. There are obviously more factors to consider, including one's VPIP, but I would be interested to hear some thoughts and compare numbers.

[krishanleong]

Alobar 5/10 - 80K overall sample

High card - 651 out of 1655 (39.3%) with W$@SD 17.7
One pair - 3198 out of 5533 (57.8%) with W$@SD 42.3

Joseki - 100K overall sample

High card - 903 out of 1858 (48.6%) with W$@SD 21.1
One pair - 4070 out of 6582 (61.2%) with W$@SD 46.6

Max - 61K overall sample

High card - 674 out of 1372 (49.1%) with W$@SD 18.1
One pair - 2603 out of 4517 (57.6%) with W$@SD 44.3

Arkady

High card - 176 out of ???? with W$@SD 22.7
One pair - 767 out of ???? with W$@SD 47.6

Some data from earlier stats post so we have something to work with. Incidentaly this is one way I try and make myself feel better when I'm in a downswing. See if my 1 pair hands are underperforming at SD. /images/graemlins/laugh.gif

Like WTSD and W$@SD, I think the W$ASD for 1 pair hands depends on the frequency that you take them to showdown.

Also the variance on these numbers is going to be huge dues to sample size issues.

Krishan

Link to original data (http://forumserver.twoplustwo.com/showflat.php?Cat=&amp;Board=headsup&amp;Number=1883174&amp;fpa rt=&amp;PHPSESSID=)

[adios]

FWIW I'm going to highlight some of your comments and if you feel I'm taking them out of context apologies in advance as I don't mean to do that.

[ QUOTE ]
I think my went to SD is too low. I suspect I am folding way too many winners with middle or even low pair. Not snapping off bluffs w/A or even K high.

Hands won $ is probably too high. I am not actually even sure what this number means, but I have a feeling its way too high especially for High Card.

[/ QUOTE ]

Not necessarily IMO. If this was a result of superior hand reading ability then I'd maintain that your statement is absolutely wrong. Mason's pointed out many times over the years that the relationship between your win rate and standard deviation is the critical factor. Mason's also pointed out over the years that as one gets more adept at reading hands one's standard deviation should get lower and the top players actually do have quite low standard deviations.

[ QUOTE ]
Won at SD%. As usual in my case, this number is just too high. Seems like if I am going to insist on going to SD ONLY with winners, I will simply be folding the winner too often. This is a reflection of my game as a whole, but it is profoundly apparent in these two categories.

[/ QUOTE ]

I think I know what you mean and thus I think you'd want to amend this statement slightly. You meant to say that you're going to the showdown mostly when you know the chances of you having the winner are far greater than what the pot is laying you. Evaluating when it's close to break even is the problem for you to put it another way.

[ QUOTE ]
Overall obtaining a high SD seems to be the most challenging hump in my poker game to date. ____1____ made a post recently that his SD is too low overall.

[/ QUOTE ]

I'd reiterate that it's the relationship of win rate to standard deviation that's important and that superior hand reading skills (I believe Mason's correct) tend to lower one's standard deviation.

[ QUOTE ]
Which indicates to me that this is definitely a difficult aspect to deal with. Because it trascends something one can teach and requires a poker sense that maybe some simply do not have. Or at the very least requires a different perspective to obtain.

[/ QUOTE ]

Well I agree and I think what's important is evaluating the EV implications.

[Silverback]

[ QUOTE ]
I think I may have found another fundamental difference between marginal winners and big winners.

[/ QUOTE ]

Can I just ask, what do you consider a marginal winner and a big winner, what bb/100?

[arkady]

Marginal: .8-1.2 bb / 100
Good: 1.8 and up

Problem with these win rates that even within a 100k sample you might be off by 1 bb. But I will just use that rough rule of thumb for now...from experience.

[arkady]

I believe that you may have gotten SD confused with Standard Deviation, while my entire post was concentrating on ShowDown.

Sorry for being unclear....:(

[arkady]

It is exactly the frequency of going to showdown that to me is the difference. Better players go there more often.

How they get there and why they do it is probably the key question.

[krishanleong]

Are you sure it's with high card and 1 pair hands?

Post the total hands and come up with a % so we can determine if it's in the range of what's expected or way out of whack.

Krishan

[cartman]

Here are my numbers in krishanleong's format:

High card 494 out of 1127 (43.8%), W$@SD 16.4%
One Pair 2184 out of 3638(60.0%), W$@SD 41.4%

[arkady]

I am sure, I can take a screen shot I suppose to make it easier. I can already see that something is not right...the last numbers are too high. Most people are about 40-42 mine is 47+ /images/graemlins/frown.gif

wish i knew why.

[cartman]

I think there is some confusion about this statistic (maybe I am the one confused). When you click the "show only hands that were not folded" box, I think the "Went to SD" does not include the hands that you took down the pot because everyone else folded. If this is the case, then we would want it to be as small as possible.

In other words, "Went to SD" with the box checked doesn't mean "Didn't fold". Instead it means "Didn't fold and at least one opponent didn't fold either".

To get a decent handle on who does the best with their high card hands, I think you would have to look at a statistic such as:

1) Hands won $ / Total Times
2) Hands won $ / Saw Flop
3) Net Amount Lost / Total Times (I think this one is best)

I may be way out in left field here, so please correct me if I'm crazy. My main point is that, if you check the box, "Went to SD" is a BAD thing. That means you were unable to take down the pot by making everyone fold therefore you had to sheepishly turn over your measly high card hand when you both checked on the river (or even worse after you bet and your opponent called or after he bet and you called!)

What do you think?

Cartman

[arkady]

To clarify any confusion, here is an image. I presume if I fix those two and restore my BB loss to .15 or so, I would be in much better shape.

http://www.computerecycle.com/images/PtSCreen.jpg

[___1___]

Through 20k 10/20 hands (VPIP: 23.8%)

High Card: Went to SD: 149/1591 - Money lost: -25.23k - hands won $: 259 - won $ at sd: 24.83

One Pair: Wento to SD: 700/2387 - Money lost: -12.57k - hands won $: 893 - won $ at sd: 45.29%

___1___

[arkady]

anyone see a glaring issue w/going to sd?

[Interested]

I have one question - Did you separate cases with what number of players saw flop or the question was about HU?
Any differences with kind of pair(Button, Med, High) you go to SD with that number of players saw flop?
My sample size is too low(4K). But for my full tables I made some PT queries, then exported them to excell, then macros find out:
- if I saw the flop with that number of opponents and I have mid/but/high pair - how much I net/won, % of the time.
I think if to change the query and add Went To SD, the same could be finded. And may be a little improve the stats/results.

Sorry for my crappy English

[cartman]

I didn't mean I was confused about which numbers you were talking about. I think you may be confused (maybe I am instead) about what "went to SD" means. Reread my post above and maybe you can either correct my supposition or, if I am, correct, redirect your investigation. Having a low "Went to SD" doesn't necessarily mean that you are folding too much. It may mean that players are folding to your bets or raises and you are winning pots uncontested, which would be a very good thing. I will think about this some more, but do you understand what I am saying about what the definition of "went to SD" appears to be? It looks to me like it means "you went to SD and so did at least one opponent". Maybe the number you are actually trying to investigate is "I went to SD" + "I won because everyone folded". This would just be the "total times" column when the "show only hands that were not folded" button is checked. I may be completely missing the point. I'm just trying to be helpful.

Cartman

[krishanleong]

Arkady,

Click the show only hands that are not folded and repost your misc tab please.

Krishan

[arkady]

cartman asked me to get back to this thread...so here i am.

what exactly is the purpose of filtering it that way? i will do it though once i get a chance.

[arkady]

show only hands that were not folded may be a start to filter out the situation that you are trying to isolate. but would that also not throw everything due to the pre-flop change? maybe i am folding pre flop too much! maybe thats where the problem is? I for one know that in 10/20 my BB vpip is extremely low. Perhaps because the 10/20 pfr aggression is so much higher that I do no get a free look as often. But in reality I think checking this box will skew everything off. Something not natural about it.

[arkady]

This is over 27.5k hands and VPIP 21.

http://www.computerecycle.com/images/NFmiscTab.jpg

[J.R.]

I don't get this, its the opposite of what u want. You want to identify the hands you are folding that you shouldn't fold. After a session/day I usually filter my hands on the general info page for folded to a river bet. Sure there is stuff you might have folded incorrectly on the turn but looking into folded to a river bet is a start.

[cartman]

I'm not sure what krishanleong's intention was for this request, but I'm sure he had one. I just wanted to see the discussion continue. Arkady had uncovered an area where he thought he differed quite a bit from the players who won substantially more than him. I thought (and still do think) that the pursuit was a bit misdirected because of our failure to understand the definition of "went to SD" in the Misc. Stats tab. I explain where I think the misconception is in a couple of my responses above. But once we clear it up, there may be some important things to learn. Does my definition appear to be accurate?:

"went to SD"= "I didn't fold AND at least one other player didn't fold"

I think the statistic that arkady is intending to compare would be a "% of the time didn't fold high card" instead of "went to SD %". We can find this by checking the "show only hands that weren't folded" box and then looking in the "times" column (which is the numerator, 476 for arkady's high card category). One possible denominator is the "times" column with the box unchecked (5488 for arkady's high card category). This fraction (476/5488=8.67%) would give us the percent of his total high card hands that he either won uncontested or showed down vs at least one opponent. Another possible denominator is the "saw flop" column with the box unchecked (1570 for arkady's high card category). This fraction (476/1570=30.32%) would give us the percent of his total high card hands that he either won contested or showed down vs at least one opponent GIVEN that he saw the flop. I think one of these fractions is probably what arkady meant when he talked about "going to the showdown" too much or too little.

What do you guys think?

Cartman

[arkady]

Cartman,

Hey that's pretty good...although I was initially reluctact to post that screen due to exactly what JR said. I think I am just folding too many potential winners. However your numbers makes sense and something I would like to compare to others in 10/20.

rkd