A study I'm going to do of my heads up play--what do you think?

[Bigwig]

Here are the stats from my last 258 $50 Party STT's:

Finishes:
1. 30
2. 45
3. 37

ITM: 43.4%
ROI: 26.5%

Obviously, that 2nd place business is getting out of hand. Now, I know that I more often have the shorter stack (from memory) when getting to heads up play, so I think it is incorrect to automatically assume that I'm as bad as the numbers say.

So I was thinking of keeping track of the size of the stacks when I reach heads up, then comparing them to the win/loss %. This would give me a better idea of whether or not my heads up play really sucks, or is just below par.

The real question I have is what adjustment I should make to the equity based on my skill. Obviously, if it works out that I'm exactly where I should be based on stack size (50% stack = 50% win rate) I'm actually playing poorly heads up because I'm a winning player.

So, what adjustment should I make upward for a winning player?

And, is there anything in my evaluation process that doesn't look right?

Thanks.

[mackthefork]

Surely if you are a winning player then no adjustment would be best, as your heads up opponents are likely to be on average as good as you are.

Mack

[Slim Pickens]

I don't think you should translate your skill edge over the field to an edge against your ITM opponents. You are more likely to be up against above-average players at that point in the tournament. How much more likely? I can't say without data on every single player's finish distribution. My answer is I think you should make no adjustment for skill difference, as that is what you are trying to discover, right?


[ QUOTE ]
So I was thinking of keeping track of the size of the stacks when I reach heads up, then comparing them to the win/loss %. This would give me a better idea of whether or not my heads up play really sucks, or is just below par.

[/ QUOTE ]

I would also track:
1) preflop blind steal attempt and success rate off your SB
2) preflop steal attempt and success rate of your BB when the SB just completes.
3) preflop steal attempt and success rate of your BB with a raise to you.
4) Your average pot EV in situations where you bet all-in (or PC'ed).
5) Your average pot EV in situations where you call all-in.

[Apathy]

[ QUOTE ]
Obviously, if it works out that I'm exactly where I should be based on stack size (50% stack = 50% win rate) I'm actually playing poorly heads up because I'm a winning player.


[/ QUOTE ]

This is not necesarily true. If your are finishing worse then your chip count indicates you should then that can be cause for concern.

A few cautions:

1. Small sample sizes can lead to erroneous beliefs
2. Some people play better HU when thay have the chip lead and play worse HU when they have a chip disadvantage or visa-versa, or they have trouble against certain types of opponents and can crush others, so finding out that you 'suck' at HU is not going to help you determine what stage of HU play you need to work on.

Your evaluation should still be interesting just make sure to take it with a grain of salt.

[Bigwig]

[ QUOTE ]
Surely if you are a winning player then no adjustment would be best, as your heads up opponents are likely to be on average as good as you are.

Mack

[/ QUOTE ]

I thought about that. But do you really think the 'average' winning player is even at 15% ROI, let alone 20% or 25%?

[TheUsher]

Nice stats Big. I wouldn't necessarily try to change how you're playing right now since the 2nd place finishes could just be from you having the shorter stack HU. Your ROI is great for 4tabling (vonmises right?). My ROI is the same as yours but 1>2>3 from aggression which gives me the big stack HU but this causes my ITM to be about 5% lower than yours.

I can't really comment about your game as I haven't really seen you in action but you're doing very well to say the least. Keep it up. /images/graemlins/smile.gif

[Bigwig]

[ QUOTE ]
Nice stats Big. I wouldn't necessarily try to change how you're playing right now since the 2nd place finishes could just be from you having the shorter stack HU. Your ROI is great for 4tabling (vonmises right?). My ROI is the same as yours but 1>2>3 from usually being the big stack HU but this causes my ITM to be about 5% lower than yours.

I can't really comment about your game as I haven't really seen you in action but you're doing very well to say the least. Keep it up. /images/graemlins/smile.gif

[/ QUOTE ]

Yeah, it's vonMises. Thanks.

It's about 1/2 3 tabling, 1/2 4 tabling. My system acts up with 4 open sometimes. Depends on it's mood.

[curtains]

Sample size is too small IMO. You could easily just be getting unlucky headsup or starting out with a smaller stack regularly. It'd be much more useful if we were actually shown exact hands you played or tough situations you felt you got into.

[valenzuela]

2 to 3 when u start HU with a shorter stack?? Why are we panicking?

[Misfire]

Shouldn't you be more concerned with why you're getting heads up with a shorter stack?

I have slightly a higher ITM%, but my ROI absolutely SUCKS, and from what I've been reading, my bubble play is what's really lacking. I don't think I am getting ITM with enough chips to even compete.

[sofere]

Meh...screw skill...push every hand.

[Benholio]

Do you use pokertracker? Go to "position stats" and filter it for 2-handed only, and look at your VP$IP/PFR for the SB and BB. For the SB, your PFR should be in the area of 50% (at least).

Also, your PFR% from the SB should be some double digit number, too.

I'd look at these stats before I'd look at your results, since even 75-100 trials might give you very skewed results.

[1C5]

[ QUOTE ]
Do you use pokertracker? Go to "position stats" and filter it for 2-handed only, and look at your VP$IP/PFR for the SB and BB. For the SB, your PFR should be in the area of 50% (at least).

Also, your PFR% from the SB should be some double digit number, too.

I'd look at these stats before I'd look at your results, since even 75-100 trials might give you very skewed results.

[/ QUOTE ]


You said SB twice, you mean BB for the 2nd part of the post?

[citanul]

if you want useful study, track how many chips you have 4 handed, 3 handed, then 2 handed. /images/graemlins/smile.gif

i've been thinking about setting up a spreadsheet myself to look at similar issues with my own results.

citanul

[Benholio]

Yeah, I did mean BB for the second part... thanks.

[curtains]

Sometimes good players will start headsup with shorter stacks, as they are better at surviving on the bubble.

[Nottom]

[ QUOTE ]
if you want useful study, track how many chips you have 4 handed, 3 handed, then 2 handed. /images/graemlins/smile.gif

i've been thinking about setting up a spreadsheet myself to look at similar issues with my own results.

citanul

[/ QUOTE ]

I actually did this for myself for the heads up case and plan on trying to expand it to the 3, 4 handed cases whan and if I find some time.

I'm in the same boat as the poster here. I'm currently winning about 45% of the time when I get heads up (it was less when I started my spreadsheet) but after runnign the numbers I only start with about 41% of the chips so I'm actually outperforming my "expected" win rate by a relatively comfortable margin. This indicates my problem lies in the stages prior to the actual heads-up battle, and I hope to find where that problem is soon.

(Quickly looking at my bubble numbers would indicate that my leak is likely there since I tend to bubble off a bit more than I should, although I also have an unexplainable spike of 6-7th place finishes.)