One of my main interests before programming was poker. I spent more than a year playing online poker quite intensively. A long time has passed since then, so now could be a good time to write what I think about it. As I'm removed from the game and don't care for it to the same extent, the bias about the opinions could be quite small.
Full disclaimer that, while I was not a losing player, I did not manage to move up the stakes. So take everything I say with a grain of salt.
There's something really fascinating about the game. Money might be one of the things that attracts people in the first place, but I don't really care about money in that scenario. It turns out that the strategic aspect is 100x more interesting than the money.
This is a slightly tricky subject. Poker is in the gray area. Meaning that, for people who perceive this game to be gambling, it will most definitely be gambling. One of the fastest ways to lose money is to gamble, that's why there will always be stories like
Yeah, I went to gamble in poker and lost 10k in one evening. WHOOPS!
People who think that poker is a gambling game are called whales, and it's very easy to extract value from them. As the old saying goes, if you don't know who's the sucker at the table, you're the sucker.
Expected Value (EV) basically means:
How much value you can extract in situation x
Gambling is considered bad because it has negative EV. If the odds of a coin flip become 49/51, it's going to become an EV losing game. Same applies to roulette, same applies to slots. These games don't need to have 30/70 EV odds to become gambling. All you need is 1% of negative expected value for something to become unprofitable in the long term. That 1% is where all of the winnings of the casinos come from, I'm assuming.
If it's aces vs kings pre-flop all in type of scenario, your chance of winning with aces is about 80%. This means that if both players have $10 and aces win, it seems like you win the $10, but in terms of EV you win only $8.
As the game is based on incomplete information, most of the decisions are not that simple. There are a lot of situations where players fight for assumed 60% positive EV. The hard part of the game starts when you're in situations where the assumed EV is slightly higher than 50%. In theory any action that has positive expected value is a profitable move, but once there is a fight between very thin margins, variance kicks in quite fast.
Variance manifests in multiple ways. The worst possible card can come on the river, fulfilling the range of the opponent. Aces versus kings can lose 10 times in a row. This is where tilt can kick in.
Tilt is a fancy term that means:
Deviation from the optimal strategy due to emotions.
Optimal strategy is a slightly tricky thing in this case, tho. It is possible that what you perceive to be the correct move, may in fact not be. But for the sake of definitions, we can say that optimal strategy = what you think is the most optimal action, based on what you know currently. There can be instances where you make an incorrect move because of tilt, but it turns out that it was actually the correct move. A rare-ish situation, but can happen.
Tilt is one of the things that contributes to your winrate the most. Tilt is such a big part of the game, that there are books, podcasts on tackling it.
I don't think that tilt is a problem with which software people deal that much. And just because you can control tilt in poker, that doesn't really mean that you will tackle it in real-life scenarios. Knowing that it exists is good and may knock off some negative EV points from decisions you make in real life, but you definitely won't become immune to it.
There seems to be a massive difference between the average skill levels online vs live players.
People who are average / not-losing online players are by default 3x better players than the average live player. Over a very very large sample size of hands, average online players will absolutely demolish average live players. This can be explained by the fact that online players can play a lot more hands than live players. My memory is a bit shaky on the details, but an online player playing 12 tables at once can go through roughly 700 hands in an hour (or somewhere in that range), while live players go through ... 20. This is a massive difference. Thus, skill levels improve at vastly different speeds.
As variance is a big part of the game, that 3x online player can lose to average players in live situations easily. In terms of sample size, let's say a full live game could have about 100 hands per person. 100 hands is not even a drop in the the ocean for variance. It literally doesn't show anything. You can't evaluate a player on how good he is based on how much money he made / lost in 100 hands. These sample sizes are meaningless.
It's quite frequent that a player can lose over a stretch of 10k hands. Even bigger sample sizes. Either due to variance, bad play or tilt.
Casual players will think that John is good because he won the two previous games in a row, but those opinions are white noise. Even louder white noise can be heard when players who limp by default start to talk about strategy.
In terms of variance, tournaments are more swingy. From what i understood, tournaments also have a lot more bad players in them, that's why they might be a bit easier than cash games. I played exclusively cash games, because I liked them more, but tournaments might be a better option in terms of the amount of money you can win. 9max cash games are tight. 6max cash games are filled with a lot more aggressive players.
There's no money. There's only how many big blinds you have. The decisions you make are based on how many big blinds you have. If it's an auto-rebuy online cash game, then those decisions stay quite similar in nature. If you're a tournament player, then your gameplan can change almost every hand. If you've watched live tournaments, then people going all-in with 25 million chips pre-flop seems ridiculous, but in reality, they might have 3 big blinds left and one of them is already in the pot, so that would be the only move they can make to survive.
Just like in software, books on a subject become outdated fast. Fundamental stuff holds up longer, but a lot of the strategy mentioned in some books becomes exploited. Not all, but most.
If you're playing against a regular online player, you will find that they use the same strategies that you read about in some random online article. For example, they might have read to check-raise on turns after defending blinds, if the stealer checks on a dry flop. Reversing the tale of the opponent is quite important. If the tale he's trying to tell you doesn't make any sense, you can throw out some wild bluffs to punish them.
As you don't know what your opponent has in most situations, you have to create a range of possible hands they may hold, based on what they did and didn't do previously. There are multiple factors that can help you to narrow down the range of the opponent, to exploit it better.
One of the indications on how wide the range can be, is based on their playstyle + the position. For example (assuming it's a 9max cash game), if the player is using a TAG strategy (tight aggressive), then his opening range for the UTG position (under the gun, first position after the big blind) will be pretty tight. If it's a LAG (loose aggressive) rasing from the button (last position before blinds), then his range will be a lot looser.
If the button knows that the players in the blinds are nits (people who fold when they shouldn't, and don't fight back), then he can profitably open like ~70% of hands to steal blinds. If you're in the big blind, and the button opens that frequently, then you have to use a wider range to resteal. If you're a TAG trying to steal blinds from the button, and you know that a there's a LAG defending from big blind by re-stealing 53% of the time, you can 4bet more frequently to re-steal the re-steal.
Once you go to high levels of the game, Occam's razor seems to be broken. If you watch some moves made by professional online high stakes players, you will find that they make a lot more decisions that seem unintuitive and quite wrong, but in fact will be correct.
If you want to improve your skills, you do have to question a lot of the moves you make. Fighting against your intuition is a hard thing to do. Especially when there's money involved. Resetting habits that have some amount of reinforced award is a hard thing to do.
Over a large sample size, player A only wins when player B makes a mistake. That's it.
The golden age of online poker was when it was new. That's why there are more stories of good players making good money at the start of the online boom. What happened in reality was that a small percentage of players could extract a lot more value from extremely bad players.
As time goes on, the average skill level of online poker players grows steadily each year. This means that, you have to become better each year, just to stay at the same place. 9max online cash games are filled with extremely tight players that are hard to exploit. It's extremely hard to extract value from players who are not there to lose money. Sure there are still ways to win against them, but the margins are definitely limited.
There will always be players who can be profitable online players. There will always be players who can make a lot of money in online games. Isldur won't have any problems winning in current games, I'm assuming. But the thing to consider is that they have to be extremely good (and even better than that) to hold that title. Sure, the same thing applies to stock traders, or any other profession, but man, you have no idea how good you need to be to succeed in these games in the long run.
There's also the dependency of whether or not the online games you play are illegal. Pokerstars can be banned in a single evening, and you can lose everything. Sure, there are ways around local restrictions, but they are always on the sketchy side. And I don't like doing sketchy stuff. So that's a consideration.
This concept of expected value has stuck with me. It's definitely a helpful way to think about problems, outside of poker. One of the ways it has manifested in software, has been by saying no to stupid ideas I have about stupid projects I could create. Engineering for engineering sake seems to be a problem in software development. Some projects just have a higher expected value than others. Making the best possible EV decision is one of the hardest thing to do. And it seems like most people don't think about that stuff too much. When it comes to thinking about EV, paralysis by analysis is not one of the worst things that can happen, if you're dealing with long-term problems.
The mental model of long term thinking is also quite valuable. Being less prone to variance and not deviating from the correct strategy because of bad short term results is a useful skill, in any scenario.
If you want to become better at something, you have to start evaluating the correctness of a perspective. That's why I'm quite skeptical of people who seem to be 100% sure that they are right, in certain areas. Especially when they don't have any background in them.
At the end, the long-term winners will be people / companies who make the most optimal EV decisions, in comparison to their competition. Have enough leaks in your stategy that are easily exploitable and you will fail. Focus on metrics that don't mean anything and you will fail.
Would I recommend online poker as the main occupation? Most likely no. You don't have to just be slightly better than an average online poker to win actual money. You have to be 3x better than them. And it's extremely hard to become one, because the regular online players are already pretty good.
It seems like poker is all about a fight between ranges. There can be so much nuance for a lot of actions. People who watch UFC and don't know anything about martial arts will see two people fighting. At the same time, a professional commentator will see every detail about each action both opponents make. You can't really appreciate things you don't know anything about to the same extent that people with knowledge in that area can.
You can't appreciate poker if you think it's just gambling. You can't appreciate UFC if you think it's just two guys trying to make the other one go night night. You can't appreciate 3D art if you haven't tried to create something in Blender.
Maybe one of the sub-goals in life is picking up as many hobbies as you can, so that you could see what that MMA commentator sees when two people fight in a cage, to a closer degree. That appreciation for how much effort needs to go into developing a high level skill is definitely a valuable thing. Knowing that you won't understand 99% of the things that are going around you is a magical thing, in a way.