My AK/AQ ain't hitting at 32.4% -- normal?

[randomfish]

I just ran a query on 11,000 hands in PokerTracker. Turns out I've been dealt AK a total of 129 times, but have hit at least one A or K a total of 37 times, for a 28.7% hit percentage. For AQ it's a mere 32/135 or 23.7%.

If I'm dealing cards to myself, my unpaired hands should hit at least a pair around 32.4% of the time, right? So is the difference due to sample size, or because those times I get called when I raise AK/AQ preflop my opponents are more likely to hold some of my cards, or something else?

I'm curious because if this abnormal and I really should hit around 32% of the time, is this a sign that I'm getting cold-decked?

[meanjean]

I'm no expert but I think you're going to hear that your sample size is still too small. What do you need to hit? 3 or 4 in a row to get back to average?

[Peter-23]

You could count 32.4% of the times you didn't see a flop as a hit and see where you get.

[TEKEE]

ACCORDING TO THE BOOK WRITTEN BY KEN WARREN AK WILL MISS FLOP 75% OF THE TIME.

[k_squared]

[ QUOTE ]
ACCORDING TO THE BOOK WRITTEN BY KEN WARREN AK WILL MISS FLOP 75% OF THE TIME.

[/ QUOTE ]

not knowing exactly what Ken Warren said, I can't say he is just plain wrong... but your characterization of what he said is wrong. You will not hit an ace or a king ~67.6%of the time, meaning that you will hit at least an ace or a king on the flop 32.4% of the time... although this does not count times when you hit flushes or straights... So, Ken Warren is either oversimplifying, or you are oversimplifying what he said.

As far as whether or not it is too small a sample size... the answer is obviously yes. Furthermore, focusing on your statistics in this way does not seem particularly helpful. Whether or not you hit your hand 25% or the time or 35% of the time a major difference in winning or losing will be how you play it based on the situation you are in. Knowing when to bet it aggressively (whether it hit or not) and when to slow down. Don't waste your time pouring over your stats so that you can worry about how unlucky you have been in the past. Focus on pluggin leaks, which means looking at where you have lost money and understanding why and what if anything you can do to improve your game.

-K_squared
-k_squared

[lucas9000]

[ QUOTE ]
ACCORDING TO THE BOOK WRITTEN BY KEN WARREN AK WILL MISS FLOP 75% OF THE TIME.

[/ QUOTE ]

THAT IS ONLY FOR 12TH POSITION THOUGH. EVERY OTHER POSITION AK WILL MISS THE FLOP 27.389743% OF THE TIME. I ALREADY MISS THE FLOP...AT LEAST I HAVE THE MEMORIES.

[Kaz The Original]

When you raise preflop, it's more likely people with ace's or kings or queens in their hands call you.

[asofel]

[ QUOTE ]
[ QUOTE ]
ACCORDING TO THE BOOK WRITTEN BY KEN WARREN AK WILL MISS FLOP 75% OF THE TIME.

[/ QUOTE ]

THAT IS ONLY FOR 12TH POSITION THOUGH. EVERY OTHER POSITION AK WILL MISS THE FLOP 27.389743% OF THE TIME. I ALREADY MISS THE FLOP...AT LEAST I HAVE THE MEMORIES.

[/ QUOTE ]

MEMORY, ALL ALONE IN THE MOONLIGHT.
I CAN SMILE AT THE OLD DAYS.
I HAD A SMALL SAMPLE SIZE THEN.
I REMEMBER THE TIME I KNEW WHAT SUCKING OUT MEANT.
LET THE MEMORY, LIVE AGAIN.

[randomfish]

Obviously. This was just something I did while going over my hands and trying to find leaks.

Then again, wouldn't I be happy to be called more often by other aces and kings when I'm holding AK? Those times an ace or king hits the flop, I should be in a good position to make money off of the guy with the dominated hand.

If the sample size issue is prohibitive, would it help if we got a lot of people to run the same query and see if the results averaged out to 32%?

[stabn]

[ QUOTE ]
When you raise preflop, it's more likely people with ace's or kings or queens in their hands call you.

[/ QUOTE ]

I THINK HE SHOULD STOP RAISING AK/AQ PREFLOP. THEY ARE DRAWING HANDS!!!

[dabluebery]

Either your sample size is too small, or this thread is worthless without pics. Not sure.

Definitely don't ignore the instances when you have AK / AQ and there is no flop, as an earlier poster suggested.

[randomfish]

And how do you propose I check the result of flops of hands when there was no flop? /images/graemlins/smile.gif

But honestly, how can those hands have any bearing on the result? No matter which AK hands I choose -- say I decided to look only at A/images/graemlins/heart.gifK/images/graemlins/heart.gif, for instance -- I should still hit at least one A or K ~32.4% of the time with that subset of my AK hands. If the sample size is good (and I'm well aware that such may not be the case in the OP), and my hit percentage is less than this, I'm either getting cold-decked or there is some other reason -- some of the aces and kings are in the hands of the callers, for instance.

I guess I could rephrase the original question to something like: how do you calculate the SD of your hit percentage with AK?

[barongreenback]

[ QUOTE ]

I guess I could rephrase the original question to something like: how do you calculate the SD of your hit percentage with AK?

[/ QUOTE ]
If you give the hit chance between 0 and 1( so you'd expect a mean of about .32) then I think the SD is around 0.5(bit less).
If this is right you'll find that your sample size is too small( 95% confindence interval +-0.09 from the mean).
There'll actually be people out there with this sample size having <20% or >45% hit rates.

I've done this in a hurry so if I'm wrong don't jump on me anyone. I'd advise a better pattern mapper anyway. Why leave things to chance. /images/graemlins/wink.gif

[Cobra]

This can be thought of as a binomial distribution with probability of success as .324 and probability of failure of .676. With 129 AK you should have had a hit 129*.324=41.8 times. Your std deviation would be the square root of n*ps*pf=129*.324*.676=square root of 28.26 which is 5.31. You are less than one std deviation away from the mean. You should be within 1.1 std deviation at least 17.3% and outside that 82.6%.

Cobra

[AngusThermopyle]

But honestly, how can those hands have any bearing on the result?

You said you were dealt AK a total of 129 times.
Did you get a flop every single time? All 129 times?
If you raised every time and got no calls 19 times, you actually saw the flop with AK 110 times.
You said you have hit at least one A or K a total of 37 times. If this was in 110 flops ( instead of 129 hands ), your percentage is 33.6%. Closer to the expected results.

If the 129 hands are just the times that there was a flop, then you have a different situation. Then if you are playing against hands of BC and DF, you might infer that at least one of B,C,D, or F is an A or K, perhaps 2 are. That would mean your odds of flopping an A or K would decrease.

[blingice]

TEKEE

RIP THE "CAPS LOCK" BUTTON OFF OF YOUR KEYBOARD.

Bling.