maximizing AK

[rhinoceros]

I've been collecting stats for awhile and have some surprising results with 1100 NL hands in my database, mostly PP$100 (some $50).

My standout worst starting hand is AA. This is mostly attributable to 1 bad beat. The database is too small to take this very seriously.

However, AK (both suited and not) each have shown losses. Additionally, nearly all Ax show losses. Focusing on AK, AQ, AJ, AT, I wonder why I am losing money on these.

Hypothesis 1:
The database is still too small. This is an easy excuse, but I don't believe it. I have 46 hands AK-AT, certainly not a huge number. But my results with non-ace starting cards are ordered pretty much as expected. KK is way out in front, and the next 5 include QQ, JJ, and KQs. KQ, KJ, KT line up perfectly. There is indeed a lot of noise, but the results do not look random, wild, or kooky.

Hypothesis 2:
I am underplaying these hands (AK-AT). This seems likely, since I am losing money with premium hands (even as a subset of a database with significantly positive results). But I have not taken a lot of big losses and/or bad beats with these hands. What is curiously missing is sufficient big wins.

Hypothesis 3:
My opponents make a preliminary read of AK for all raising hands. Thus I can't win big with this hand (or its weak sisters) because they won't make their customary loose calls after I have made a hand. This is my tentative conclusion. However, I am not sure what to do with it.

Solution 1:
play more deceptively with AK (now the posterchild for the whole class AK-AT). Small raise? Limp? I don't really like this idea. My results improved dramatically (from big minus to big plus) when I learned the perils of small raises.

Solution 2:
Do very little. When I have raised pre-flop, but completely miss the flop of, say, A 9 5 rainbow, I tend to bet out. Perhaps I am stealing enough on these hands against a weak Ax to compensate for not killing those same hands with AK. Maybe I should be more aggressive in trying to steal those hands. With TT, do you bet the A98 flop too? out of position? If the 98 are suited? How about AQ8? Any suggestions/examples on where to draw this line?

Now, 2 questions for those few still reading:

First, have you observed similar results in your own play? And, if so, is it anecdotal or have you collected thorough data?

Second, if you have similar experience, what do you do about it?

If you're losing money with them, one solution, of course, is not to play them or play them less often. I know your sample size is small, but have you looked into what position you're playing these hands from? Are you playing them from early position and then making lay downs to people who you feel have to have better Aces? You may want to limit yourself to playing AKo, AKs, and AQs from early and mid position, and not playing AT at all except as a steal.

Also, I hate to say it, but don't rule out that your sample size is too small. (I know you haven't ruled it out, but you sound like you want to). If you're a consistant winner, then you may not want to change anything until you get a bigger sample size, since you could be hurting yourself. And as you realize, your play with aces may be making you more money elsewhere. Change the way you play the aces and the other areas of your game might fall apart.

Along those lines, say you only raise big cards pre-flop. Then when you raise, your opponents assume you have an ace. Now you make money with your other big card hands, espcially when you hit top two pair or a straight and your opponents assume you cannot have that hand because they KNOW you have an ace in your hand. That looks kind of like your results.

Finally, it seems like you're trying to decide whether limping AK or raising with it is better. Why not try mixing it up. Sometimes limp and sometimes raise (especially with hands like AKs early). And make sure you raise with other hands sometimes like T9s or 77 (especially in late position). Just don't get carried away with it, or you'll lose in raw EV what you gain in deception.

I don't seem to win with AK that much. Saturday at Foxwoods, Big Suck held up so many times that the whole table was laughing abvout it.

[sizeup]

My opinion based on limited experience: AK can be difficult to play in big bet games since it builds top pair, top kicker. If you raise preflop, and flop an A or K, you are unlikely to get action unless you are beat. So it is generally good for winning small pots or losing big pots. I don't really know how best to play it, I usually (but not always) raise preflop and would almost always bet the pot on the flop if I hit, but I won't back top pair with my whole stack unless I don't have many chips. I like your idea of stealing more on the flop. Hope this helps.

[Eric P]

if you hold AK the you hit the flop, assuming there is no legitamte flush or straight draws, why not try to trap? Check raise the turn, or just slowplay it until the river, of course if something happens you should obviously bet it so that you are not outdraw. I'm not saying do it every time, but you should consider mixing it up like this.

[rigoletto]

I'm very new to NL and PL, so judge my comments accordingly.

My opponents make a preliminary read of AK for all raising hands. Thus I can't win big with this hand (or its weak sisters) because they won't make their customary loose calls after I have made a hand. This is my tentative conclusion. However, I am not sure what to do with it.

You might be raising with to few hands preflop. I play mainly PL with blinds of $0.50/1and a maximum buy in of $100 (the babygame). This game has a lot of weak tight players, so in late position (and sometimes in tight games) I raise with a variety of hands. This gives me many upportunities to steal and people are less inclined to put me on AK or QQ-AA. I also raise the pot size almost all of the time.

[sizeup]

Absolutely, this sounds good to me as a way to mix it up and get some action from a hand you're ahead of. I would be hoping for the opponent to put in exactly one bet though, I still have a hard time playing on if more than one pot-sized bet goes in post-flop. The second one (assuming you raised preflop) is starting to get to be a lot of chips to back your 1 pair with. I wouldn't check-raise the turn for that reason - I don't think you get played with unless you're beat in most big-bet games. Of course, as you mention, the appearance of draws on the board changes this situation. Thanks for the input.

[angelo alba]

"Hypothesis 2:
I am underplaying these hands (AK-AT). This seems likely, since I am
losing money with premium hands . . .
sBut I have not taken a lot of big losses
and/or bad beats with these hands. What is curiously missing is
sufficient big wins."

Isn't this partly answered by:

"Hypothesis 3: My opponents make a preliminary read of AK for all raising hands. Thus I can't win"

Well if you're getting read THAT well by your opponents.

I know this is ,oh slightly dangerous in NL, but to 'vary' have you considered raising with AK AND
AQ AND AJ suited or not, in or out of position vs 1 limper or a full table?

Just a brilliant (or suicidal) thought.

BTW perhaps you could tell us a bit more about your post-flop play.