Let me be the first to point out there's a HU forum section and this isn't the first HU SnG thread. Nevertheless...
I usually play 2 tables, gives me enough time to concentrate on both opponents' game and lines, and make notes on them. Unlike FR, there's no solid proven-to-work winning strategy for HU. You have to adapt to the opponent, that's how you win. You can't just rely on automatic decisions and usual lines, so you can't proportionally increase your winrate by adding more tables. Every opponent is different and requires attention to his game. I never played more than 4 HU tables, and wouldn't advise (relative) HU beginners to play more than 2 simultaneously.
A HU match is usually decided in one of the three ways:
- opponent bluffs till the bitter end against your strong hand (tilt is often the cause). It is important that you are fully tilt-resistant at all times. If you can put your opponent on tilt, the game is half won;
- opponent's strong hand loses to your stronger hand. Pot control is important, you don't want to take your bluffs too far without reads. Only go all-in with real good hands, a simple advice that many of your opponents will ignore;
- blinds grow too big so you have to gamble if you want to win (starting 50/100 for a 1k5 stack). Generally you should have outplayed your opponent before this phase, but if you make it here, just use ICM;
HU can get really swingy. I once played a total fish, either that or he was on complete mental tilt (probably both) and literally went all-in every hand. He later stopped pushing all the time, but still about 1/4 of all hands. 5 times I called his push, 3 of which on the first hand. Every single time I had the best hand, overcards, domination, overpair, everything just lost to a suckout regardless of odds. Cost me over $125. And the prick just kept laughing in the chat, probably thinking he actually is a good player. That's how variance goes.
As for the limits, unlike FR/SH the skill isn't proportional to the limit. It is much more opponent-dependent than limit dependent. Granted, there are more fish on lower limits because they can't afford more expensive games, but that doesn't mean higher limits mean less profit due to better opponents. HU is particularly interesting because you're only playing against one person, and usually you're fortunate enough to be able to choose your opponent. If he's a fish, you'll wish he'd play higher limits so you could earn more off his weaknesses. If you're playing a regular, however, it gets complicated. So don't. Stay away from regulars, seek out the fish and rip them off.
Don't shove on the first hand unless you know there's a realistic chance the opponent will call. Meaning: he's bad, he wants to gamble and he's called first hand pushes before. Don't do it against unknowns. And never do it with (low) pocket pairs. You might as well put a buy-in on black, saves you time.
Finally, some people tend to fall into a psychological trap. Unless they're not passive by style or have decent tilt control, most opponents can't take losing 5 or more hands in a row. In villain's eyes, you're trying to run over him, or he's not getting proper cards which he deems "unfair". So he might try to fight back just because he feels like he "deserves" to win, seeking some sort of justice in the statistics as if it's his "turn" to win some chips. It's kind of like the gambler's fallacy, but projected on you with reverse psychology: "he won the last 8 pots, he can't possibly have it this time..." or my favorite "I folded the last 8 hands, so if I raise now he'll surely believe I have the better hand." Some go too far in meta-thinking and start believing you have figured their strategy out so they have to make unexpected moves to make you think he doesn't know you know. Try to avoid this mistake, overthinking it might lead to a lot of confusing thought processes like "I think he thinks I think he knows I know he knows," which rarely lead to correct decisions.
