Online Draw Situations

Some interesting situations came up recently in an online draw game. They illustrate some of the principles unique to this game. These examples are all from a $3-$6 game.

I was on the button, and everyone passed to me. I had a pair of tens, and I opened for a raise. In the strategy I have detailed before, two tens is my approximate minimum opening requirement when I'm on the button. There are exceptions, but this is in general.

The small blind folded. SuSuSuzie called from the big blind. She drew three cards, and so did I.

SuSuSuzie is a weak-tight player. She would bet after the draw in a raised pot only if she improved to aces up or three of a kind. With any two pair less than aces up, she would check. She would call after the draw with some pairs and any two pair. Bluffing did not seem to be part of her strategy. Since I knew how she played, I would not bet my own hand unless I improved. Since she would call my raise-open initially with almost any pair, if I did not improve my pair of tens, I would just show down the hand – and win more than half the time. A pair of tens is higher than a random pair about two-thirds of the time. Deuces through nines is eight possibilities, jacks through aces four, and tens are 50-50. The smaller pair improves approximately 30 percent of the time, making the unimproved pair of tens about a 5.6-to-4 favorite.

My edge increased due to my knowledge of her play. If she bet, I would not call with two pair smaller than kings up. I would raise with three tens. I knew she would not then reraise with worse than three aces. (Yes, she would not put a third bet in after the draw with three kings.) So, if she reraised, I would know three tens were beaten. Most other regulars in this game, even if their strategies approximated mine, would call with tens up if she bet – but I would not – and if they raised with trips, would call her reraise. Money not lost calling spends just as well as money won.

After the draw, SuSuSuzie checked. I caught a pair of threes for two pair, and bet. I could lose, because she would pass with hands better than mine, but I was better than a 2-to-1 favorite. The 2-to-1 came from the a priori chances of my pair being higher than hers. The odds improved, however, because she would also call a bet after the draw with several hands that did not improve – namely, aces, kings, and sometimes worse. Actually, she would call more than sometimes, because she is very suspicious of my play and thinks I bluff considerably more than I actually do. In this exact situation, I would never bluff. I would always have at least a pair of tens, and would always play them as described. Math serves me much better than psychology here.

SuSuSuzie called. I showed my hand, and was awarded the pot. She didn't have to show her cards, and I then could have requested a hand history if I wanted to know what she had called with. I didn't need to, though, because she showed her two pair, eights and sixes. She likes to show losing hands, as if she needs to prove to someone that she didn't make a bad play by calling. Or, maybe she wants to demonstrate how bad her luck is, as many players seem to do.

I didn't care why she showed her cards, but, in fact, she had just demonstrated how poorly she had played. If she was paying the least bit of attention, she would have observed that on her big blind, I never came in with less than a pair of tens. (Of course, most players are not observant.) She would know that it was almost certain that her pair was smaller than my opening hand. The very best she could be was about a 3.25-to-1 underdog. That's a pair of eights against a pair of tens, a comparison I ran with Mike Caro's Poker Probe. Against higher pairs, her chances decreased. Some percentage of the time, I would have two pair or better. Part of the time, my two pair would be smaller than her pair – say, fours and deuces against her pair of eights – in which case her chances would actually improve to only about 2.75-to-1 against. Part of the time, my two pair would be higher than her pair, in which case her chances would drop to somewhere between about 4.5-to-1 and 5.7-to-1. When I had trips or better, her chances would drop significantly. Given that, her overall chances were worse than 3.25-to-1 – more like 4-to-1. Also, the times that her good hand would be beat by my better hand would happen more often than the other way round – just because my starting hand was better than hers. The big confrontations she lost would cost her more in general than her profit from those she won. Her winnings would also be shaved somewhat by her after-the-draw conservatism. I would be far more willing to push a good hand than she would. With all of that, her overall underdog odds were probably close to 5-to-1. But when I opened for a raise, she was getting only 3-to-1. That made calling a play with considerable negative expectation.

The preceding is a somewhat lengthy detailing of one justification for the truism that playing shorts (small pairs) in draw poker is a losing proposition.

On my button, everyone passed. QueeQueg99 had the big blind. I classified him in my player notes as tight-aggressive.

I raise-opened with three tens.

The small blind folded. QueeQueg99 reraised. I knew he would do this with any two pair or better. He would probably only call with one pair of jacks through aces, and would fold anything smaller. My hand would be better than his most of the time. I capped.

He drew one card. QueeQueg99 could not be drawing to a straight or flush. He might call, but would not raise. He could have trips, of course. Like many players, he often drew one card to trips to disguise the hand, but two pair was more likely. So, I drew one card, also. I knew if I drew two cards, he would check after the draw no matter what he made. He would call with two pair, because most players tend to disbelieve two-card draws. They think that if they always have a kicker when they draw two cards, so must everyone else. And he would check-raise if he made a complete hand, thinking that if I drew two cards and really had trips, I would bet, and if I didn't – and thus was bluffing – I would still bet. But I didn't want him to check; I wanted him to bet, because it was my intention to raise.

Yes, he could have me beat already. But, except for quads, only four hands – trip jacks, queens, kings, or aces – had me beat. Combined with the chances of him completing his hand, one in 12, I was still nearly a 2-to-1 favorite to end up with the best hand after the draw. Thus, I drew one card. I kept my three tens plus a 7. I discarded a queen, making it slightly less likely that I was beat already (since his chances of having three queens were lessened). I decreased my chances of improving by not taking two cards, but increased my chances of winning two bets versus one after the draw.

I caught a queen, which made me wince slightly, but missing my chance to fill up was only a small downside compared with the value of disguising the hand. And both of us would complete our hands only about one time in 144 that the situation came up.

QueeQueg99 bet. I raised. He called. I showed my hand, and took the pot. QueeQueg99 did not show his cards, but I requested a hand history to see what hand I had beat: two pair, aces and kings. I made an extra $6 by playing the hand as I did, since he would have checked had I drawn two. If QueeQueg99 had had a smaller two pair – say, tens up or lower – he still would have checked against my one-card draw, and my play would have made no difference. But for those times that he had a better yet still not as good a hand as mine, I had maximized my after-the-draw profit.

I was in the cutoff seat with a pair of jacks. SwinginSammy limped in to my right. He was a player I characterized as weak-loose. He would open with any pair and any draw to a straight or flush. Overall, my hand was better than a 4-to-1 favorite over his, so I raised. Sure, he might have queens or kings, but probably not aces, because he usually raise-opened with aces. He also might sometimes slow-play a big hand, but the odds were decidedly in my favor. Two jacks was also a favorite over three random hands, and I didn't want to let the button or the two blinds in cheaply. In particular, I didn't want to give the big blind infinite odds to beat my hand.

The button and small blind folded.

Few2, the big blind, called. Few2 I characterized as tight but tricky. He would not reraise with most two pair, but would likely check and call after the draw. He would probably reraise with trips, although not always with small trips. He would also call with many pairs, maybe from about sevens on up, plus any straight or flush draw.

SwinginSammy called.

Few2 drew one card. That didn't tell me anything more than he didn't have precisely one pair.

SwinginSammy took three cards.

I took three.

After the draw, Few2 checked, as did SwinginSammy.

I caught two fives, giving me jacks up. I knew that if Few2 had started with two pair, he would call. If SwinginSammy made two pair, they would likely be smaller than mine and he would overcall if Few2 called, and call, otherwise. These factors all pushed me in the direction of betting as opposed to showing down. You may say, "Duh, of course you bet; you improved." I certainly agree, but many weak-tight players do not bet two pair in this situation. They're afraid of being beat in a showdown, or, worse, being check-raised. Of course, those same players do not raise with just a pair of jacks before the draw, either. So, I bet.

Now, something unexpected happened. Few2 raised, and SwinginSammy folded.

I thought about this for a long time – well, 10 seconds. That's a long time online.

Here's what went through my mind: Few2 might check a big hand if I drew one or two cards. Being a tricky player, he was quite capable of that. But what could he have that he would check into two three-card draws? If he had completed his hand, he definitely would have bet, because he would have been afraid of the hand being checked around. Pure mathematics showed that more than half the time, neither SwinginSammy nor I would improve. Many of the times that SwinginSammy improved, he wouldn't bet anyway. SwinginSammy rarely bet two medium pair or smaller in a raised pot, and Few2 was observant enough to know that. If Few2 had started with two pair, he would not bet without filling up. So, in the case of any completed hand, he would bet in this situation, because he would want to be sure to get one and maybe two calls – and maybe even a raise. He knew that SwinginSammy would call with two pair, but if I bet and he check-raised, SwinginSammy would give up. It wasn't that Few2 was incapable of a check-raise, as he had demonstrated that to me many times. It was just that a check-raise in this exact situation made no sense whatsoever, but a bluff did.

It was a lot to contemplate, but my call came fairly quickly. Few2 could not muck his hand, even though I am sure he would have liked to, because he was first to show. I was not surprised to see that he had four spades and a diamond. My two pair took a nice pot. I won two more big bets than I would have had I just shown down the hand, all because of an illogical check-raise.