Check out our brand new Local Poker Communities! Get updates and interact with poker players in your area.
Visit the United States Poker Community | Visit the California Poker Community | Read more about the Launch of P5s Local

All-in Against Random Aces[ return to main articles page ]

By: thearthurdog
Published on Sep 10th, 2009
I’m on the final table of a Full Tilt Poker $11 Knockout MTT. I’m 6th in chips, with about 23 big blinds. I’m on the button, and grin as I look down at two red kings. A player in middle position (who has me covered) raises to three times the big blind. The action folds to me and I ship my entire stack in without hesitation. The initial raiser calls and flips A 5. The flop comes J-9-6 with no clubs. Safe so far. The turn is a Q. Still good. The river delivers the only card I didn’t want to see, an Ace... thearthurdog finishes the tournament in 9th place, $67 has been credited to your account. Meh….

I want to be really clear: this is not a tale of woe about bad beats. But, I admit the curse of A-x has brought me unstuck at critical points in MTTs more times than I can remember. In fact, I would go as far to say that my bankroll would be substantially bigger if that damn ace hadn’t come in a few deep and important spots.

It used to drive me crazy until I actually sat down and worked out the mathematics of why that ace keeps hitting the table. Combine the math with a real understanding of the way people play an ace pre-flop in lower buy-in MTTs, and now I am at a point where my K-K gets cracked and I don’t blame the curse; it’s just how it is. Allow me to explain.

I play lower buy-in MTTs and play most of them these days at Titan and Party (smaller field sizes which suit my schedule a lot better). Both of these sites have 10-handed tables, which actually gives the curse of A-x a little more impact.

Let’s imagine we are playing a $6 MTT at Party Poker. 20 cards are dealt pre-flop. Which means that on average there are 1.54 aces dealt each time. Of course there are occasions when there are 1, 2, 3, 4 or none. For the purposes of our example let’s say there is 1 (which mathematically is likely to occur often).

Now imagine that you have any good pocket pair that is not A-A (let’s say J-J). You have a stack of about 20 big blinds. The action folds to you in middle position and you raise the standard three times the big blind. The button calls without hesitation. The flop comes T 6 5 rainbow. You bet around half the pot. The button calls again. The turn is a 2. Perfect. You ship it in. The button calls and flips his A-7. You can hardly believe your luck, until (you guessed it) the river comes an A ending your tournament life. How did this happen? There are three major influencing factors here.

Firstly, you have to understand the way many people play aces pre-flop in low buy-in MTTs. They simply cannot fold them. I think a lot of this comes down to that fact that these players do not have a good understanding of the rank order of starting hands and have ‘any hand with an ace’ way up at the top of the list. In their minds then, it would be foolish to fold such a strong hand pre-flop.

The second part of this phenomenon is that when they don’t hit their flop, they honestly believe that A high is good. Players in this situation will rationalize their play by convincing themselves that you are on a draw (despite the absence of a draw heavy board), have an ace with a weaker kicker, or a weaker pair they still have a chance of beating if (when?) the ace comes. The fact that they are up against a good overpair rarely enters their mind. If they are, well, they have still got two more chances of hitting that ace right?

If you don’t think some players behave like this with an ace in their hand, ask anyone who has put in any volume in low buy-in MTTs.

The third part of this scenario is the mathematics behind another ace hitting the table. Now in our example, let's assume there were still 3 aces in the deck. 3 aces out of a possible 32 cards. Let’s examine the probabilities for this specific scenario when there are still three cards out that could pair our opponent's ace:

• Flop card 1 = 3/32 or 9.4%
• Flop card 2 = 3/31 or 9.6%
• Flop card 3 = 3/30 or 10%
• Turn card = 3/29 or 10.3%
• River card = 3/28 or 10.7%

When I worked that out I instantly saw why A-x seemed to be popping my big pairs so often. In our example, it is roughly a 1 in 10 shot at an A on each of 5 cards. No wonder some players remember the times they’ve spiked an Ace, and are looking for any reason to try again.

Combine those probabilities with the previously examined tendency of bad players falling in love with A-x, and it starts to become clear. There is no curse, no bad-beat algorithm crushing your big pairs and your dreams. There’s just the tendencies of bad players, and simple probability. The beats might not be much easier to take, mind you, but I felt better after seeing the numbers and rationalizing what’s going on. I hope you do too.

* thearthurdog Arthur Russell is a post-grad university student in Queensland, Australia who enjoys online poker in his free time.

More Articles by thearthurdog

How Does Poker Fit into Your Life? Jul 12, 2009

Adjusting to Micro Stakes Online Poker Games May 19, 2009

Comments

  1. <p>the same happened to me in a tourny...10th place...my kings get crack by a/4 off...i almost want to throw my cpu out the window..</p>
  2. <p>where did you get that 32 cards number? shouldnt it be 48? My guess is that u are considering each other player to not have a ace and thats so wrong...</p>
  3. <p>As he said in the article, for purposes of this example, he calculated probabilities of 1 ace being dealt to a player, then 2, 3, and 4 aces being out.</p>
    <p>For this example, he used at least 1 ace being dealt to another player as being highly probable.  So, the example is based on the times only 1 ace was dealt to one of the 10 players, leaving 3 in the deck.</p>
  4. <p>Assuming that nobody else has an ace ruins your calculations.  It is certainly less likely than random that someone will have an ace, but saying that no one will ever have an ace pretty much makes this numbers worthless.</p>
     
  5. <p>its just variance. Your opponent will hit an ace sometimes, as you worked out 1 time in 10, but the other 9 times you would of doubled up. </p>
    <p>Its Poker.</p>
  6. <p>In a 9 handed game it actually IS correct that more often than not there is only one ace dealt in the 18 hole cards (1.38 aces for each 18 cards dealt). Of course there will be plenty of times with none, 2, 3 or 4, but the most frequently occurring scenario WILL be that solitary ace.</p>
    Thread Starter
  7. <p>Great analysis, thanks.  I love the people who just don't believe in math or accept theoretical situations... it's what makes the low (and hopefully medium) stakes games go round!</p>
  8. <p>I don't like this.  It's too simplistic.</p>
  9. <p>If the conclusion is that people overplay ax in the lower stakes mtts that is fine. If you are trying to make a meaningful mathematical argument for this being the case I would argue that all you've shown mathematically is that: 1. in the case where only one ace was delt to anyone pf and 2. it was paired with a rag (or at least a card not strong enough to reraise and put you off your hand)  and 3. it went to the weak player who was willing to play it and 4. he was willing to play it to the river no matter what the board or your betting might be,  then he has a coinflip's chance of hitting his ace (over the 5 cards to come). That is such a small subspace of the original problem that I think it's rendered unmeaningful.</p>
     1
  10. <p>LOL @ this...like Ax doesn't win deep in mtts...if it holds for 9 times early in the mtt and but always comes the 10th time deep, u won't make any money obv.  Do a report on where Ax beats you in every mtt instead.</p>
  11. <p>Wasn't this article posted earlier this week? </p>
  12. <p>article needs more exclamation points imo</p>
     
  13. <p>If you said that AK was a small favourite vs pp if it was fullring and folded to button, then I would agree, this is just plain stupidity, hopefully somebody take ur advice, so p5-members get some value for the $ u get for writing this article.</p>
     
  14. <p>It doesn't take a lot of math to know exactly how your gonna get jailsexed as soon as you snap call the shove with KK and see the Ace-trey off, and everyone that reads this knows what I mean. </p>
  15. <p>When using the number 32 for your calculations, you eliminate 16 of 48 cards, or one-third.  There are 3 aces left in the deck, so eliminating a third of the cards without eliminating an ace totally ruins the math you provide...in fact, without doing the math myself, I would say every number you provide is 33% too high.  Also, even if you are assuming only 1 ace dealt, that leaves 34 cards, not 32.  9 players multiplied by 2 is 18, and 52 - 18 is 34.  Definitely not a good article to be featured on the front page of a site as awesome as P5s.</p>
     
  16. <p>Voff the example I used was Party Poker (10 handed). 52 - 20 cards = 32.</p>
    Thread Starter
  17. <p>lol I wrote a poem about this kinda thing. Anyone can read it here <a rel="nofollow" target="_new" href="http://donkdestroyer.blogspot.com/2009/08/donkey-donk.html">donkdestroyer.blogspot.com/.../donkey-donk.html</a> </p>
    <p>if they feel like it. </p>
    <p>I think shoving 23bbs was a bad move. Why not smooth call if no one else has called him or put in a small reraise to keep the blinds from getting involved? Evaluate from the flop- if favorable then shove (i.e no Ace, 3 to a str8 or flush). You have position on him so why not take a flop? If you had smooth called you would have 20bb left and with no one else in pot the pot would have 7.5 bbs you bet pot or raise if he bets and he is probably going to fold since he has npnd . Maybe he calls your flop bet so you really amp up the aggression on the turn and I'd say he is folding and you would have won a nice pot. Even though the low buyins are populated with donkeys by the time you get to the final table those players are usually the better players and they can fold to pressure. Unless you've been acting like a reckless fool I can't see him calling a turn bet. BTW I pokerstoved the hands in question and KK  isn't ahead (preflop) by as much as I would think vs A5s :  66.3% vs 33.6%- so he's gonna win a third of the time. </p>
  18. <p>I used to be one of those donks that fell in love wit ace-rags. But I soon learned that more often than me cracking kings, I was losing kicker battles. I now fold ace rages unless on the button in an unopened pot. Or when they are sooooted. Eeeeehm...</p>
  19. <p>OK fine, use 20 cards.  It doesn't change the fact that you remove one-third of the deck for your calculations, rendering them useless.  I understand that you wanted to emphasize how often A-x can beat KK, but you should do it in an accurate way, without all the number-fudging and exaggeration.</p>
     
  20. <p>It is impossible to remove one third of the aces. It is only possible to remove none, 1, 2, 3 or 4. The most mathematically likely scenario in a 9 handed game is 1 ace dealt and 3 left in the deck. No fudging, no exaggeration, just maths. In hindsight I should have used a 9 handed game in my example. </p>
    Thread Starter
 
Page 1 of 21 2

Return to Articles

Quick Navigation