I can think of times when playing 5-3o is reasonable:
When it’s very cheap to do so (ie, you’re in BB and everyone who enters the pot just limps, or when you’re super deep stacked and it’s only a couple of blinds, so speculating on a weak hand that has the potential to surprise can be a good thing, for the once in a while monster draw that you might make with it.)
When you have no choice (ie you’re super short stacked and all-in or mostly committed due to blinds/antes). I got QQ cracked by 74 hitting a 3-7 straight like this, and I see it all the time on Replay that two weak rags forced all in hit miracle boards. It’s unlikely to win, but you have to defend when you’re short stacked, or you just get run over.
When you likely won’t have to show it down to win the pot (against an isolated known weak player who over folds and is easy to bluff). Picking starting hands intending to bluff with them is not a great way to bluff, though. Good bluffs are sold by playing the entire hand consistently, in a way that convinces your opponent that you likely have them beat, when they in turn have played the hand in a way that suggests to you that they are on a weak holding.
I agree about the importance of unpredictability, but just like anything in poker there needs to be balance. If you can read your opponents perfectly, you can win without even looking at your own cards. But being a winning player generally involves choosing your spots. Generally, I would not choose 53o as a bluff open because there are many other hands that make better bluff hands, and if you are choosing a hand as weak as 53o then you are likely to be bluffing too much. I would much rather bluff with 53s because 10% of the time you will flop a flush draw that gives you some equity.
For the most part, I have no problem with opponents knowing that I do not have 53o in my range. The frequency of situations where that would impact their decision-making is tiny compared to the number of situations where 53o is a losing hand. In some kinds of heads up GTO situation against an extremely balanced yet unpredictable opponent (like Doug Polk?), maybe 53o needs to be played; I am not a good enough poker player to pretend I know how to best approach that situation.
But in general, the play chip/microstakes/low-stakes population loves to call too much, so why try to bluff with really weak hands like 53o?
Your AA example is a great example of the importance of having some kind of balance. Obviously, tournament situations are dynamic, so I am not going to try to define what your range should be, but if you are only 3-betting AA/KK then your opponent’s should almost always fold (unless they are pot-committed or your 3-bet is too small). That shows the importance of throwing in bluffs or balancing your range. If your opponents are going to open a range that includes all pocket pairs or hands as weak as say AJ or ATs, you can three bet hands like JJ+/AQ for value. That way you can have a range wider than KK+ without having to even include any bluffs.
The trap that people fall into on Replay is playing the way everyone else here plays. I think the most important advice is: do the opposite of what everyone else does. Specifically, players here will limp their entire range and try to see most flops. Even at the highest stakes 9/10 players will open limp and limp more hands than they open. When people open-raise it often means they have AA/KK or at the very least AQ+/JJ+. Against these players you should mostly just fold and never bluff 3-bet because they are playing extremely predictably, but you can profitably set mine with pocket pairs.
But my actual advice is not about other people’s ranges, it’s about what you should do. Do not limp. Never open-limp (be the first limper). It can be ok to limp behind, but you should do it rarely (let’s say 1/10 hands you play), not 9/10 hands. If you have a hand you want to play you almost always want to raise. You can look at hand charts to get a basic sense of where to start, but you can open a hand like 33 from UTG in 6-max. You can open 87s from the CO. Having a reasonable range puts you at a huge advantage over the field. Simply by opening more often you apply a great deal of pressure to limpers and potential limpers. You can win dead money, you get the initiative in the hand, and if one of these nits 3-bets you can easily fold (or set mine). By opening a normal range you are already standing out from the crowd and making yourself unpredictable without have to play every single trash hand like 53o. You may find yourself in new and uncomfortable postflop situations, but that is part of learning how to play poker and refining your range. But I seem to have gone on a tangent…
WOW, I don’t think I’ve ever seen a more condescending, self righteous, self-congratulatory statement here on RP! Being that rigid and predictable will get you killed at the table.
If both Polk and Negreanu have poker strategy including hand ranges , their little battle must be pretty close to dead even. Is it?
Superstitous? Polk must be doing a lot of voodoo to be that far ahead playing Negreanu…
“Poker is unpredictable, cry me a river.” That is what I was what I was responding to, which is quite condescending and self-righteous. But apparently I was correct that you were triggered, based on your response.
I assume you are trolling because from an actual poker standpoint whatever you are saying is incomprehensible. You think Polk and Negreanu don’t both “have poker strategy including hand ranges”? That is hilarious. I am not even sure I am understanding what you are saying because that couldn’t possibly be it. Their hand ranges aren’t static, but are flexible to specific situations and table dynamics, but they 1000% are thinking about ranges all the time and have an extremely strong understanding of ranges. That is part of why they are good players. If they didn’t, they would not be able to be winning poker players, let alone professionals, let alone top professionals.
Because analogies are fun, here is one: what you seem to be saying is that because an elite chess player makes a creative, unorthodox move, that they clearly aren’t just using rote memorization of opening strategy. To be an elite chess player, memorizing opening strategies is a fundamental part of learning the game. Similarly, in poker, learning about hand ranges is a complex yet fundamental part of learning the game. Players start by learning basic default hand ranges, but range-based decisions become almost infinitely complex when considering past actions, board textures, and balance. The best players are the players who understand ranges the best. If an elite player makes a move with an unorthodox hand (like 53o), they are using their understanding of the entire situation (including their opponent’s range and their own perceived range), not simply saying “oh well, hands flop weird ways, let’s just see what happens, lol”.
I think you are making a great point, except it is the opposite of the one you intended to. Both players understand ranges very well. However, Polk’s understanding is far superior to Negreanu’s. At one time Polk was the top HUNLHE player in the world. Negreanu was never even top-tier in HU poker. This is why over a large sample of hands, Polk will win the match, period. It is his superior understanding of range interactions across board textures that give him a permanent edge in this match.
Everyone can play their cards any way they want to. People can have favorite hands or superstitions or whatever else they want to incorporate into their games. Its the reason there is money to be made in the game. If everyone was playing well, we’d all lose to the rake and no one would play anymore.
As I mentioned above, 5/3o in a HU match is going to be borderline if everyone is playing well. If people are playing for hand strength alone or are being overly passive (or aggressive), this hand becomes strongly negative. In fact, many hands that people think are EV+ actually aren’t for them. Suited connectors as cold calls kill people’s winrates because they tend to play them terribly. Most people should fold small and medium pairs rather than cold call with them for the same reason. In many low stakes games, people would do far better only playing 88+, AQ+ and folding everything else. That’s just the data and data don’t lie across large enough samples.
I think Polk has been widely felt to be a heavy favorite, but I don’t think it is accurate to suggest that he was 100% likely to win the match, just because they agreed to play a pretty large number of hands. I’ve seen some statistical analysis that had a pretty reasonable percentage of outcomes with Daniel winning, even assuming the skill gap was quite large in Doug’s favor. I’m not sure how accurate that was, but it felt reasonable to me. Even over 20,000 hands, only a fairly small percentage (I imagine less than 1%) of those hands will get full stacks into the middle. In that smaller sample of 100 to 200 big hands, luck can have a pretty sizeable effect. And in general, I think volatility in heads up play seems quite a bit higher than at say 6 max or full ring, especially with pros that will be throwing in 3 bets and 4 bets and 5 bets pre-flop a fair amount.
I have not seen any of the matchup, but I agree with your points. Even a swing of several dozen buy-ins over 20,000 hands does not necessarily equate to a huge difference. Heads up poker has a ton more variance than larger games and playing a lot against only one opponent is bound to make the dynamics and patterns go in weird directions.
More broadly, when talking about skill advantage one player may be better at heads up while another may be better at full ring, or one player’s style may match up particularly well against the other player’s, while they may both perform differently against another opponent. Poker is so full of variance, it is difficult to draw conclusions. All of that being said, it seems like game theory is driving play in a particular direction, which is still full of variance but could somehow cut down on individual variability of play styles.
I’ve always had the impression that DNegs was old school and less mathematical than Polk, so not surprised that Doug would do better.
The larger the sample size, the more certain it is that the player with the skill edge will win. The betting lines all had Polk winning handily. I think pre match, the lowest predicted rate for a Polk win was at 5bb/100. Is it 100% certain that a >5bb/100 winrate will in fact win over 25K hands? No, but its close enough to make the point - a permanent skill advantage over infinite hands will generate a certain winrate.
ADDED: just ran a sim using a 5bb/100 winrate over 25K hands. It is a statistical certainty that this winrate will produce an overall win until the standard deviation reached 25bb/100. At a SD 5x the winrate, the worst case scenario would be for the advantaged player to lose a total of ~120bb for the entire match.
Another factor which is hard to calculate is that Negreanu and Polk are not static entities. Doug has acknowledged that Negreanu has been playing a much better game in their recent sessions, as he has come back to reduce his losses from close to 1M down to about half that.
I think it’s been an interesting and fun match to watch, at least the highlights that I’ve been able to catch.
It’s also worth noting that Negreanu won the opening session, which was played live. You can’t make too much of this, due to the small number of hands, but even playing a superior GTO strategy, Polk could be at a disadvantage against Negreanu at a live table game, if DNegs has good tells on Polk. And Negreanu is one of the best there is at reading. It seems clear that Polk is better at heads up GTO, but he also benefitted from a couple of huge sessions where he ran spectacularly good. I read an article sometime last week that said Negreanu’s been running almost 300k under EV for the way he’s played his hands. So if we adjust for that, it would have Polk up still, but only by around 200k. Which, is obviously a significant amount of money, but I think this match is more evenly balanced than a lot of people think.
FWIW - If DNegs was running $300K below EV, as of today, he is losing at almost exactly 5bb/100 ($330K over 16,500 hands at 200/400). That was the best case predicted scenario according to the betting lines before the match began.
I don’t think DNegs is running that far below EV but maybe he is. I’ve watched a good amount of the hands and seen the data available and think DNegs is running very well. Its not a popular opinion but I think the skill differential is far greater than most everyone estimated before the match. Yes, DNegs is getting better but its just not close. People want him to be better than he actually is because he’s a hero to many poker players. That leads to observation bias.
IMO, it is pretty remarkable that DNegs is even mildly competitive in this match. Yes, Polk is out of practice and wouldn’t be near the top HU player in the world if he started playing again now. Still, he was a HU specialist and was in fact the best there was when he did play. DNegs is a top level tournament player. He’s a pretty good cash game player. He is not a great cash player and he certainly isn’t a top level HU player. If DNegs were to play the top HU players right now, I think he’s be run over.
He’s probably made up more ground since the story i read that in was published. I think at the time it was published, he was just starting his comeback, and was down around -700-800k, from his bottom of -970k. Adjusting that by 300k, he’d still be behind about -400-500k. He’s obviously been motivated to improve his HU game, and I think it sounds like he’s done so impressively quickly. He’s not just losing less fast, but had had a string of winning days for several days in a row, and that’s an impressive turnaround.
I haven’t watched as much as you have, from the sound of it,but it’s been a great story to follow.
“Poker is unpredictable, cry me a river.”
Sorry JD but that isn’t condescending but a play on words.
The last card is called the ‘river’ and it can make you cry if it turns your hand.
You seem to be very sensitive about being ‘right’ about your knowledge of poker…