With good reason, most poker players cut their teeth in big-bet games like no-limit hold 'em by playing in games where stacks are relatively shallow. Small stakes tournaments may begin with stacks of 100 big blinds or more, but within an hour or two the average stack will usually be more like 30 big blinds. Small stakes cash games usually cap the buy-in at 100 big blinds, and it's common for less experienced players to buy in for smaller amounts.
Playing with a short stack can be a good way to learn: it simplifies your decisions, minimizes the magnitude of your mistakes, and helps you focus on fundamentals like pre-flop hand selection and evaluating flop textures. However, it's important to keep learning as you start playing in deeper games, because strategies that are at least reasonably effective when shallow often become liabilities.
When shallow, it's easy and generally effective to narrow your decisions down to just a few options. If you have 8 big blinds in a tournament, you still in theory have quite a few options (call, fold, raise to 2.1 big blinds, raise to 3.6 big blinds, etc.), but in actuality you can collapse that to a binary choice between going all-in or folding. You'll almost always be correct to do one of those two things, and even in the rare case that some third option might be viable, ignoring it will cost you very little in terms of expected value.
Even once you get deep enough to play a flop, you'll often have just a few choices: bet and call a check-raise all-in, bet and fold to a check-raise, check and fold to a turn bet, or check and shove over a turn bet. This does represent a significant branching of the game tree relative to a binary pre-flop decision, especially as you now need to consider not just your initial action but also your response to an opponent's action and an additional community cards. Still, this pales in comparison to the complexity of a much deeper game, where you must consider possibilities such as a check-raise followed by two bets, or an overbet after you check.
Most players quickly learn that, without further information, you can't answer a question like, “Is QJs a good hand for going all-in before the flop?” You'd need to have more information about the size of the pre-flop pot, the size of your stack, and your position.
However, not so many seem to recognize that a question like, “Should I continuation bet Q[club]Q[spade] on a 9[heart]8[heart]7[diamond] flop?” depends on many of the same variables. If your experience is mostly in shallow-stacked games and/or against extremely straight-forward opponents, this probably seems like an obvious bet to you. Your hand benefits from protection (“I want to take it down now” is how I often see this expressed, though that's not quite right) and, in a sufficiently shallow game, can still be in good shape if all of the money goes in.
That last point is really important and gets at why betting is more hazardous in a deep-stacked game. Betting is probably correct if you can profitably call a check-raise. It may even be correct to bet if you can easily fold to a check-raise, as this will also mean that you are unlikely to face a check-raise. It's not easy to out-flop an overpair, even on this flop, which means that a player who only check-raises with two-pair or better will rarely check-raise.
However, against an opponent who will check-raise a good mix of bluffs, semi-bluffs, and strong hands, you can expect a check-raise somewhat often but also expect to have no good options when it comes. Folding, of course, will have an EV of 0, but calling is unlikely to be much better. In essence, the check-raise reduces the value of your hand to 0, or almost 0. In such situations, it may well be the case that you can find higher EV options elsewhere in the game tree.
Deeper stacks require considering a wider array of options. Your choice is not simply between bet-folding or bet-calling. If you bet and get called or raised, you'll need to make meaningful choices on the next two streets, and it's best to plan for those before betting the flop. You need to consider not just a betting and a checking range, but a bet-bet-check range and a check-call-call range and a check-call-fold range and a bet-check-bet range and etc.
On top of all that, you need to consider how those ranges will change depending on the turn and river cards. It's easy to see why people prefer just to play the simple, familiar strategy they would play with a shallow stack, but it's also easy to see how much they're neglecting when they do that. A flop strategy that works for you with a 20BB stack will not necessarily be profitable with a 100BB stack, and it's extremely unlikely to be optimal.
This kind of multi-street planning can change your flop decision in other ways as well. With shallow stacks, it's more frequently correct to check and fold weak hands on the flop. There simply isn't room to push opponents off of moderately strong hands like top pair.
Deeper stacks, though, offer the potential for multi-street bluffs, not to mention some really high implied odds should you happen to back into an unlikely monster hand. There are many situations where a single bet or raise might not be immediately profitable but will set up enough profitable turn and river barrels that it ends up having positive expected value.
If this article so far has convinced you to start thinking about your own deep-stacked play, great! That's the first step.
The second step is to recognize that most of your opponents aren't doing this kind of thinking either, and they're opening themselves up to a lot of exploitation as a result. Remember how we said that getting check-raised can kill the value of some seemingly strong hands? If you know that your opponent is auto-betting hands like those on dangerous flops, check-raising aggressively may well be a profitable response.
To players accustomed to shallow stacks, trying to bluff someone off of an overpair may seem recklessly aggressive. This is generally the result of a failure to think across multiple streets. “Will my opponent fold an overpair to a flop check-raise?” is not quite the right question to ask. Rather, you must ask, “Will my opponent stack off with an overpair if I continue applying pressure on the turn and river?”
Recognizing these opportunities, in turn, can be the difference between a profitable and an unprofitable call before the flop. In other words, players accustomed to playing with shallower stacks might habitually fold 75s to a single raise, even on the button. In such games, this will often be a slightly losing call.
However, with deeper stacks and a willingness to take advantage of them, it's entirely possible that you can find a few extra bluffing opportunities, as well as additional implied odds with your strong hands that will swing this from a slightly losing to a slightly winning call.
Conversely, if your opponents are able to make such calls profitably, then certain early position raises that might be profitable for you in shallower games will become folds when deeper. Classic examples of this include offsuit Broadway hands such as Ace-Jack or King-Queen.
This is in no way an exhaustive list of the kinds of adjustments that should be made when playing with deeper stacks. It barely scratches the surface. My goal is only to convince you that there is more to playing with deep stacks than simply adding turn and river tactics onto an existing strategy for handling pre-flop and flop decisions.No-limit hold 'em really is a completely different animal at 200 big blinds than it is at 30 big blinds. Succeeding in deeper games requires re-assessing all of your decisions, starting from before the flop. Failure to do this means not just missing out on profitable opportunities but opening yourself up to exploits that may well turn a winning strategy into a losing one.