Small Question on SnG Wizard

P

pat3392

Visionary
Silver Level
Joined
Jan 19, 2010
Total posts
565
Chips
0
How come A5s is considered to be the top 12.9%, whereas KQo is considered to be the top 17.1% hand? This seems wrong to me; I consider KQo a much better hand than A5s for. It also says that KTs is better than KQ.

Do I simply have a flawed preconception on hand strengths or is SnG wizard flawed in this aspect?
 
Logan2

Logan2

Legend
Silver Level
Joined
Nov 5, 2008
Total posts
4,054
Chips
0
How come A5s is considered to be the top 12.9%, whereas KQo is considered to be the top 17.1% hand? This seems wrong to me; I consider KQo a much better

Do I simply have a flawed preconception on hand strengths or is SnG wizard flawed in this aspect?
As5s=60.609%
KcQd=39.391%
 
P

pat3392

Visionary
Silver Level
Joined
Jan 19, 2010
Total posts
565
Chips
0
As5s=60.609%
KcQd=39.391%

Yeah but one must think about it vs. ranges, not an individual hand. Ace rag has a decent chance of being dominated, whereas KQ has little chance and may dominate others

vs. a tight range ( 66+,A8s+,KJs+,ATo+,KQo)

A5s wins 36.832%
KQo wins 37.656%


vs. a average range (22+,A5s+,K8s+,QJs,A7o+,KTo+)

A5s wins 42.371%
KQo wins 57.629%

vs. a loose range (22+,A2s+,K5s+,Q9s+,J9s+,T9s,A4o+,K7o+,Q9o+,JTo)

A5s wins 48.662%
KQo wins 51.338%


I think this shows that KQo is a better hand than A5s, yet SnG wizard believes it is 30% or so better
 
P

pat3392

Visionary
Silver Level
Joined
Jan 19, 2010
Total posts
565
Chips
0
I wonder what I have to do to make A5s be better?

A5s/KQo vs. 22+,ATs+,A5s-A2s,KQs,AJo+,KQo,32o

A5s wins 41.869%
KQo wins 40.779%

Success!! However, I'm fairly sure this range is not that relasitic, let's make it a tad bit more feasible

A5s/KQo vs. 22+,A9s+,A5s-A2s,KQs,ATo+,KQo,32o (2 more aces , 32o was kept in there for tilt value + I actually want A5s to win)

A5s wins 41.301%
KQ wins 40.710%

So for the A5s to be ahead the opponent must like A(2-5)s, 22 and 32o more than A9o/A8s/KJo. hmm I think this proves something

That range is a tad bit dumb/unrealistic imho though. Let's do one more to make sure I'm not overlooking something


A5s/KQo vs. 22+,A8s+,A5s-A2s,KJs+,A9o+,KQo

A5s wins 39.356%
KQo wins 39.535%


I think this proves no doubt about it that KQo is a better hand that A5s. It's hard to make a range that A5s is better against. SnG Wizard thinks it significantly better aswell, which is imo very wrong. Looks like I will have to add every hand as the slide percentage of hands is inaccurate
 
W

WiZZiM

Legend
Silver Level
Joined
Oct 28, 2009
Total posts
5,008
Chips
0
You can read up on sklansky/chubokov hand rankings.

As you have found out, differant hands play better against differant ranges than others. For example, if you know villian is shoving with 100%, it's better to call with high card strength like K7 rather than a hand like 22. This is because against that particular range 22 will always be at best a 50/50, whereas the K high hand will be in 60/40's a lot, and also has a chance to dominate hands in villains range.

So yeah, KQ will probably play better against a range which is prodiminately high Ax type hands, sure. Generally speaking though, shoving isn't much of an issue, but when we are planning to call a shove, shoving ranges are almost always wider than calling, thus if you widen the ranges out to 50%+ you should find the Ax hands are better than KQo.
 
W

WiZZiM

Legend
Silver Level
Joined
Oct 28, 2009
Total posts
5,008
Chips
0
And while it's not perfect, it's pretty close to being accurate, i know the slide scale sucks a bit, it uses the sklansky hand rankings i believe. so it will favour suited over non suited and what not, but we can never really get a perfect range anyways, there will always be that little bit of error both ways, either your read of the player is wrong , or whatever. As long as you have a decent approximation thats going to be the best we can get. But if you make a -.2.5 mistake, theres no way errors can be responsible there, even with the close decisions, where it's close to neutral equity, little errors won't really affect it too much.
 
P

pat3392

Visionary
Silver Level
Joined
Jan 19, 2010
Total posts
565
Chips
0
A5s/KQo vs. (22+,A2s+,K2s+,Q4s+,J6s+,T7s+,97s+,86s+,75s+,64s+,53s+,43s,A2o+,K5o+,Q7o+,J7o+,T8o+,98o)

A5s wins 53.331%
KQo wins 55.465%

A5s/KQo vs. (22+,A2s+,K2s+,Q2s+,J2s+,T4s+,94s+,84s+,73s+,62s+,52s+,42s+,32s,A2o+,K2o+,Q2o+,J3o+,T6o+,96o+,85o+,74o+)

A5s wins 57.434%
KQo wins 59.740%


A5s/KQo vs 100%

A5s wins 59.923%
KQo wins 61.456%



I think this makes it clear that KQo is a better hand than A5s. I think it's quite absurd that it believes A5s is 30% better. I'll look into the slaskive/chubokve range thingo.

I suspect that these ranges where designed when poker was dominated by tight nits; A5s was considered better because it had the flush/straight potential and could spike a A if the semi-bluff failed, whereas KQ was dominated too much because KJ/QJ was not played enough.

Is there anyway to program my own slider range? By using the default slider I fear that I'm teaching myself wrongly
 
W

WiZZiM

Legend
Silver Level
Joined
Oct 28, 2009
Total posts
5,008
Chips
0
Hmm interesting, there isn't a lot of differance between the two though, i figured the ace would be slighltly favoured over the King but there you have it, really suprised the KQo is ahead against 100% strange.

Anyways to be honest, most top pros have used, or still use this program. So really, if it's accurate enough for them, it's going to be accurate enough for us, even if there may be a slight discrepancy. I'm really not sure where you are getting your numbers from though, how do you figure that SNG is telling you that A5s is 30% better than KQo?

there is probably a way to get into the program and change it, but i have no idea how to, and i wouldnt waste my time with it.

And again, to reassure you, as long as you have semi decent accurate ranges, it will give you the information required to determine if the shove is close good or bad, the slight variation you get when it preselects hands for you really just wont make that much differance.

Like for example, you think villain calls with decent aces up to A7, but the program selects hands like A3s. if you go and do all the math and whatever it may work out to be a slight .10 differance, or even lower i'd suspect. Plus, we are probably not completely accurate with the ranges we give villians anyway, so for me, i'd focus on being able to put players on pretty accurate ranges, rather than focusing on stuff we probably cannot change and probably won't make a lot of differance.
 
P

pat3392

Visionary
Silver Level
Joined
Jan 19, 2010
Total posts
565
Chips
0
Hmm interesting, there isn't a lot of differance between the two though, i figured the ace would be slighltly favoured over the King but there you have it, really suprised the KQo is ahead against 100% strange.

Anyways to be honest, most top pros have used, or still use this program. So really, if it's accurate enough for them, it's going to be accurate enough for us, even if there may be a slight discrepancy. I'm really not sure where you are getting your numbers from though, how do you figure that SNG is telling you that A5s is 30% better than KQo?

there is probably a way to get into the program and change it, but i have no idea how to, and i wouldnt waste my time with it.

And again, to reassure you, as long as you have semi decent accurate ranges, it will give you the information required to determine if the shove is close good or bad, the slight variation you get when it preselects hands for you really just wont make that much differance.

Like for example, you think villain calls with decent aces up to A7, but the program selects hands like A3s. if you go and do all the math and whatever it may work out to be a slight .10 differance, or even lower i'd suspect. Plus, we are probably not completely accurate with the ranges we give villians anyway, so for me, i'd focus on being able to put players on pretty accurate ranges, rather than focusing on stuff we probably cannot change and probably won't make a lot of differance.

A5s is considered to be the top 12.9% hand, whereas KQo is considered to be the top 17.1% hand, a little less than a 1/3 difference or 30% better. I see now what I wrote was probably misleading.

It turns out one can't add in their own ranges. The reason is that it would take too long to compute all the answers that way.

I've also read that it uses a different slider for different situations. So a suited connector would be weighted as a stronger hand in a pushing situation and an ace would be weighted as stronger in a calling situation(not sure how that one works..) So even though I'm skeptical of the ranking system this doesn't make it seem so bad + the author plans on improving it + you're right it won't make that big of a difference
 
Top