Many players refuse to play unless a game is nearly full. Still more will play if there are five or six players but not if there are fewer than that. Sometimes, while sitting with three or four other players who are unwilling to play until more players arrive, I ask them why they don't like to play shorthanded. One such player told me that short-handed games are too expensive because the blinds come around too often. Sometimes it's simply that they don't want to play shorthanded in a game with me in it. Some players come for the social atmosphere and find that the competition gets too personal with fewer players. Other players seem to just like to sit around folding most of the time while waiting for AA or KK. Still other players probably have other concerns.
So, why do I think short-handed games are not only worthwhile but are actually better for poker players? Let's look at the three main concerns of a poker player: winnings, social experience, and game purity.
Effect on winnings: The old adage that "you need to play more hands when you play short-handed" is overly vague. The truth is that there is a much better way to transfer your strategy from a full-ring game to a short-handed game: When playing X-handed, play exactly as you would play in the same position relative to the button in a 9 handed game after the first (9-X) players fold. As I will now demonstrate, the two scenarios are very similar.
Suppose you are playing in the cutoff or button of a 9-handed game and the first five players after the big blind fold before the flop. If you have an idea of how to play in this situation, then you can play in a 4-handed game. There are only three very minor differences in the two situations. First, the 4-handed scenario would move faster because the deal would be quicker and you wouldn't need to wait for the first five players to fold. Second, the hand in the 4-handed game would be cheaper in most casinos because in most casinos the short-handed game would have a smaller rake and maybe no jackpot drop. (Both scenarios have four players left, so you are about equally likely to win the hand and end up paying the drop.) Third, in the four-handed game, you wouldn't need to worry about any bunching effects, and so the game is slightly easier if you are trying to combinatorially guess your opponents' hands. (As we will see presently, this is probably a negligible concern anyway.) So, the game is faster, cheaper, and easier. Faster and cheaper are clear advantages. Easier is a wash because it applies to your opponents as well as to you.
So, the strategy in a short-handed game has a nearly direct parallel to a situation most full-ring players are already familiar with: the situation where the first several players fold. Consider the implications of this. The first five hands after the blind are the least-often-played hands. They are the spots where you are most often sitting and doing nothing except waiting for your opponents to finish playing (and maybe gathering information to use against them). If you could just eliminate these five spots, you would vastly increase the rate at which you would have decisions to make. That is exactly what happens when you play short-handed. Winning poker players make their money by making better decisions than their opponents, so having more decisions in any given period of time translates directly to a higher win rate.
In addition, short-handed games require players to quickly gain a good sense of their opponent's strategies, because they will be going up against the same opponents in each hand. Also, the nature of short handed games is that they tend to fluctuate in size, requiring an ability to adjust to changing circumstances. These are skills that good players can put to better use in short-handed games than full-ring games.
I anticipate two main objections to my argument that short-handed games are better for your win rate. First, the situations I've described are not exactly the same. After all, in a 9-handed game where the first five players have folded, there is a "bunching" effect: because folded hands are slightly more likely to have contained low cards, the remaining cards are slightly more likely to be high (and thus strong). Just for a quick example, suppose the first five players will each play if and only if they are dealt an ace. Suppose they all fold this hand. If you are in the cutoff and also don't have an ace, then all four aces are still left and there are only 40 cards left (52 minus 12). For each of your opponents, the chance of having pocket aces is now 1 in 130 (4/40 times 3/39), much better than the usual 1 in 221. However, this is an extreme example. In practice, players do play some low cards and they do fold with hands like A2 (most players, anyway). The "bunching" effect is just not all that big a factor. Maybe I'll do a deeper analysis of this later. For now, I'll just say that I don't believe this is a big enough effect to contradict my claim that you can view them as being essentially the same. I concede that it makes sense to be tiny bit more wary of your opponents in the 9-handed game after five folds than in the 4-handed scenario, but making a big adjustment would be a mistake. You would be safe if you made no adjustment at all.
Second, you may argue something like:
Sure, those scenarios are essentially the same, but you are overlooking some real advantages to having those early-position hands that exist only in full-ring games. These afford me important opportunities to apply my superior poker skills. For one thing, I can use the time after I fold these hands to watch my opponents play. Since I'm more observant than my opponents and since I'm better able to apply new information, this represents a real advantage to me. Even ignoring this, my basic strategy in these EP hands is far superior to some of my opponents. The biggest weakness of many of my opponents is that they play too loosely, and this weakness is most pronounced in EP. Why remove these opportunities for my opponents to play badly?These are reasonably strong arguments. It's true that playing short-handed eliminates some opportunities for strategic advantage, and for a thorough analysis, I would need to consider the value of these strategic advantages. For now, let me just give an intuitive explanation for why I don't think these arguments are strong enough contradict my claim that short-handed games are better. Yes, we are giving up some opportunities to gain value over our opponents through superior use of observation and superior play in EP. However, there is an opportunity cost to these advantages. For the benefit of getting to sit and observe your opponents, you are losing the opportunity to actually be playing more hands and applying your strategic advantages; if you really want extra time to observe your opponents, you can achieve that in a short-handed game by simply sitting out a few rounds and watching. As for your advantage over your opponents who play terribly in EP, this comes at the cost of the advantage you have over your opponents' LP and blinds strategy. Every hand we would get to play in EP is replaced in short-handed games by extra hands in LP and the blinds. Since these are hands that you are much more likely to be playing (rather than folding), it seems to me that your advantage over your opponents here is likely even greater than the advantage you have over your opponents in early position. I agree that it is much easier for your opponents to play terribly in EP, but it is also much easier for you to play great in LP and the blinds, because they often involve lots of decision making deep into hands. Playing "great" in EP basically just means having the discipline to fold a lot.
There are some rational explanations for why a winning player would have a higher EV in full-ring games. Perhaps they really aren't good at playing in the cutoff and on the button compared to their proficiency at playing in early position. Perhaps they thrive on post-flop situations that are multi-handed. Perhaps they don't have the ability to concentrate in the way that is required when you need to make many decisions each hour. Perhaps they make their profit by preying on players who play poorly in early position, or perhaps they play in a venue where such players are particularly prevalent. Perhaps their game does not lower the fees for short-handed games. If these describe your situation, you may have a strong case that you have a better EV when playing full-ring games. However, I don't think these are common characteristics of poker players or venues, and, in any case, any advantage you may have in EV per hand when playing full-ring games would have to outweigh the increased number of hands you are able to play in the faster short-handed games. A 4-handed game likely has twice as many hands per hour than a 9-handed game. Asking a player to make twice as much per hand in a 9-handed game than a 4-handed game is a very tall order.
In the end, I think that with a smaller rake and a greatly increased decision-rate, short handed games make for much more efficient winnings.
Big advantage to short-handed games, assuming you are better than your opponents in late postion and on the blinds.
Effect on social atmostphere: Full-ring games afford players the opportunity to sit back and relax. They chat with their opponents and even discuss poker strategy without worrying too much that the player they are speaking with will play many hands with them. Basically, the game is less contentious. On the other hand, they can sometimes get boring and players get impatient with each other for playing slowly.
Small advantage to full-ring games.
Effect on game purity: Smaller rake means slightly less negative effect on the game. Fluctuating game size requires a flexible strategy, which I view as a slight positive. However, some would argue that a full-ring game, with its slower pace and emphasis on patience and observation, is the way "pure" poker was meant to be played.
Very small advantage to short-handed games.
Unless you come mostly for the social experience, you should prefer short-handed games.