w$sd is a worthless stat (semi-brag)

[JanelleBB7]

My show down # is only 25.92% do I fold too often when I have the winner? I think I certainly do -- not as much now though as I did in the beginning. Grrr I don't do well when I am out of position and I end up folding when I may at times have a winner.

[iggymcfly]

We're talking about W$SD (won money at showdown), not WTSD. 25.9's a perfectly reasonable number for WTSD.

[JanelleBB7]

Dur i see now mine is 43.97% so I am doing fine. Ty

[Troll_Inc]

[ QUOTE ]
[ QUOTE ]
[ QUOTE ]
if your W$SD is too high you're probably giving up value by not semi-bluffing/looking people up enough.

[/ QUOTE ]
if you can tell this from someone's W$SD, how is it a worthless stat?

[/ QUOTE ]

oops

[/ QUOTE ]

Isnt it super easy to just download all the stats of people that you have 5,000+ hands of and see if there is a correlation to winrate?

I think that your thread title will end up being w$sd is a non-correlative (to winrate) stat.

[iggymcfly]

There's a definite correlation. It's just that one doesn't necessarily increase with the other. For instance, I'd guess that players who's W$SD is between say 44 and 54 have much higher winrates than players who's W$SD is outside of those bounds.

It's a stat just like VPIP where you're trying to get close to an optimal range rather than just trying to make it as small or as large as possible.

[Troll_Inc]

Good point Iggy, it may not be linear.

Also, you shouldn't just look at people with x,000+ hands because it probably eliminates losers. Or I guess if you do limit it that way the question is more like, "Of players who play a lot, what is the best range for this stat".

[iggymcfly]

If you look at it in the short-term though, there probably would be a linear relationship since the people with really high W$SD in the short-term would be the ones were running hot. I think the data you get for people with >5000 hands would be much more accurate than the data for people who played less, even if there was still a little bit of a bias.

[pete fabrizio]

the problem is that i don't have a database nearly big enough to do any meaningful analysis on this stuff, since both winrate and w$sd are extremely volatile. and on top of that there are going to be tons of biases -- e.g., people who run hot in their showdowns are more likely to keep playing, etc.

[Troll_Inc]

[ QUOTE ]
If you look at it in the short-term though, there probably would be a linear relationship since the people with really high W$SD in the short-term would be the ones were running hot. I think the data you get for people with >5000 hands would be much more accurate than the data for people who played less, even if there was still a little bit of a bias.

[/ QUOTE ]

This hypothesis could easily be tested.

It's a well known assumption that losing gamblers quit. So the people that only play a little bit, what is THEIR w$sd?

Along the same lines, I would bet that of all the people in pf's database with <500 hands, their VPIP is greater than the people in pf's database with >5,000 hands.

[jipster]

my stats are 53.3 w$sd, 6.14 bb/100...60k hands

although imho i've run bad for long periods

am i to infer i don't look people up enough?