I've been playing around with the trial version, viewing my hand histories, but all it does is frustrate me.

It tells me to push when I folded, and tells me I should have folded when I pushed, but unless I memorize the position, stack size, and % of playable hands it suggests based on that tournament structure, how the hell am I supposed to remember this info when I'm at the table playing?

Is there some sort of guide or book companion that I'm missing here? Or if I review enough hand histories am I just supposed to eventually remember all of this p00p?

Thanks in advance for your help.

Someone made a HUD overlay that pulls the stats from your database and puts them over top Wiz's program. Can't remember the name, but someone will come along who does.

Basically, you should be able to get a rough estimate on the calling/shoving ranges based on game conditions and stack sizes once you've played a certain amount. It's also important to note that you shouldn't only look at correct/incorrect plays in Wiz. Go through your HH's, look at interesting hands, and play with stack sizes, ranges, and positions to understand WHY a play is or isn't +EV. That will teach you much more than simply going through and trying to get green check marks on every hand.

For instance this hand.



It tells me to fold AJs to the push in front of me.

BUT, if I change it so the guy to my right folds, and the guy to his right pushed, (dshrow), it tells me to push.

I guess what's frustrating is there's no explanation for why it thinks those two scenarios are different.

halp plz

It sounds like you're lacking from a basic understanding of ICM?

Have you read the FAQ, and all of the archived threads there?

I've read alot of it.

I can't do ICM calculations at the table. I understand what it is and how it works....but I'm just effing confused as hell right now because I review my hands and it tells me I shoulda pushed 93o from the BTN and called an all in shove from the BTN into my BB with J6o, but then tells me to fold AJs to a shove from the 4 seat, but to call a shove from the 5 seat with the same hand, and to fold 77 UTG 5 handed when I only have 8 big blinds.

If some sort of explanations were offered along with the words push or fold I think that would help immensely, since I can't possibly memorize every possible scenario, it would help if the logic was explained because I can apply logic at the table.

FAQ helps.

Personally I think this does a little bit better job of explaining ICM to noobs/people bad at math.
http://www.flopturnriver.com/phpBB2/...es-t17868.html

To answer your question in regards to the above hand, it is because there is a 'Bias Against Confrontation' in SNGs (concept introduced to me courtesy of AMT's videos on deucescracked, free 7 day trial worth checking out, /end shameless plug), particularly 9 man SNGs.

Any time your chips are all at risk, if you are a 50% favorite, that means 50% of the time you lose and cannot make any money in the SNG. If you win in the above situation, then while you are doubling your chips, you are not doubling your expected $'s won for the tourney... so taking at 50/50% coinflip has a lower $ expectation than not taking the coinflip.

Hope this helps.

Cliff notes, studying wiz is going to be largely pointless in my opinion if you do not understand why wiz is giving you the results it is.

Start with a really common scenario (such as shoving BvB with <=10BB) and then spend some time working out what factors should influence your decision. After a while you'll have a good idea of when (roughly) you should deviate from the "standard" play (pushing atc in this case) and then you can move onto looking at another common scenario and so on. Don't waste your time looking at extremely rare scenraios yet (see below).

Then play some SNGs and while playing note down any hand numbers where you were unsure of what you were doing (ie: you had to think lots in game) and then after you've finished playing, look through these in SNGWiz and see if you can work out how they fit in with your "common scenarios" mental model. After doing this a few times you should be pretty good (and won't have to think very much in game), but you will probably still have some massive leaks that crop up occationally (see below).

Finally (you might have to do this a lot when you first start), run though your played hands setting the ranges as accuratly as possible and try to find your biggest most costly mistakes (ie: missing an open push >= 0.5% or making a bad call <= -1%) and try to work out how to avoid making the same mistake again (it only takes a couple of these big mistakes in every SNG to turn a +10% ROI into a 0% ROI!).

Juk

So if the player 2 to my right shoved, and I called and lost, I would be crippled, but not eliminated, but if I lose to the guy who did shove, I'm out.

So sngwiz doesn't really make a distinction there? It'll recommend a looser call if you have villain covered?

But if you are down to 300 chips you are practically eliminated, which will be reflected in your tournament equity (expected $ winnings with the end result stack) if you calculate it using ICM.

Do you set ranges based on villain's hud stats or do you use a pretty standard range based on the buy-in level you're playing?

I've just been using the default, and using it to review my hands afterwards, to see if I missed any good spots to push....oddly though, I'm actually pushing too much, and I figured it would be the opposite. (I play about 15/11/2)

Set them based on stats, feel, experience, and your image at the table.

So for example, shoves will be worse if you have been shoving a lot or if you have shoved and shown down with poor hands.

Ok, that's what I was thinking...but for some reason, if the player 2 to my right pushed, and the player directly to my right folded, SNGwiz recommends I push.

I can't figure out why it's ok to call one guy's push and not the other guy's, with the same hand and approximately the same stack sizes.

If I could figure out the logic or math behind the recommendations then at least I could try to apply it in the future. As it is, it feels like the program is throwing darts at the board and saying "push this trash, fold that top 20 hand" mwahahahahahaha.

This. Can I this myself? I feel dirty.

You can look through some hand histories to get a good idea of some default ranges to use (and even split these into reg and non-reg), but the last time I suggested this a guy spent like a month doing just this (please don't ).

As a very rough guide: as you analyze some common scenario (ie: BvB, etc) write down the Karlson-Sklansky rank of the hand a player (or player type) calls with (over say 20-30 samples, 50 tops). Then double the average of the recorded ranks to get an idea of their range - this will give you an idea of what the average player's range is within a few percentage points (but please don't spend 4 weeks doing this!).

Just using the default is a pretty bad idea as SNGWiz will sometimes use very wrong values (especially for low limit players).

Juk

I have not read through this thread really, but OP wiz is not a magic fix with which we all memorize every situation available.

most of us have spentliterally hundreds of hours going over common spots and themes, which over time you begin to intuitively learn what feels like a push or a shove. Some spots come up so often that you know all the subtle stack variantions that make a certain hand a push or shove.....

even then there will still be spots you are unsure of or suprise you

unfortuately its just down to hard work, not a photographic memory...

Here's a good example of what I mean when I say I'm confused and dunno how to use this to improve my game. Both of these hands are from the same sitngo, same structure, and basically the same hand in the same position.

First hand, A9o in the BB, with 4bb's in my stack, facing a push.


2nd hand. A8o in the BB with 3bb's in my stack, facing a push.


To me, these are basically the same hand in the same situation. (As a matter of fact, the suggested percentages of hands in the 2nd example are much wider than in the 1st, but A8 is a fold for 36% and A9 is a push for 26%?)

Plus, in the 2nd hand the bubble has already broken and in the first it hasn't.

So I'm having a hard time figuring how I can take the recommendations and apply them when I'm at the table.

In the two pictures above, note the opener's "Open percentage". The first one opens 32%, and A9 does pretty well against that range (53% according to stove). The second one opens only 8.1%, so A8o doesn't look so good (31%). The reason why SNGWiz assigns these wildly different ranges may be due to position or the distribution of stack sizes.

I'm sure this has been said, but getting the ranges down just take lots of time, memorization, and practice. Take what I say w/ a grain of salt because i'm merely a decent player, but I've spent a looooooooooooonnnnnggg time going over wiz before parts started to click. OH, and ignore the standard ranges you're going to have to learn to put your own in. Lastly, something I've learned recently (just post this actually) is wiz isn't god. It tends to err towards caution in my opinion because it doesn't include the important of fold equity and blinds rising fast in turbos, hopefully I can get responses in my post for you to read what I mean.

I am trying understand your range estimate technique. If you double the average Sklansky numbers on villains calling hands does that give you the top of his calling range. If he has a better hand he would raise. Do you use the average Sklansky number as the bottom or the middle of his range?

The program is WizHud written by Hood. It is a great tool for reviewing. It helps with ranges by overlaying ur HUD on Wiz.

"The software is free for the basic version. With that you get unlimited use, hud stats on all opps with vpip pfr data, and some handy shortcuts for navigating wiz. But for a small donation of $16 gets you lots of rad awesome cool features - like loads of new stats, full hud customisation, filtering by # of players, instant nash calculations using holdemresources.net, and lots more."

The features of the full version are the way to go imo

here is a link to the thread
http://forumserver.twoplustwo.com/45...re-itt-482242/

There it is.

I tried the free version, and will definitely be donating just to be able to filter via players per hand. The stats will be much more useful that way.

Yeah I guess that could be it. The villain in the 2nd hand is opening from a much earlier position, and if it assumes villains will be more aggressive at bubble time, which is when the 1st hand occured, then it would make sense to call in the 1st hand and fold the 2nd. (even though I did the opposite)

I just wish there were explanations or shortcuts or something like that so I could understand "why" something is a fold or a push. Otherwise, short of memorizing every possible scenarion I don't see how this can help me.

Sadly that only works for HEM and I use PT3

Well first you need to get a list of the Sklansky-Karlson Rankings in terms of percentages (ie: like SNGWiz uses them so: {AA}=0.45%, {KK+}=0.9%, {AKs+}=1.2%, ..., {32o+}=100%). I did have a link that listed these, but can only seem to find one that lists the rank now.

Keep in mind this is only a very rough technique, but here is the idea: imagine if a player only calls in a certain scenario with the top 20% of hands (and assume he uses the Karlson-Sklansky ranking). If you were to take a good number of observations and write down the K/S rank of each, then what would be the mean of these observations be? Well (ignoring the different gap sizes between different hand types and the fact that he probably won't be following the K/S rank exactly...) the mean would turn out to be ~10% so to get back to him calling with the top 20% you just need to double it (the same goes for a player who will call atc: your observed mean % rank will be 50%).

Juk

http://sngwiz.com/forum/index.php?topic=7.0