April 25, 2012
If I could redesign NLHE...
So I posted about my hypothetical poker variant a few days ago (NLHE played with 2 card flop, 1 card river, and new 4 card hand rankings), and although some people liked the idea, it didn't generate that much discussion, which I was hoping for.
The concept was interesting to think about though, because it's good to break down the fundamental aspects of a poker game to see why they play like they do, and what features create an interesting game.
Every game can be broken down into a few main features:
- The betting system. Limits (NL, FL etc.) is an obvious and clearly stated one and plays an important role, which is shown clearly by the wild differences between NLHE and FLHE. Amount of betting rounds is also important. A lot of games are regularly four street games, while Stud variants are usually five, 5 Card Draw is merely 2!
- The source of information available. This is everything that helps to define the ranges of each player and is mainly just up cards like in Hold'Em or Stud. It can also be the drawing aspect in Draw games.
- The equity distribution between hands and ranges at each point. This is probably the most important factor and also very complex to figure out, though things like Stove have come a long way for HE.
So far, I've been talking mostly objectively about game features. However, the concept of what makes an interesting game can be somewhat subjective. In my opinion, a game is more interesting the more decisions a player has to make. This means that the best action in each situation has to be unobvious to the player. Each of the features can be designed for this goal:
- Betting needs to allow for viable choices for the player. Take Fixed Limit Omaha, a game that is never played. In such a game the limit bet sizing is too small that it has removed the option of folding - your default strategy becomes to call down in a lot of situations that it doesn't leave you any room to make a real decision.
- The information available needs to not define the opponent's range too much. Take a game like 5 Card Stud, with 1 down card and 4 up cards. The only information hidden is a single card out of a 5 card hand, meaning what he could potentially have is severely limited in a lot of situations. If his board is weak and you can beat him regardless of what his hole card is, then you don't have a decision, and if your board automatically beats what his 5 card hand, neither does he.
- The equity distribution goes very much hand in hand with the information available. In the above 5CS example, one players displayed range could already be 100% versus the other player's range regardless of betting. Generally, high equity favourites will limit the option of aggression for the equity underdog. This will leave one player calling and folding, and the other betting or checking, which means way fewer decisions. Another important aspect is how the equity switches between players per street, and this is best illustrated when the nuts has changed on a new street in PLO, and drawing player's range now looks stronger. This concept shifts power away from the person with the uncapped range on the previous street and gives power to someone who has a capped range, as he now has a new chance to be aggressive.
Anyway, this brings me to the main problems with NLHE. First, I should point out that it has a ton of good features, and that is why it became popular. One of the best parts is that you give no information away about your individual hand unlike in Stud. Your displayed range is never weaker or stronger than your opponent's - you could still technically have anything and so can he. It also became popular because it's the first real big bet games that works, and the nature of big bet lead to a lot more decisions, as well as more opportunities to bluff. The game is also very simple in nature (you have 2 cards, now try to match it with the board) which is a very important feature that I'm not including under 'interesting'.
However, if I had the opportunity to slightly change it, while retaining most of it's features I'd make two adjustments:
1) Remove the turn: Yep, the turn is my least favourite part of the way the board runs out. Simply put, not enough changes on enough boards. It is uninteresting, due to what I described as the way equity switches between the players. Having one card roll off at a time means very few hands which were behind catch up - it is a feature that works very well for limit: when your bets are a small percentage of the pot, it makes sense that the chance of improving on the next card isn't that high.
How would I do it? Have the turn and river come off at the same time, or a 'two card river'. You will notice that this changes the game into a 3 street game rather than a 4 street game, which is also great. 4 streets is too many. It's hard for me to articulate why I think this, but it leads to way fewer showdowns because it takes longer to get there.
2) Remove pocket pairs from the game: Or at least, remove all pocket pairs higher than 99/TT (though this rule isn't as intuitive) If you get one, your hand has to only play one hole card. The problem with big pocket pairs is that they are way too strong, it completely limits what hands you can see the flop with. They are big favourites pre-flop, big favourites to hold up for the flop, and then often big favourites after the flop. This has created a game that is aggressive pre-flop and very commonly ends pre-flop or on the flop. If they were gone, you'd get a preflop game more similar to PLO, where you don't have favourites pre-flop that high. If you are 3-bet, you can more comfortably call with something more speculative knowing that your pairs are more often live and that you will get to see two more cards for one flop bet.
Addendum: Another way I thought of reducing the power of pocket pairs is to say that they are not allowed to interact with the board - AA must remain a pair of aces no kickers. This will mean AA loses if JT makes two pair on a J4674 board, if any Ax catches the case A, or the board ends up showing more than AA or better (a nice kick in the teeth if you flop quads and can't beat the board, but you do get to have both trip blockers). In this scenario, AA is around a 55-65% favourite to most hands pre-flop, but becomes incredibly weak post-flop, which could lead to interesting play. This would also remove all sets from the game, which I think is a positive.
-
It's important to note that none of the features of NL were a big problem until people got very good at it. When people are bad at the game, it is interesting for them under my earlier definition because no decisions seem obvious. However, now people are much closer to optimal strategy than before and it turns out, it is boring.
I think these two feature changes will make the game completely change from a pre-flop non-showdown oriented game that it is now, to a post-flop game that involves a fair amount showdowns and varied aggression. Of course, it'd never happen, but it's still interesting to think about.
12 Comments:
KasinoKrime posted on April 25, 2012 at 04:04 AM
Interesting perspective, I like it. I especially like your comments about removing pocket pairs. I agree that it's frustrating how strong big pairs are in NL. What drew me to PLO in the first place was the amount of postflop skill it takes to be successful. In NL when I play aggressively, I feel helpless when I get 3b by tight players (or players that 3b correct ranges) because it's difficult to fight back correctly postflop. In PLO, if a nit 3b's me, I take comfort in knowing that there are ways to win the pot without actually making the hand.
Xerod posted on April 25, 2012 at 05:44 AM
"In my opinion, a game is more interesting the more decisions a player has to make."
I haven't analyzed your game enough to know whether it takes this into account, but along with making more decisions a game has to be such that a skilled player will make the correct decisions more often than an unskilled one. Money is made in poker by making less mistakes than your opponent.
That's one thing I don't like about PLO (aside from being terrible at it). It's very easy to shovel money into the middle and have no one make a mistake because the equities are so close.
Schweig posted on April 25, 2012 at 06:06 AM
Yeah, exactly KasinoKrime. The same applies to flops too - the power of top pair is way too high and anything worse almost never catches up, and I think the removal of a street addresses that really well.
I imagine if this game were played, there will be so much chasing on the flop when someone has a runner-runner straight or flush draw, except it won't be runner-runner here!
Xerod, that's a fair point about PLO. In Hold'em, the equities are too far apart on the flop. In PLO, the equities are too close together. I think the balance must be in the middle somewhere.
I think PLO would probably benefit from the turn change too. You are basically deterred from not getting it in on the flop if the pot is already sizeable in a lot of situations, so that you can get to showdown without making a mistake/being priced out on turns.
If you could get straight to the river with one call, that would making calling way more attractive than getting it in in a ton of a spots.
Schweig posted on April 25, 2012 at 06:13 AM
Good point about skilled players making the correct decisions more often than unskilled one. It was kinda what I meant by the decisions being "unobvious", but yours is a better way of wording it.
That feature is the main reason I think stud games are bad, especially Stud and Razz. In Stud, there are so many situations where your board is so much stronger than your opponent that even fish won't make many mistakes.
Schweig posted on April 25, 2012 at 06:34 AM
Thinking about whether my game takes it into account, I feel that it does because it leads to more post-flop play, specifically river play.
In NL betting, the exponential pot size growth means the river bet is going to be very important. In Hold'em, river play is very limited due to the vast amount of folds pre and on the flop, and one player is usually fairly face up on most rivers anyway and forced to take a passive line.
With the changes I proposed, you'd see the river far more often as it takes 2 bets to get to it. The equity distribution with 2 cards falling at once also means that the situation changes significantly - this will mean that either player is rarely completely face up and creates opportunity for aggression from both sides.
I think there's a ton of room for more hand-reading to be employed, and there are way more things to rep.
shuttle posted on April 26, 2012 at 00:57 AM
I think the main downside to NL is a bit like stud/razz in that it's possible to massively increase someone's win rate by just giving them a quick sheet with how to play some spots. Sure that won't be making them elite but it would make some high rake games unbeatable pretty quickly. In stud/razz that might be something like "fold when your opponents board has you crushed" (not sure exactly as I suck at stud) and in NL that might be the equivalent of giving someone a preflop opening range chart and say bet/fold when you have a strong non nut hand.
shoryuken posted on April 26, 2012 at 01:39 AM
I like where you're going with this and the other game idea you mentioned in the forums. I like your three card flop and two card river. Regarding how you handle pocket pairs, have you considered having players being dealt 3 hole cards but you could only use two non-paired hole cards? I think the equities would run closer but you wouldn't kill paired hands completely. I'm kind of new to poker so I can't do the math. Maybe the hands made would be too strong, making the river a little easier to fold? If that's the case, maybe a two card flop and a two card river?
Schweig posted on April 26, 2012 at 02:51 AM
Interesting idea. By "two non-paired hole cards" do you mean at max two hole cards like in Lazy Pineapple. It certainly could be a viable alternative, but would change a lot about the game that I wasn't looking to do - was trying to maintain the 2 cards in hand nature.
Something a friend suggested was to add four straights and four flushes as rankings, which beat pairs but don't beat two pairs. It would certainly reduce the power of aces, but I feel it puts a lot of that power in the hands of suited connectors which are already quite playable, and makes offsuit hands much weaker than before.
Luke00016 posted on May 01, 2012 at 18:49 PM
It looks like you're analyzing the game carefully, but you aren't giving much weight to the aspects of the game from a casino or recreational player's perspective.
A casino wants a game that is fast, generates rake, encourages action, and is fun for the players.
A recreational player wants a game that is 'fun'. How we define 'fun' is up for debate. I think recreational players want games that appear easily understood. I think they want games with action and are exciting. At some level, recreational players are competing, so they want a game where they can feel like champions. The funny thing about poker is that recreational players will feel like champions when they get pocket aces and win a big pot, but they'll feel the same way when they hit their flush to crack pocket aces.
How do you think these different perspectives fit in to your analysis above?
Schweig posted on May 01, 2012 at 18:54 PM
I consciously ignored the aspect of 'fun' or the appeal to recreational players, because I was specifically focussing on the game theoretical aspects. I feel I can analyse that while I would merely be speculating on what aspects would make the game popular or fun.
Schweig posted on May 01, 2012 at 19:08 PM
Here's my take though: recreational players don't survive long in NL anymore. NLHE sure was fun for them until everyone started learning the game, playing tighter, and now if you're a recreational player who wants to play loose and gambly, you'll just get crushed by people playing solid TAG.
This game is naturally less punishing to bad players given the 2 street aspect and how hands are much less behind preflop. If you want to peel 43o preflop it probably wouldn't be that huge of a mistake anymore. A game that has a lot of flops seen, a lot of preflop VPIP brings more action and is more exciting.
We actually tried this variant in a homegame the other day and I will post some results.
shuttle posted on May 01, 2012 at 19:54 PM
I think the no pocket pairs thing makes it harder to understand to a recreational player, I think amongst stronger players it might have the potential to make the game better. But I think removing the turn and having the 2 last cards dealt at the same time has to be a great thing the more I think about it, I don't see any downside to this.
Log in or sign up to leave a comment!